@iD i ing Conso ium 2024 – 2027 – h ps//id i ing-p ojec .eu
In elligen & Digi al Roadway In as uc u e o
Vehicles In eg a ed wi h Nex -Gen Technologies
D2.3 – Technical Requi emen s and Digi al
Ecosys em A chi ec u e
Au ho (s):
Ioannis Tampakis, Geo ge Lalas, Da iya Rublo a
Leading Pa ne :
Ne company S.A.
Ve sion - S a us:
V0.9
Submission Da e:
06/2025
Dissemina ion Le el:
PUBLIC
Disclaime : Funded by he Eu opean Union. Views and opinions exp essed a e
howe e hose o he au ho (s) only and do no necessa ily e lec hose o he
Eu opean Union o CINEA. Nei he he Eu opean Union no CINEA can be held
esponsible o hem.
Copy igh message: ©iD i ing Conso ium, 2024-2027. This deli e able con ains
o iginal unpublished wo k excep whe e clea ly indica ed o he wise.
Acknowledgemen o p e iously published ma e ial and o he wo k o o he s has
been made h ough app op ia e ci a ion, quo a ion, o bo h. Rep oduc ion is
au ho ised p o ided he sou ce is acknowledged.
@iD i ing Conso ium – 101147004
Page 2 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Repo in o ma ion
Pu pose o he Repo
This epo ou lines he echnical equi emen s and
sys em a chi ec u e o iD i ing, ansla ing use
needs in o echnical speci ica ions. I p o ides a
ounda ion o he de elopmen , in eg a ion, and
alida ion o all p ojec componen s.
Rele an Wo k package:
WP2
Rele an Task:
T2.5 Technical equi emen s speci ica ions and ecosys em
a chi ec u e
Con ibu o s:
INTRA, CERTH, TEKNIKER, DREVEN, MBL, TUC, UNI.EIFFEL,
INFRA PLAN, AUSTRIATECH, ACCELI, SIMAVI
Nex e sion Ti le:
NA
Nex e sion Da e:
NA
O icial Submission Da e:
30/06/2025
Ac ual Submission Da e:
30/06/2025
ABOUT iDRIVING
iD i ing, a 3-yea Ho izon- unded p ojec , uni es 17 Eu opean pa ne s in a mission
o enhance oad sa e y. The p ojec ocuses on ans o ming u ban and seconda y
u al oad in as uc u e h ough inno a ion. I aligns wi h he EU’s goals o sma
anspo and ac i ely emb aces eme ging echnologies. Key a eas o impac
include enhancing d i e beha iou , imp o ing in as uc u e sa e y, and
empowe ing fi s esponde s. Th ough a comp ehensi e Sa e y C i e ia Ca alogue,
inno a i e senso s, AI-based wa nings, and a Digi al Twin, iD i ing pa es he way
o sa e oads.
The iD i ing conso ium consis s o he ollowing pa ne s:
No
Pa icipan o ganisa ion name
Sho name
Coun y
1
ETHNIKO KENTRO EREVNAS KAI TECHNOLOGIKIS
ANAPTYXIS
CERTH
EL
2
POLYTECHNEIO KRITIS
TUC
EL
3
AUSTRIATECH - GESELLSCHAFT DES BUNDES
FUR AT TECHNOLOGIEPOLITISCHE
MASSNAHMEN GMBH
AUSTRIATECH
AT
4
UNIVERSITE GUSTAVE EIFFEL
UNI.EIFFEL
FR
5
FUNDACION TEKNIKER
TEKNIKER
ES
6
INGARTEK CONSULTING SL
ING
ES
7
INFRA PLAN KONZALTNIG JDOO ZA USLUGE
INFRA PLAN
HR
8
ACCELIGENCE LTD
ACCELI
CY
@iD i ing Conso ium – 101147004
Page 3 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
9
NETCOMPANY S.A.
INTRA
LU
10
SOFTWARE IMAGINATION & VISION SRL
SIMAVI
RO
11
ALP.Lab GmbH
ALP.LAB
AT
12
MUNICIPALITY OF ALBA IULIA
AIM
RO
13
DRAXIS RESEARCH VENTURES ASTIKI MI EL
KERDOSKOPIKI ETAIRIA
DREVEN
EL
14
PRAVO I INTERNET FOUNDATION
LIF
BG
15
DIMOS THESSALONIKIS
THESSALONIKI
EL
16
GRAD KARLOVAC
COK
HR
17
MOBILYSIS SARL
MOBILYSIS SARL
CH
Ve sion His o y
Ve sion
Da e
Au ho
Pa ne
Desc ip ion
0.1
25/11/2024
Ioannis Tampakis
INTRA
Fi s e sion o ToC
0.2
18/03/2025
Ioannis Tampakis,
Geo ge Lalas
INTRA
Upda ed ToC
0.3
19/05/2025
Dimi is Pe ikleous,
And eas Le ka is
(ACCELI),
Alexand os
Pe opoulos,
S a os Paspalakis,
Panos V achnos,
Emmanuel Rap is,
Kons an inos
Ioannidis,
Alexand os S y idis,
Nikos Dou as
(CERTH), Mos a a
Ameli, Thomas
Bapaume, La i a
Oukhellou
(UNI.EIFFEL),
Emmanouil
Ba mpounakis,
Dimi ios Tsi sokas,
Jasso Espadale
Clapes (MBL),
Panagio is Tsalis
(Thessaloniki)
Susana Fe ei o,
Gonzalo Gil, Ike
Na baiza, Elena
Ga cía (TEK)
ACCELI,
CERTH,
UNI.EIFFEL,
MBL,
THESSALONIKI,
TEKNIKER,
TUC,
SIMAVI,
DREVEN,
INFRA PLAN,
ALP.LAB,
AUSTRIATECH
All ele an inpu s
om WP2, WP3, WP4
and WP5 Tasks
@iD i ing Conso ium – 101147004
Page 4 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Diana Go nea,
Daniel
Ghe ghiceanu,
La inia Popa
(SIMAVI), Vasileios
Ma kan onakis
(TUC), Zoi
Dimi iadou
(DREVEN),
I ina S ipano ic
(INFRA PLAN),
Mohamed
Be azouane
(ALP.LAB), Helena
Ko ndoe e
(AUSTRIATECH)
0.4
20/05/2025
Ioannis Tampakis
INTRA
iD i ing A chi ec u al
iews added
0.5
28/05/2025
Ioannis Tampakis,
Geo ge Lalas
INTRA
Upda ed in oduc ion,
me hodology,
conclusion, execu i e
summa y, and
e e ences.
0.6
05/06/2025
Rele an pa ne s
Rele an
pa ne s
Fu he upda es om
he con ibu ing
pa ne s
0.7
24/06/2025
Dimi ios Tsi sokas
(MBL), I ina
S ipano ic (INFRA
PLAN)
MBL, INFRA
PLAN
In e nal e iew
0.8
26/06/2025
Rele an pa ne s
Rele an
pa ne s
All commen s
add essed
0.9
26/06/2025
Ioannis Tampakis,
Da iya Rublo a
(INTRA)
INTRA
P e- inal documen
sen o Coo dina o
1.0
30/06/2025
Alexand os S y idis
CERTH
Submission o EC
@iD i ing Conso ium – 101147004
Page 5 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table o Con en s
1 In oduc ion ............................................................................................................ 13
1.1 Deli e able con ex .................................................................................................. 13
1.2 App oach o de ining he equi emen s .......................................................... 14
2 O e iew o he iD i ing sys em ..................................................................... 15
2.1 Sys em con ex and pu pose ................................................................................ 15
2.2 Sys em Desc ip ion and Main Capabili ies ...................................................... 15
2.3 Summa y o use cases, scena ios, and use equi emen s ........................16
2.3.1 Use Case 1.1: G az, Aus ia ......................................................................................................................... 16
2.3.2 Use Case 1.2: Ne e s, F ance .............................................................................................................. 17
2.3.3 Use Case 2.1: Ka lo ac, C oa ia ......................................................................................................... 18
2.3.4 Use Case 2.2: Thessaloniki, G eece ............................................................................................... 18
2.3.5 Use Case 3.1: Alba Iulia, Romania .................................................................................................... 19
2.3.6 Use Case 3.2: Bizkaia, Spain .............................................................................................................. 20
3 Technical equi emen s ...................................................................................... 21
3.1 Func ional & Non-Func ional equi emen s .................................................... 21
3.1.1 Gene al Non-Func ional equi emen s .......................................................................................... 21
3.1.2 Requi emen s ela ed o WP4............................................................................................................. 35
3.1.3 Requi emen s ela ed o WP5 ............................................................................................................ 50
4 Technical & use equi emen s Mappings .................................................. 59
5 O e all A chi ec u e & Componen s ............................................................. 67
5.1 Me hodology o De ining he A chi ec u e ................................................... 67
5.2 iD i ing Con ex View ............................................................................................. 67
5.3 iD i ing Con aine s View ...................................................................................... 68
5.3.1 Con aine s View – Consolida ed ........................................................................................................ 69
5.3.2 Con aine s View Pe Use Case ......................................................................................................... 71
5.3.2.1 Use case 1.1 G az, Aus ia ...................................................................................................... 71
5.3.2.2 Use Case 1.2 Ne e s, F ance................................................................................................... 72
5.3.2.3 Use Case 2.1 Ka lo ac, C oa ia ............................................................................................... 73
5.3.2.4 Use case 2.2 Thessaloniki, G eece .......................................................................................... 74
5.3.2.5 Use case 3.1 Alba Iulia, Romania ............................................................................................ 75
@iD i ing Conso ium – 101147004
Page 6 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5.3.2.6 Use case 3.2 Bizkaia, Spain ..................................................................................................... 76
5.4 iD i ing Componen s Views & Speci ica ions ................................................ 76
5.5 iDRIVING P ocess Views ...................................................................................... 128
5.5.1 Teknike Da aspace Connec o ......................................................................................................... 129
5.5.2 Rou e Guidance ool ........................................................................................................................... 130
5.5.3 Signal Con ol Tool ................................................................................................................................ 131
5.5.4 Mobile Applica ion & In-Vehicle Applica ion ....................................................................... 132
5.5.5 Ae ial Su eillance UAV Sys em ................................................................................................... 133
5.5.6 Au onomous UAV Deploymen o A ea Co e age ........................................................ 134
5.5.7 Sea Bel and Cell Phone De ec ion Tool............................................................................... 135
5.5.8 License Pla e De ec ion Tool ......................................................................................................... 136
5.5.9 Helme De ec ion Tool....................................................................................................................... 137
5.5.10 Zeb a C ossing De ec ion Tool ..................................................................................................... 138
5.5.11 Fallen T ee and Rockslide De ec ion Tool ............................................................................. 138
5.5.12 C ashed Vehicle De ec ion Tool .................................................................................................. 139
5.5.13 Red Ligh Viola ion De ec o ........................................................................................................ 140
5.5.14 Imp ope Lane Usage De ec o .................................................................................................... 141
5.5.15 Zeb a C ossing Viola ion De ec o ............................................................................................. 142
5.5.16 Ab up Mo emen De ec o .......................................................................................................... 143
5.5.17 Po hole De ec ion and Se e i y Assessmen Tool .......................................................... 144
5.5.18 Wea he P edic ion Tool (WRF and Da a Assimila ion) ............................................... 145
5.5.19 AI-Based Real-Time Wea he Ale Sys em .........................................................................146
5.5.20 3D-SMART Tool ....................................................................................................................................... 147
5.5.21 Clai eSITI Pla o m ................................................................................................................................ 147
5.5.22 SUMO (Simula ion o U ban Mobili y) ..................................................................................... 148
5.5.23 CARLA (Au onomous D i ing Simula ion Pla o m) ......................................................149
5.5.24 AI-Op imized Main enance h ough Digi al Twin ........................................................... 150
5.5.25 Digi al Twin-Based Con ol Cen e wi h XR ea u es o Enhanced Si ua ional
Awa eness ...................................................................................................................................................................... 155
6 Conclusion / Fu u e Wo k ............................................................................... 156
Re e ences ................................................................................................................... 157
7 PROJECT FACTS .................................................................................................. 158
@iD i ing Conso ium – 101147004
Page 7 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table o ables
Table 1 - Use Case 1.1: G az, Aus ia ............................................................................................................................... 16
Table 2 - Use Case 1.2: Ne e s, F ance ........................................................................................................................ 17
Table 3 - Use Case 2.1: Ka lo ac, C oa ia ................................................................................................................... 18
Table 4 - Use Case 2.2: Thessaloniki, G eece ......................................................................................................... 18
Table 5 - Use Case 3.1: Alba Iulia, Romania .............................................................................................................. 19
Table 6 - Use Case 2.3: Bizkaia, Spain ....................................................................................................................... 20
Table 7 – Technical equi emen o use equi emen mapping ........................................................... 59
Table 8 - Use equi emen o echnical equi emen mapping ........................................................... 62
Table 9 - Teknike Da aspace Connec o - A chi ec u al Design and Desc ip ion .................... 78
Table 10 - Rou e Guidance ool A chi ec u al Design and Desc ip ion ..............................................79
Table 11 - Signal Con ol Tool – A chi ec u al Design and Desc ip ion ............................................... 82
Table 12 - Mobile Applica ion – A chi ec u al Design and Desc ip ion ...............................................85
Table 13 - In-Vehicle Applica ion – A chi ec u al Design and Desc ip ion ...................................... 86
Table 14 - Ae ial Su eillance UAV Sys em – A chi ec u al Design and Desc ip ion ................. 89
Table 15 - Au onomous UAV Deploymen o A ea Co e age – A chi ec u al Design and
Desc ip ion ........................................................................................................................................................................ 91
Table 16 - Sea Bel and Cell Phone De ec ion Tool – A chi ec u al Design and Desc ip ion
................................................................................................................................................................................................. 94
Table 17 - Coun Ca s De ec ion Tool – A chi ec u al Design and Desc ip ion ............................. 95
Table 18 - License Pla e De ec ion Tool – A chi ec u al Design and Desc ip ion ....................... 96
Table 19 - Helme De ec ion Tool – A chi ec u al Design and Desc ip ion ..................................... 98
Table 20 - Zeb a C ossing De ec ion Tool – A chi ec u al Design and Desc ip ion .................. 99
Table 21 - Fallen T ee and Rockslide De ec ion Tool – A chi ec u al Design and Desc ip ion
............................................................................................................................................................................................... 100
Table 22 - C ashed Vehicle De ec ion Tool – A chi ec u al Design and Desc ip ion ............... 102
Table 23 - Red Ligh Viola ion De ec o – A chi ec u al Design and Desc ip ion ..................... 103
Table 24 - Imp ope Lane Usage De ec o – A chi ec u al Design and Desc ip ion .............. 104
Table 25 - Zeb a C ossing Viola ion De ec o – A chi ec u al Design and Desc ip ion ......... 105
Table 26 - Ab up Mo emen s De ec o – A chi ec u al Design and Desc ip ion .................... 107
Table 27 - Po hole De ec ion and Se e i y Assessmen Tool – A chi ec u al Design and
Desc ip ion .....................................................................................................................................................................109
@iD i ing Conso ium – 101147004
Page 8 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table 28 - Wea he P edic ion Tool (WRF and Da a Assimila ion) – A chi ec u al Design and
Desc ip ion ...................................................................................................................................................................... 110
Table 29 - AI-Based Real-Time Wea he Ale Sys em – A chi ec u al Design and Desc ip ion
................................................................................................................................................................................................. 113
Table 30 - 3D-Sma Tool – A chi ec u al Design and Desc ip ion ....................................................... 115
Table 31 - Dynamic Moni o ing Pla o m – A chi ec u al Design and Desc ip ion .................... 117
Table 32 - SUMO: Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning Sys ems –
A chi ec u al Design and Desc ip ion .......................................................................................................... 118
Table 33 - CARLA: Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning Sys ems –
A chi ec u al Design and Desc ip ion ......................................................................................................... 120
Table 34 - AI-Op imized Main enance Th ough Digi al Twin – A chi ec u al Design and
Desc ip ion ..................................................................................................................................................................... 123
Table 35 - Dynamic Risk Assessmen Module – A chi ec u al Design and Desc ip ion ........ 124
Table 36 - Heal h and Logis ics Managemen Module – A chi ec u al Design and Desc ip ion
................................................................................................................................................................................................ 125
Table 37 - Main enance Scheduling Tool – A chi ec u al Design and Desc ip ion .................. 127
Table 38 - Digi al Twin-Based Con ol Cen e – A chi ec u al Design and Desc ip ion ........ 127
Table O Figu es
Figu e 1 - iD i ing Sys em O e iew ........................................................................................................................... 16
Figu e 2 - iD i ing Con ex View ................................................................................................................................. 68
Figu e 3 - Consolida ed iD i ing Con aine s iew ........................................................................................... 70
Figu e 4 - Con aine s View Colou Legend .......................................................................................................... 70
Figu e 5 - Use Case 1.1 Con aine s View .................................................................................................................... 71
Figu e 6 - Use Case 1.2 Con aine s View .................................................................................................................. 72
Figu e 7 - Use Case 2.1 Con aine s View .................................................................................................................. 73
Figu e 8 - Use Case 2.2 Con aine s View ................................................................................................................ 74
Figu e 9 - Use Case 3.1 Con aine s View .................................................................................................................. 75
Figu e 10 - Use Case 3.2 Con aine s View ...............................................................................................................76
Figu e 11 - Teknike da aspace connec o sequence diag am ............................................................... 129
Figu e 12 - Rou e Guidance ool sequence diag am .................................................................................... 130
Figu e 13 - Signal Con ol ool sequence diag am .......................................................................................... 131
Figu e 14 - Mobile Applica ion & In-Vehicle Applica ion sequence diag am ............................... 132
Figu e 15 - Ae ial Su eillance UAV Sys em sequence diag am ............................................................ 133
@iD i ing Conso ium – 101147004
Page 9 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Figu e 16 - Au onomous UAV Deploymen o A ea Co e age sequence diag am ................. 134
Figu e 17 - Sea Bel and Cell Phone De ec ion Tool sequence diag am ....................................... 135
Figu e 18 - License Pla e De ec ion Tool sequence diag am .................................................................. 136
Figu e 19 - Helme De ec ion Tool sequence diag am ............................................................................... 137
Figu e 20 - Zeb a C ossing De ec ion Tool sequence diag am ............................................................ 138
Figu e 21 - Fallen T ee and Rockslide De ec ion Tool sequence diag am ...................................... 138
Figu e 22 - C ashed Vehicle De ec ion Tool sequence diag am .......................................................... 139
Figu e 23 - Red Ligh Viola ion De ec o sequence diag am ................................................................ 140
Figu e 24 - Imp ope Lane Usage De ec o sequence diag am ........................................................... 141
Figu e 25 - Zeb a C ossing Viola ion De ec o sequence diag am ..................................................... 142
Figu e 26 - Ab up Mo emen De ec o sequence diag am .................................................................. 143
Figu e 27 - Po hole De ec ion and Se e i y Assessmen Tool sequence diag am .................. 144
Figu e 28 - Wea he P edic ion Tool sequence diag am .......................................................................... 145
Figu e 29 - AI-Based Real-Time Wea he Ale Sys em sequence diag am .................................146
Figu e 30 - 3D-SMART Tool sequence diag am ............................................................................................... 147
Figu e 31 - Clai eSITI Pla o m sequence diag am ......................................................................................... 147
Figu e 32 - SUMO sequence diag am ................................................................................................................... 148
Figu e 33 - CARLA sequence diag am ...................................................................................................................149
Figu e 34 - AI-Op imized Main enance h ough Digi al Twin - Global Componen sequence
diag am ............................................................................................................................................................................ 150
Figu e 35 - AI-Op imized Main enance h ough Digi al Twin – Module 1.1 Risk Assessmen
sequence diag am ..................................................................................................................................................... 151
Figu e 36 - AI-Op imized Main enance h ough Digi al Twin – Module 1.2 Heal h & Logis ics
Managemen sequence diag am ................................................................................................................... 153
Figu e 37 - AI-Op imized Main enance h ough Digi al Twin - Module 1.3 Main enance
Scheduling sequence diag am ........................................................................................................................ 154
Figu e 38 - Digi al Twin-Based Con ol Cen e wi h XR ea u es o Enhanced Si ua ional
Awa eness sequence diag am ......................................................................................................................... 155
@iD i ing Conso ium – 101147004
Page 16 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Figu e 1 - iD i ing Sys em O e iew
This digi al ecosys em enables seamless in o ma ion low among sys em
componen s and s akeholde s, suppo ing bo h cen alized con ol and
decen alized, on-si e ac ion.
2.3 Summa y o use cases, scena ios, and use
equi emen s
The ollowing sub-chap e s p o ide an o e iew o he six use cases, including
scena io desc ip ions and use equi emen s.
2.3.1 Use Case 1.1: G az, Aus ia
Table 1 - Use Case 1.1: G az, Aus ia
Scena io Ti le
Desc ip ion
Conges ion P edic ion and Dynamic Re-
ou ing
Con inuous moni o ing o a ic low is
conduc ed using a ious senso s and AI-
enabled came as. P edic i e models analyse
eal- ime and his o ical da a o o ecas
conges ion. Based on hese p edic ions,
al e na i e ou es a e communica ed o oad
use s ia VMS and mobile o In- ehicle
applica ions o op imize a ic dis ibu ion.
@iD i ing Conso ium – 101147004
Page 17 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Au oma ed De ec ion o Road Use
Viola ions
Video analy ics sys ems au oma ically de ec
iola ions such as non-usage o helme s o
sea bel s, a ic ligh in ac ions, and
unau ho ized lane usage, including w ong
way d i ing o imp ope use o u ning lanes.
Iden i ied iola ions a e documen ed o oad
ope a o s and communica ed o ele an
oad use s in o de o inc ease compliance.
Sa e y-C i ical Inciden s De ec ion
The sys em con inuously analyses senso s and
ideo da a o iden i y dange ous o c i ical
si ua ions. Upon de ec ion, immedia e ale s
a e issued o oad use s and oad ope a o s o
enable p omp in e en ion and minimize
po en ial isks.
2.3.2 Use Case 1.2: Ne e s, F ance
Table 2 - Use Case 1.2: Ne e s, F ance
Scena io i le
Desc ip ion
Deploymen o Moni o ing sys ems
Ins all came as a key loca ions like
in e sec ions, pa king zones, and zeb a
c ossings, while deploying UAVs o e c i ical
a eas, including bicycle lanes and pedes ian-
hea y zones, o p o ide a comp ehensi e
g ound and ae ial iew o oads
Real Time in eg a ion and analysis
In eg a e g ound came a and UAV da a in o
iD i ing’s Digi al Twin, using AI o analyse a ic
pa e ns, de ec in ac ions, iden i y isky
beha iou s, and highligh a eas o
in as uc u e imp o emen s.
Communica ion and ale s o oad use s
Ac i a e messaging on oad o online sys ems o
ale oad use s o po en ial inciden s. No i y
d i e s app oaching zeb a c ossings o yield ia
VMS o in- ehicle messages. Use VLC o
communica ion and wa n dis ac ed cyclis s o
enhance compliance and sa e y.
@iD i ing Conso ium – 101147004
Page 18 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
2.3.3 Use Case 2.1: Ka lo ac, C oa ia
Table 3 - Use Case 2.1: Ka lo ac, C oa ia
Scena io i le
Desc ip ion
De ec ion o oad damages (po holes,
c acks)
Deploy se e al echnologies o suppo
inspec ion p ocedu e: da a collec ion using
came as on inspec ion ehicles, ixed came as
on c i ical sec ions, like in e sec ions o sec ions
wi h hea y a ic, and UAVs o e emo e a eas.
En i onmen al moni o ing using sma wea he
s a ions, o co ela e oad damages occu ence.
P o ide main enance plan o
in as uc u e manage s o epai o
de ec ed po holes
De elop DT-based op imized model o
main enance plan while conside ing a ic low
(came as and a ic coun e s da a) and wea he
o ecas (wea he s a ion)
Communica ion and ale s o oad use s
be o e and du ing main enance wo ks
Ac i a e messaging on oad o online sys ems o
ale oad use s o main enance oad ac i i ies.
No i y d i e s app oaching oad wo k zones o
yield ia VMS o in- ehicle messages. Use VLC
o communica ion and wa n dis ac ed cyclis s
o enhance compliance and sa e y.
Moni o wo ke s sa e y du ing
main enance wo ks
Deploy UAV du ing he main enance wo ks o
de ec any sa e y iola ions in he wo k zone
a eas
2.3.4 Use Case 2.2: Thessaloniki, G eece
Table 4 - Use Case 2.2: Thessaloniki, G eece
Scena io i le
Desc ip ion
Accumula ing wa e om ain all/ Gus y
winds
Ad anced wea he s a ions moni o wea he
da a. Da a analysed o iden i y pa e ns
indica ing po en ial oad haza ds, such as wa e
accumula ion ha could lead o slippe y
condi ions o looding o s ong winds o wo
wheele s. In- ehicle wa ning applica ion, digi al
oad signs, and he dedica ed iD i ing mobile
applica ion all s a issuing ale s and in o m on
al e na i e ou es. T a ic manage is in o med
Fallen T ees/Rockslides
UAVs ly au onomously pa olling he isk a eas
o de ec obs acles on he oad like allen ees
o ocks. Exis ence o obs acles on he oad
@iD i ing Conso ium – 101147004
Page 19 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
leads o wa ning o iD i ing pla o m. Wa ning
applica ions and digi al oad signs all s a
issuing ale s and in o m on al e na i e ou es.
T a ic manage is in o med
2.3.5 Use Case 3.1: Alba Iulia, Romania
Table 5 - Use Case 3.1: Alba Iulia, Romania
Scena io i le
Desc ip ion
Agg essi e D i ing De ec ion
De ec ion o agg essi e beha iou in a ic (e.g.,
sudden lane changes, apid
accele a ion/decele a ion).
The scena io implies using ideo came as wi h
ideo analy ics o iden i y isky beha iou s, (e.g.,
a ic senso s o de ec sudden lane changes
and ha d b aking,) dedica ed T a ic
Managemen Cen e pla o m o analysing and
p ocessing ale s.
Speeding de ec ion
Real- ime speeding de ec ion using sma
came as and speed came as
The scena io implies using ideo came as o
ehicle iden i ica ion, sma ada s o speed
measu emen , pla o m dedica ed o he
exis ing T a ic Managemen Sys em in Alba Iulia,
mobile applica ion o nea eal- ime use
no i ica ion.
The scena io a ge s speed o e he legal limi
de ec ion, au oma ically gene a ing ale s o
d i e s and sending hem ia he mobile app,
analysis o collec ed da a o iden i y high- isk
a eas, implemen a ion o p oac i e measu es o
op imize a ic low ( a ic ligh adjus men )
Speeding and Agg essi e D i ing
De ec ion
Combining de ec ion o excessi e speed and
agg essi e beha iou o p o ide a comple e
a ic isk moni o ing solu ion.
The scena io implies using ideo came as and
sma ada s o speed and agg essi e beha iou
de ec ion, a ic senso s o moni o sudden lane
changes and dange ous b aking, dedica ed
pla o m o a ic managemen sys em in Alba
@iD i ing Conso ium – 101147004
Page 20 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Iulia, mobile applica ion o nea eal- ime ale s
and no i ica ions.
The scena io a ge s simul aneous de ec ion o
excessi e speeding and agg essi e beha iou ,
d i e no i ica ion ia VMS panels and mobile
app.
C ea ing hea maps o iden i y high- isk a eas.
2.3.6 Use Case 3.2: Bizkaia, Spain
Table 6 - Use Case 2.3: Bizkaia, Spain
Scena io i le
Desc ip ion
Real- ime de ec ion
iD i ing senso s ins alled in one o he ehicles
in ol ed in he acciden de ec and epo he
sudden s op. O he d i e s in he a ea, using
he iD i ing in- ehicle o he mobile app,
con i m he acciden .
Deploymen o Moni o ing sys ems
Came as ins alled, and UAVs dispa ched o e
he a ea p o ide a comp ehensi e iew o he
si ua ion, co e ing bo h g ound and ae ial
pe spec i es.
Digi al win gene a ion
NeRFs and isual analysis c ea e an accu a e 3D
model o he zone. This eal- ime da a is sha ed
wi h local au ho i ies, p o iding a i ual
eplica ion.
Re ou ing and planning
Using his o ical a ic da a and eal- ime inpu ,
iD i ing’s algo i hms de e mine he as es
ou e o eme gency ehicles and he
al e na i e ou es o app oaching d i e s.
Real- ime ale s
Th ough he mobile app, use s a e in o med
abou he al e na i e ou es, while VMS sugges
slowing down wi hin he a ea.
@iD i ing Conso ium – 101147004
Page 21 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
3 Technical equi emen s
This sec ion p esen s he iD i ing echnical equi emen s, g ouped acco ding o
hei scope and ela ionship o he p ojec ’s wo k s uc u e. Unlike a s ic spli
be ween unc ional and non- unc ional equi emen s, ou app oach is as ollows:
• Gene al Non-Func ional Requi emen s a e lis ed in a dedica ed sec ion.
These cap u e c oss-cu ing quali y a ibu es ha apply o he sys em, such
as eal- ime pe o mance, in e ope abili y, access con ol, usabili y, da a
p o ec ion, and compliance. These a e iden i ied wi h he o ma TR-NFUN-
GEN-y, whe e y is a unning index o gene al non- unc ional equi emen s.
• Wo k Package-Speci ic Requi emen s a e hen de ailed pe Wo k Package
(WP). Fo each WP, we p esen bo h unc ional and non- unc ional
equi emen s oge he , e lec ing he p ac ical in eg a ion o quali y
a ibu es and sys em unc ionali ies in he p ojec ’s implemen a ion.
Requi emen s he e a e labelled as ollows:
o TR-FUN-x-y o unc ional equi emen s, whe e x is he WP numbe
and y is he index wi hin ha WP.
o TR-NFUN-x-y o non- unc ional equi emen s speci ic o a WP,
ollowing he same numbe ing logic.
This s uc u e ensu es ha bo h ypes o equi emen s, unc ional (wha he sys em
does) and non- unc ional (how he sys em pe o ms o mus beha e), a e
add essed in di ec connec ion o he esponsible p ojec a eas and ools. I also
makes clea which equi emen s apply p ojec -wide and which a e WP-speci ic.
The se o equi emen s can e ol e du ing he p ojec , due o e.g. an upda e o he
use equi emen s. This may lead o addi ional equi emen s, which can be
assigned consecu i e highe numbe s o al eady exis ing numbe s wi h lowe case
le e s appended (e.g. …-1a), o o some equi emen s being d opped, wi hou
enumbe ing o he ull se .
3.1 Func ional & Non-Func ional equi emen s
3.1.1 Gene al Non-Func ional equi emen s
ID
TR-NFUN-GEN-1
Name (Op ional)
Real-Time Pe o mance
Desc ip ion
C i ical ale s, sa e y p ocessing, and con ol unc ions mus
ope a e wi h eal- ime o nea eal- ime la ency
Ca ego y
Pe o mance
@iD i ing Conso ium – 101147004
Page 22 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Rela ed Tasks
T3.2, T3.3, T4.1, T4.2, T5.5
Rela ed Tool
-
Rela ed Use Cases
All
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
-
Ve i ica ion Me hod
Pe o mance es ing, simula ion
Commen s
-
ID
TR-NFUN-GEN-2
Name (Op ional)
In e ope abili y ia S anda d Fo ma s
Desc ip ion
All ools mus exchange da a using open, well-de ined o ma s
(JSON, CSV, e c.) wi h schemas.
Ca ego y
In e ope abili y
Rela ed Tasks
All
Rela ed Tool
-
Rela ed Use Cases
All
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
-
Ve i ica ion Me hod
In eg a ion es ing, schema alida ion
Commen s
-
ID
TR-NFUN-GEN-3
Name (Op ional)
Access Con ol
Desc ip ion
The pla o m mus p o ide ole-based access o unc ions (e.g.,
admin/ope a o ), wi h clea sepa a ion o p i ileges.
Ca ego y
Secu i y, Design
Rela ed Tasks
All
Rela ed Tool
-
Rela ed Use Cases
All
MoSCoW Scale
mus -ha e
@iD i ing Conso ium – 101147004
Page 23 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Dependencies
-
Inpu Needed
-
Ve i ica ion Me hod
Access con ol es ing, ole simula ion
Commen s
E.g. Admin, ope a o , and use oles.
ID
TR-NFUN-GEN-4
Name (Op ional)
Usabili y o C i ical UIs
Desc ip ion
Mobile, in o ainmen , and ope a o UIs mus comply wi h
usabili y s anda ds (e.g., non-dis ac ion, ISO 15005), and be
es ed wi h end use s.
Ca ego y
Usabili y, Design
Rela ed Tasks
T3.3, T5.5
Rela ed Tool
-
Rela ed Use Cases
All
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
-
Ve i ica ion Me hod
Usabili y es ing, su eys,
Commen s
-
ID
TR-NFUN-GEN-5
Name (Op ional)
Inpu Da a Quali y
Desc ip ion
All image/ ideo inpu s mus ha e ≥1080p esolu ion and ≥5 ps
ame a e o accu a e de ec ion.
Ca ego y
Da a Quali y
Rela ed Tasks
T3.4, T4.1
Rela ed Tool
-
Rela ed Use Cases
All
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
-
Ve i ica ion Me hod
Da a alida ion
@iD i ing Conso ium – 101147004
Page 24 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Commen s
-
ID
TR-NFUN-GEN-6
Name (Op ional)
Ha dwa e Compa ibili y
Desc ip ion
All edge p ocessing modules mus be deployable on an
embedded de ice such as a NVIDIA Je son
Ca ego y
Compa ibili y
Rela ed Tasks
T4.1
Rela ed Tool
-
Rela ed Use Cases
All
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
-
Ve i ica ion Me hod
Deploymen es ing on Je son
Commen s
Requi ed o ield deploymen
ID
TR-NFUN-GEN-7
Name (Op ional)
Ne wo k Connec i i y
Desc ip ion
The sys em mus main ain an ac i e in e ne connec ion o
ensu e eal- ime p ocessing and communica ion.
Ca ego y
A ailabili y
Rela ed Tasks
All
Rela ed Tool
-
Rela ed Use Cases
All
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
-
Ve i ica ion Me hod
-
Commen s
Applies o UAVs, edge de ices, se e comms.
ID
TR-NFUN-GEN-8
Name (Op ional)
Came a Field o View
@iD i ing Conso ium – 101147004
Page 25 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Desc ip ion
Came as mus be ins alled o maximize ield o iew and
op imize o he de ec ion algo i hm’s needs.
Ca ego y
Deploymen , Quali y
Rela ed Tasks
T4.1, T4.2
Rela ed Tool
-
Rela ed Use Cases
All
MoSCoW Scale
should-ha e
Dependencies
-
Inpu Needed
-
Ve i ica ion Me hod
Field alida ion, sample checks
Commen s
-
ID
TR-NFUN-GEN-9
Name (Op ional)
Da a P o ec ion and E hical Compliance
Desc ip ion
The sys em shall handle all de ec ion ope a ions and da a
p ocessing in acco dance wi h GDPR equi emen s and
es ablished AI e hics p inciples (e.g., ai ness, anspa ency,
accoun abili y).
Ca ego y
Secu i y, P i acy, E hics
Rela ed Tasks
All
Rela ed Tool
All
Rela ed Use Cases
All
MoSCoW Scale
mus -ha e
Dependencies
Applicable GDPR egula ions; AI E hics F amewo ks (e.g., EU AI
Ac )
Inpu Needed
-
Ve i ica ion Me hod
-
Commen s
This NFR ensu es ha any collec ion, p ocessing, o analysis o
pe sonal da a espec s use p i acy igh s and ha AI-d i en
ope a ions emain ai , anspa en , and accoun able.
iD i ing echnical equi emen speci ica ion
ID
TR-FUN-3.1-1
Name (Op ional)
Da a Ca alog Publica ion, Policy Nego ia ion, Da a Access
@iD i ing Conso ium – 101147004
Page 32 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Inpu Needed
Fo eal ime communica ion and li e da a ansmission ne wo k
is equi ed.
Ve i ica ion Me hod
-
Commen s
ACCELI can p o ide a 4G s ick ha can be connec ed o he
g ound s a ion o di ec ly o he d one, p o iding he ne wo k
equi ed o ansmi ing da a o he cloud, i needed.
This equi emen is a ask-speci ic ex ension o TR-NFUN-GEN-7.”
ID
TR-NFUN-3.4.5
Name (Op ional)
Compliance wi h Regula ions
Desc ip ion
The UAV mus comply wi h EASA egula ions in he Open
Ca ego y A1/A3, namely:
• Maximum Take-o Mass (MTOM): D ones in his subca ego y
can ha e a maximum weigh o 25 kilog ams.
• No Fly Zones: Flying is p ohibi ed o e open-ai assemblies o
people (e.g., conce s, es i als) and densely popula ed a eas wi h
mo e han 300 people pe squa e me e .
• Visual Line o Sigh (VLOS): The pilo mus always main ain
isual con ac wi h he d one du ing he ligh .
• Maximum Al i ude: D ones canno ly abo e 120 me e s abo e
g ound le el (AGL) unless a special au ho iza ion is ob ained.
Ca ego y
Non-Func ional, Regula o y/compliance
Rela ed Tasks
T3.4. Ae ial Su eillance in Inciden Managemen and
Main enance Tasks
Rela ed Tool
iD i ing Mission UAV
Rela ed Use Cases
Spain, Aus ia, G eece, C oa ia, F ance
MoSCoW Scale
Mus Ha e
Dependencies
-
Inpu Needed
All equi ed ligh au ho iza ions mus be p o ided by he end
use s o he ACCELI eam o conduc he missions o e hei
designa ed a eas.
Ve i ica ion Me hod
-
Commen s
-
@iD i ing Conso ium – 101147004
Page 33 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
ID
TR-FUN-3.5-1
Name (Op ional)
Mission A ea De ini ion
Desc ip ion
The se ice shall accep as inpu a geog aphical a ea de ined as a
polygon in WGS84 coo dina es, including in o ma ion abou
possible no- ly zones and/o obs acles. This inpu will de ine he
spa ial bounda ies and cons ain s o he mission a ea, which will
be used o in o m and cons ain UAV pa h planning and mission
execu ion.
Ca ego y
Func ional, Design, Regula o y/Compliance
Rela ed Tasks
T3.5. Dynamic Resou ce Alloca ion o Op imized Sa e y
Su eillance
Rela ed Tool
UAV-based Pa h Planning o Co e age Ope a ions
Rela ed Use Cases
UC 1.2 Ne e s, F ance/ UC 2.1 Ka lo ac, C oa ia/ UC 2.2 Thessaloniki,
G eece/ UC 3.2 Bizkaia, Spain
MoSCoW Scale
mus -ha e
Dependencies
Geospa ial da a a ailabili y, In e ne Connec ion
Inpu Needed
WGS84 polygon, lis o obs acles/no- ly-zones (i applicable)
Ve i ica ion Me hod
Uni and in eg a ion es ing; inpu pa sing alida ion
Commen s
Founda ional o ini ia ing he mission planning p ocess
ID
TR-FUN-3.5-2
Name (Op ional)
Single-UAV Co e age Pa h Planning
Desc ip ion
The se ice shall compu e an op imal co e age pa h o a single
UAV o map/moni o a speci ied a ea, conside ing any obs acles,
no- ly zones wi hin he ope a ional a ea.
Ca ego y
Func ional, Design
Rela ed Tasks
T3.5. Dynamic Resou ce Alloca ion o Op imized Sa e y
Su eillance
Rela ed Tool
UAV-based Pa h Planning o Co e age Ope a ions
Rela ed Use Cases
UC 1.2 Ne e s, F ance/ UC 2.1 Ka lo ac, C oa ia/ UC 2.2 Thessaloniki,
G eece/ UC 3.2 Bizkaia, Spain
MoSCoW Scale
mus -ha e
Dependencies
TR-FUN-3.5-1, UAV a ailabili y, speci ica ions and capabili ies,
In e ne Connec ion
@iD i ing Conso ium – 101147004
Page 34 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Inpu Needed
UAVs cha ac e is ics i.e., Ba e y le el, Maximum ligh ime,
Maximum – Minimum al i ude o sa e y and legal cons ain s,
Maximum speed, Payload (came a/FoV) specs
Ve i ica ion Me hod
Simula ed mission co e age e i ica ion
Commen s
Ensu es mission execu ion o mapping/moni o ing an a ea o
in e es
ID
TR-FUN-3.5-3
Name (Op ional)
Mul i-UAV Co e age Pa h Planning
Desc ip ion
The se ice shall compu e coope a i e pa hs o a swa m o UAVs
o coope a i ely map/moni o a speci ied a ea, conside ing any
obs acles, no- ly zones wi hin he ope a ional a ea.
Ca ego y
Func ional, Design
Rela ed Tasks
T3.5. Dynamic Resou ce Alloca ion o Op imized Sa e y
Su eillance
Rela ed Tool
UAV-based Pa h Planning o Co e age Ope a ions
Rela ed Use Cases
UC 1.2 Ne e s, F ance/ UC 2.1 Ka lo ac, C oa ia/ UC 2.2
Thessaloniki, G eece/ UC 3.2 Bizkaia, Spain
MoSCoW Scale
mus -ha e
Dependencies
Mul iple UAVs, In e ne Connec ion
Inpu Needed
Numbe o UAVs, UAVs cha ac e is ics i.e., Ba e y le el,
Maximum ligh ime, Maximum – Minimum al i ude o sa e y
and legal cons ain s, Maximum speed, Payload (came a/FoV)
specs
Ve i ica ion Me hod
Simula ed mission co e age e i ica ion
Commen s
Ensu es dis ibu ed mission execu ion ac oss a ailable UAVs o
mapping/moni o ing an a ea o in e es
ID
TR-NFUN-3.5-4
Name (Op ional)
Da a In e ope abili y
Desc ip ion
The sys em shall suppo s anda dized da a o ma s o mission
inpu s/ou pu s, such as .JSON o a ea de ini ions and waypoin
ex ac ion.
Ca ego y
In e ope abili y
Rela ed Tasks
T3.5. Dynamic Resou ce Alloca ion o Op imized Sa e y
Su eillance
Rela ed Tool
UAV-based Pa h Planning o Co e age Ope a ions
@iD i ing Conso ium – 101147004
Page 35 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Rela ed Use Cases
UC 1.2 Ne e s, F ance/ UC 2.1 Ka lo ac, C oa ia/ UC 2.2
Thessaloniki, G eece/ UC 3.2 Bizkaia, Spain
MoSCoW Scale
mus -ha e
Dependencies
In e ne Connec ion
Inpu Needed
Da a schema de ini ions
Ve i ica ion Me hod
In eg a ion es ing
Commen s
Only he ou pu will be es ed h ough in eg a ion es ing
agains he de ined da a schema. The inpu will be inse ed
h ough he pla o m ha will be de eloped unde Task 3.5,
which will ac as he in e ace o mission a ea and pa ame e
de ini ions.
This equi emen ope a ionalizes TR-NFUN-GEN-2 o UAV
mission planning.
3.1.2 Requi emen s ela ed o WP4
ID
TR-NFUN-4.1-1
Name (Op ional)
Je son NVIDIA Pla o m
Desc ip ion
T4.1 will be implemen ed on an embedded de ice, such as he
NVIDIA Je son pla o m
Ca ego y
Equipmen
Rela ed Tasks
T4.1 - Real-Time Edge Compu ing o Visual Moni o ing
Rela ed Tool
-
Rela ed Use Cases
UC1.1/ UC1.2/ U.C. 2.2./U.C 3.2
MoSCoW Scale
-
Dependencies
-
Inpu Needed
-
Ve i ica ion Me hod
-
Commen s
-
ID
TR-FUN-4.1-1
Name (Op ional)
Sea bel de ec ion.
@iD i ing Conso ium – 101147004
Page 36 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Desc ip ion
The sys em shall execu e an objec de ec ion algo i hm on
embedded de ices connec ed o oadside came as o UAVs o
iden i y and classi y sea bel usage.
Ca ego y
Func ional
Rela ed Tasks
T4.1 - Real-Time Edge Compu ing o Visual Moni o ing
Rela ed Tool
Sea bel de ec ion ool
Rela ed Use Cases
UC 1.1-G az, Aus ia
MoSCoW Scale
mus -ha e
Dependencies
Dependence on a da ase con aining images o d i e s bo h
wea ing and no wea ing sea bel s o aining he deep lea ning
model.
Came a's ield o iew and angle
Inpu Needed
F ames o ideos wi h high ame a e
Ve i ica ion Me hod
T aining he deep lea ning model, and pe o ming in e ence on a
p o ided sample, as pa o in eg a ion es ing
Commen s
P o ided da a o ideo samples om di e en angles a e c ucial
o he unc ionali y o he ool. Mos o he images in he cu en
model a e cap u ed om oad came as a a 45-deg ee angle and
no om a long dis ance. Depending on he in eg a ion and he
placemen o he came as, we may need da a o e ain he model
om ha speci ic poin o iew.
ID
TR-FUN-4.1-2
Name (Op ional)
Mobile de ec ion.
Desc ip ion
The sys em shall execu e an objec de ec ion algo i hm on
embedded de ices connec ed o oad came as o UAVs o
iden i y and classi y mobile phone usage.
Ca ego y
Func ional
Rela ed Tasks
T4.1 - Real-Time Edge Compu ing o Visual Moni o ing
Rela ed Tool
Mobile de ec ion ool
Rela ed Use Cases
UC 1.1-G az, Aus ia/ UC 1.2-Ne e s, F ance
MoSCoW Scale
mus -ha e
Dependencies
Dependence on exis ing da ase s o aining he deep lea ning
model
Came a's ield o iew and angle
@iD i ing Conso ium – 101147004
Page 37 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Inpu Needed
F ames o ideos wi h high ame a e
Ve i ica ion Me hod
T aining he deep lea ning model, and pe o ming in e ence on a
p o ided sample, as pa o in eg a ion es ing
Commen s
P o ided da a o ideo samples om di e en angles a e c ucial
o he unc ionali y o he ool. Mos o he images in he cu en
model a e cap u ed om oad came as a a 45-deg ee angle and
no om a long dis ance. Depending on he in eg a ion and
placemen o he came as, we may need da a o e ain he model
om ha speci ic poin o iew. In UC 1.2, de ec ing bike s using
mobile phones equi es a dedica ed da ase . Howe e , we
we e unable o ind any sui able da ase . I such a da ase can
be p o ided, hen his ool can also be in eg a ed in o UC 1.2.
ID
TR-FUN-4.1-3
Name (Op ional)
Helme De ec ion
Desc ip ion
The sys em shall execu e an objec de ec ion algo i hm on
embedded de ices o e i y helme usage by ide s o bicycles
and mo o cycles
Ca ego y
Func ional
Rela ed Tasks
T4.1 - Real-Time Edge Compu ing o Visual Moni o ing
Rela ed Tool
Helme de ec ion ool
Rela ed Use Cases
UC 1.1-G az, Aus ia/ UC1.2-Ne e s, F ance
MoSCoW Scale
mus -ha e
Dependencies
Dependence on exis ing da ase s o aining he deep lea ning
model
Came a's ield o iew and angle
Inpu Needed
F ames o ideos wi h high ame a e
Ve i ica ion Me hod
T aining he deep lea ning model, and pe o ming in e ence on a
p o ided sample, as pa o in eg a ion es ing
Commen s
P o ided da a o ideo samples om di e en angles a e c ucial
o he unc ionali y o he ool. Mos o he images in he cu en
model a e cap u ed om oad came as a a 45-deg ee angle and
no om a long dis ance. Depending on he in eg a ion and he
placemen o he came as, we may need da a o e ain he model
om ha speci ic poin o iew.
@iD i ing Conso ium – 101147004
Page 38 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
ID
TR-FUN-4.1-4
Name (Op ional)
License Pla e De ec ion
Desc ip ion
The sys em shall execu e an objec de ec ion algo i hm on
embedded de ices o iden i y he license pla es o o ende s who
a e no wea ing sea bel s o helme s, o who a e using mobile
phones while d i ing.
Ca ego y
Func ional
Rela ed Tasks
T4.1 - Real-Time Edge Compu ing o Visual Moni o ing
Rela ed Tool
License pla e de ec ion ool
Rela ed Use Cases
UC 1.1-G az, Aus ia/ UC1.2-Ne e s, F ance
MoSCoW Scale
should-ha e
Dependencies
Dependence on exis ing da ase s o aining he deep lea ning
model
Came a's ield o iew and angle
Inpu Needed
F ames o ideos wi h high ame a e
Ve i ica ion Me hod
T aining he deep lea ning model, and pe o ming in e ence on a
p o ided sample, as pa o in eg a ion es ing
Commen s
The model has been ained on a da ase con aining close-up
images. Addi ionally, images om s ee came as we e included
o help gene alize he model. Howe e , due o he small size o
license pla es, i may be di icul o de ec and ead hem om
d one oo age. Fo accu a e license pla e ecogni ion, he image
mus be clea enough o dis inguish he cha ac e s.
ID
TR-FUN-4.1-5
Name (Op ional)
Coun Ca s
Desc ip ion
The sys em shall de ec and coun he numbe o ca s p esen in
each ideo ame cap u ed by connec ed came as.
Ca ego y
Func ional
Rela ed Tasks
T4.1 - Real-Time Edge Compu ing o Visual Moni o ing
Rela ed Tool
Coun Ca de ec ion ool
Rela ed Use Cases
UC 1.1-G az, Aus ia, UC2.2-Thessaloniki, G eece
MoSCoW Scale
should-ha e
Dependencies
Dependence on exis ing da ase s o aining he deep lea ning
model
@iD i ing Conso ium – 101147004
Page 39 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Came a's ield o iew and angle
P o ision o da a
Inpu Needed
F ames o ideos wi h high ame a e
Ve i ica ion Me hod
T aining he deep lea ning model, and pe o ming in e ence on
p o ided samples, as pa o in eg a ion es ing
Commen s
This ool was added la e , ollowing he plena y mee ing in G az.
P o ided da a o ideo samples om di e en angles a e c ucial
o he unc ionali y o he ool. Mos o he images in he cu en
model a e cap u ed om oad came as a a 45-deg ee angle and
no om a long dis ance. Depending on he in eg a ion and he
placemen o he came as, we may need da a o e ain he model
om ha speci ic poin o iew.
ID
TR-FUN-4.1-6
Name (Op ional)
Zeb a c ossing de ec ion
Desc ip ion
The sys em shall de ec he p esence o zeb a c ossings in ideo
ames cap u ed by oad came as o UAVs.
Ca ego y
Func ional
Rela ed Tasks
T4.1 - Real-Time Edge Compu ing o Visual Moni o ing
Rela ed Tool
Zeb a C ossing de ec ion ool
Rela ed Use Cases
UC 1.2. - Ne e s, F ance
MoSCoW Scale
should-ha e
Dependencies
Dependence on exis ing da ase s o aining he deep lea ning
model
Came a's ield o iew and angle
P o ision o da a
Inpu Needed
F ames o ideos wi h high ame a e
Ve i ica ion Me hod
T aining he deep lea ning model, and pe o ming in e ence on
p o ided samples, as pa o in eg a ion es ing
Commen s
This ool was added ollowing a bila e al mee ing wi h TEKNIKER
o T4.2. The pu pose is o de e mine whe he a pe son is walking
on a zeb a c ossing o no – simila o checking i someone is
walking on sidewalk.
I needed CERTH will p o ide only he coo dina es o he zeb a
c ossing, by combining he zeb a c ossing coo dina es wi hin he
ame wi h he de ec ed posi ion o he pe son. We can assess
@iD i ing Conso ium – 101147004
Page 40 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
whe he he indi idual is walking on he s ee o wi hin he
designed c ossing a ea.
Sidewalk de ec ion is no easible, as mos pa emen s equi e
polygon-shaped anno a ions o be accu a ely bounded. Howe e ,
ou cu en models only suppo ec angula bounding boxes.
T aining a model o handle bo h ec angula and polygon
anno a ion is no suppo ed wi hin ou amewo k, The e o e,
only zeb a c ossing can be de ec ed, gi en ha he p o ided
da ase includes ec angula bounding box anno a ions.
ID
TR-FUN-4.1-7
Name (Op ional)
Fallen T ee and Rockslide de ec ion ool
Desc ip ion
The sys em shall execu e an objec de ec ion algo i hm on
embedded de ices o moni o landscapes and oadways o allen
ees and ockslides using ideo eeds om UAVs. I analyses
ae ial image y o quickly iden i y and assess po en ial haza ds.
Ca ego y
Func ional
Rela ed Tasks
T4.1 - Real-Time Edge Compu ing o Visual Moni o ing
Rela ed Tool
Disas e de ec ion ool
Rela ed Use Cases
UC2.2 - Thessaloniki, G eece
MoSCoW Scale
mus -ha e
Dependencies
Dependence on exis ing da ase s o aining he deep lea ning
model
Came a's ield o iew and angle
P o ision o da a
Inpu Needed
F ames o ideos wi h high ame a e
Ve i ica ion Me hod
T aining he deep lea ning model, and pe o ming in e ence on
p o ided samples, as pa o in eg a ion es ing
Commen s
I is impo an o p o ide us wi h da a om he ield o in e es
ha includes he ele an objec s o in e es
ID
TR-FUN-4.1-8
Name (Op ional)
C ashed ca de ec ion ool
Desc ip ion
The sys em shall execu e an objec de ec ion algo i hm on
embedded de ices o iden i y c ashed ehicles in eal ime using
ideo eeds om CCTV came as o UAVs. I analyses he image y
o de ec unusually s a iona y ehicles o collisions on he oad
and p omp ly ale s eme gency se ices.
@iD i ing Conso ium – 101147004
Page 41 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Ca ego y
Func ional
Rela ed Tasks
T4.1 - Real-Time Edge Compu ing o Visual Moni o ing
Rela ed Tool
C ashed Ca de ec ion ool
Rela ed Use Cases
UC 3.2 - Bizkaia, Spain
MoSCoW Scale
mus -ha e
Dependencies
Dependence on exis ing da ase s o aining he deep lea ning
model
Came a's ield o iew and angle
P o ision o da a
Inpu Needed
F ames o ideos wi h high ame a e
Ve i ica ion Me hod
T aining he deep lea ning model, and pe o ming in e ence on
p o ided samples, as pa o in eg a ion es ing
Commen s
I is impo an o p o ide us wi h da a om he ele an ield
ha includes he objec s o in e es .
ID
TR-FUN-4.2-1
Name (Op ional)
Red ligh iola ion de ec o
Desc ip ion
De ec ehicles in eal- ime and de e mine i hey c oss a i ual
s op line du ing a ed a ic ligh phase.
Ca ego y
Func ional
Rela ed Tasks
Task 4.2
Rela ed Tool
Red Ligh Viola ion De ec o
Rela ed Use Cases
UR_UC1.1_17
MoSCoW Scale
Should-ha e
Dependencies
Real- ime ideo inpu s eam, accu a e a ic ligh s a e da a,
co ec i ual line con igu a ion.
Inpu Needed
Video s eam om su eillance came as, eal- ime a ic ligh
s a us da a. Ob aining his in o ma ion does no appea o be
easible. I canno be acqui ed h ough compu e ision, and i is
cu en ly unclea whe he access o he a ic ligh iming da a
will be possible o he in ended use case.
Ve i ica ion Me hod
Sys em alida ion wi h es ideo sequences including known
iola ion cases; c oss- e i ica ion o ale s and cap u ed e idence
(image + imes amp + a ic ligh s a us).
@iD i ing Conso ium – 101147004
Page 48 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Rela ed Tasks
TASK 4.4: Sma En i onmen al Condi ion Moni o ing o P oac i e
Road Sa e y Measu es
Rela ed Tool
Tool 4.4.2: Wea he P edic ion Tool using WRF and Da a
Assimila ion
Rela ed Use Cases
UC2.2 - Thessaloniki, G eece / UC 2.1 Ka lo ac, C oa ia
MoSCoW Scale
Mus -ha e
Dependencies
Gi , docke , Web F amewo k (like Django Res F amewo k),
Rela ional Da abase, ei he locally ins alled Open Sou ce LLM o
Cloud P o ided LLM (Accessed ia API), Py hon
Inpu Needed
S a ion Real Time Da a, Fo ecas Da a, Use Case da a
Ve i ica ion Me hod
Use Feedback
Commen s
-
ID
TR-FUN-4.5-1
Name (Op ional)
3D Recons uc ion o Road En i onmen s
Desc ip ion
The ool ans o ms 2D isual da a cap u ed by UAVs o g ound-
based mo ing came as in oad en i onmen s in o 3D
ep esen a ions, enhancing sa e y, enabling emo e inspec ions,
and p o iding aluable main enance insigh s. The scale o he
scene ex ended om a single de ec in a small-scale scene o a
la ge pa i ion o he oad ep esen ing b oade en i onmen s.
The econs uc ion me hod mus be adap ed acco ding o he
scale o he scene, whe he i is a localized de ec o a la ge oad
segmen . Also, he ool econs uc s scenes wi h en iched
in o ma ion, enhancing scene unde s anding and highligh ing
po en ial oad de ec s in 3D space.
Ca ego y
Func ional, Design
Rela ed Tasks
T4.5 AI-d i en 3D Scene Gene a ion o Sa e y & Main enance
Moni o ing
Rela ed Tool
3D-Sma Tool
Rela ed Use Cases
UC 2.1 Ka lo ac, C oa ia
MoSCoW Scale
should-ha e
Dependencies
• High quali y isual da a acqui ed om UAVs o g ound-
based mo ing came as.
• Inpu om Task T4.3 will be used o de ec iden i ica ion o
apply mul i esolu ion capabili ies in he egion
su ounding de ec ed anomalies
@iD i ing Conso ium – 101147004
Page 49 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Inpu Needed
• High- esolu ion ideo o sequen ial high- esolu ion
images.
• Inpu om Task T4.3
Ve i ica ion Me hod
Quali y me ics agains use s' pe cep ion, isual ideli y agains
g ound u h, ende ing quali y me ics, p ojec ion e o s
Commen s
The ool is highly dependen on he da a cap u e. The en i e a ge
a ea mus be ho oughly co e ed om mul iple iewpoin s o
ensu e accu a e econs uc ion. In he case o g ound-based
mo ing cap u e, he came a should emain as s able as possible
while in mo ion. Ex ac ed images should ha e signi ican o e lap
(ideally 70%).
The ou pu will be isualized acco ding o he me hod ha will be
used. Possible op ions:
Local b owse link o in e ac i e iewing
Expo o 3D Gaussians o .ply ile o ex e nal ools
Me hod’s in eg a ed SIBR iewe
ID
TR-FUN-4.5-2
Name (Op ional)
3D Recons uc ion o Ca Acciden
Desc ip ion
The ool ans o ms 2D isual da a cap u ed by UAVs a po en ial
ca acciden s scenes in o 3D eplica ions, enhancing sa e y,
enabling emo e inspec ions, and p o iding aluable inciden
moni o ing.
Ca ego y
Func ional, Design
Rela ed Tasks
T4.5 AI-d i en 3D Scene Gene a ion o Sa e y & Main enance
Moni o ing
Rela ed Tool
3D-Sma Tool
Rela ed Use Cases
UC 3.2 Bizkaia, Spain
MoSCoW Scale
should-ha e
Dependencies
High quali y isual da a acqui ed om UAVs o g ound-based
mo ing came as.
Inpu Needed
High- esolu ion ideo o sequen ial high- esolu ion images.
Ve i ica ion Me hod
Quali y me ics agains use s' pe cep ion, isual ideli y agains
g ound u h, ende ing quali y me ics, p ojec ion e o s
Commen s
The ool is highly dependen on he da a cap u e. The en i e a ge
a ea mus be ho oughly co e ed om mul iple iewpoin s o
ensu e accu a e econs uc ion. In he case o g ound-based
@iD i ing Conso ium – 101147004
Page 50 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
mo ing cap u e, he came a should emain as s able as possible
while in mo ion. Ex ac ed images should ha e signi ican o e lap
(ideally 70%).
The ou pu will be isualized acco ding o he me hod ha will be
used. Possible op ions:
Local b owse link o in e ac i e iewing
Expo o 3D Gaussians o .ply ile o ex e nal ools
Me hod’s in eg a ed SIBR iewe
3.1.3 Requi emen s ela ed o WP5
ID
TR-FUN-5.1-1
Name (Op ional)
Da a in eg a ion ool
Desc ip ion
The sys ems mus ake accoun mul i-sou ce da a om ID i ing
moni o ed ne wo k. Da a mus be adap ed and p ocess o
con inuous lea ning o KPI de ined in SCC
Ca ego y
Design, in e ope abili y
Rela ed Tasks
Dynamic Upda e o C i e ia Ca alogue Th ough Con inuous
Lea ning
Rela ed Tool
Dynamic moni o ing ool
Rela ed Use Cases
UC1.1, UC1.2, UC2.1, UC2.2, UC3.1, UC3.2
MoSCoW Scale
mus -ha e
Dependencies
Sa e y C i e ia Ca alogue
Inpu Needed
Da a om UCs ela i e o KPI de ine in SCC (da a equency,
quali y, o minima)
Ve i ica ion Me hod
In eg a ion es ing, unc ional es ing
Commen s
-
ID
TR-FUN-5.1-2
Name (Op ional)
KPI compu a ion and s o age
Desc ip ion
The sys em mus compu e KPI and lea n indica o s cha ac e is ics
based on in as uc u e and his o ical measu emen s. KPI a e
de ined inside a SCC py hon da abase. The esul s a e s o ed o
e alua ion, lea ning and isualisa ion.
Ca ego y
Func ional
@iD i ing Conso ium – 101147004
Page 51 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Rela ed Tasks
Dynamic Upda e o C i e ia Ca alogue Th ough Con inuous
Lea ning
Rela ed Tool
Dynamic moni o ing ool
Rela ed Use Cases
UC1.1, UC1.2, UC2.1, UC2.2, UC3.1, UC3.2
MoSCoW Scale
mus -ha e
Dependencies
Sa e y C i e ia Ca alogue
Inpu Needed
His o ical da a om moni o ed a eas
Ve i ica ion Me hod
In eg a ion es ing, unc ional es ing
Commen s
-
ID
TR-FUN-5.1-3
Name (Op ional)
KPI moni o ing dashboa d
Desc ip ion
The sys em shall p o ide a dashboa d o isualizing KPI ime
se ies, suppo ing mul iple isualiza ion o ma s such as maps,
hea maps, and g aphs.
Ca ego y
Func ional
Rela ed Tasks
Dynamic Upda e o C i e ia Ca alogue Th ough Con inuous
Lea ning
Rela ed Tool
Dynamic moni o ing ool
Rela ed Use Cases
UC1.1, UC1.2, UC2.1, UC2.2, UC3.1, UC3.2
MoSCoW Scale
mus -ha e
Dependencies
In e ne connec ion (API)
Inpu Needed
KPI da a
Ve i ica ion Me hod
In eg a ion es ing, unc ional es ing
Commen s
-
ID
TR-FUN-5.1-4
Name (Op ional)
KPI Con inuous lea ning
Desc ip ion
The sys em shall suppo o ecas ing and e alua ion o KPIs based
on incoming da a s eams.
Ca ego y
Func ional
Rela ed Tasks
Dynamic Upda e o C i e ia Ca alogue Th ough Con inuous
Lea ning
Rela ed Tool
Dynamic moni o ing ool
@iD i ing Conso ium – 101147004
Page 52 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Rela ed Use Cases
UC1.1, UC1.2, UC2.1, UC2.2, UC3.1, UC3.2
MoSCoW Scale
mus -ha e
Dependencies
Inpu Needed
KPI da a, T a ic da a
Ve i ica ion Me hod
In eg a ion es ing, unc ional es ing
Commen s
-
ID
TR-FUN-5.3-1
Name (Op ional)
Da a in eg a ion o digi al wins
Desc ip ion
The sys ems mus ake accoun mul i-sou ce da a om ID i ing
moni o ed ne wo k. Da a mus be adap ed and p ocess o send
o digi al win and s o e o pe o m o line calib a ion o digi al
win simula ion ools
Ca ego y
Func ional, in e ope abili y
Rela ed Tasks
Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning
Sys ems
Rela ed Tool
SUMO/CARLA
Rela ed Use Cases
UC1.1, UC1.2, UC2.2, UC3.1, UC3.2
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
T a ic Da a om UCs
Da a/e en s om g ound came a o UAVs
Wea he da a o UC2.2
Ve i ica ion Me hod
In eg a ion es ing, unc ional es ing
Commen s
-
ID
TR-FUN-5.3-2
Name (Op ional)
Digi al win calib a ion
Desc ip ion
The sys em shall build a digi al win o oad ne wo ks in
moni o ed a eas o simula e oad use beha iou based on eal
da a. The esul ing models shall be s o ed and made a ailable
o euse wi hin he iD i ing p ojec .
Ca ego y
Func ional
Rela ed Tasks
Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning
Sys ems
@iD i ing Conso ium – 101147004
Page 53 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Rela ed Tool
SUMO/CARLA, T3.2
Rela ed Use Cases
UC1.1, UC1.2, UC2.2, UC3.1, UC3.2
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
T a ic Da a om UCs
Ve i ica ion Me hod
In eg a ion es ing, unc ional es ing
Commen s
-
ID
TR-FUN-5.3-3
Name (Op ional)
Mul i oad use simula ion
Desc ip ion
The sys em shall use a simula ion-based digi al win o pe o m
a ic p edic ion based on eal- ime and his o ical da a. I shall
gene a e use ajec o ies unde a ious condi ions and
con igu a ions.
Ca ego y
Func ional
Rela ed Tasks
Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning
Sys ems
Rela ed Tool
SUMO
Rela ed Use Cases
UC1.1, UC1.2, UC2.2, UC3.1, UC3.2
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
T a ic da a, SUMO calib a ed models
Ve i ica ion Me hod
In eg a ion es ing, unc ional es ing
Commen s
-
ID
TR-FUN-5.3-4
Name (Op ional)
In e sec ion sa e y simula ion
Desc ip ion
The sys em shall pe o m a 3D simula ion o c i ical a ea o he
ne wo k (e.g. in e sec ions) In o de o e alua e isk o oad-
use s.
Ca ego y
Func ional
Rela ed Tasks
Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning
Sys ems
Rela ed Tool
CARLA
@iD i ing Conso ium – 101147004
Page 54 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Rela ed Use Cases
UC1.1, UC1.2, UC2.2, UC3.1, UC3.2
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
T a ic da a, SUMO calib a ed models
Ve i ica ion Me hod
In eg a ion es ing, unc ional es ing
Commen s
-
ID
TR-FUN-5.3-5
Name (Op ional)
Sa e y analysis and communica ion sys em
Desc ip ion
F om simula ion da a, p o ide ale s and sa e y measu es o
a ge ed ne wo k o iD i ing p o o ype. Ale s a e s o ed o
pe o m sa e y assessmen and ool e alua ion.
Ca ego y
Func ional
Rela ed Tasks
Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning
Sys ems
Rela ed Tool
SUMO/CARLA
Rela ed Use Cases
UC1.1, UC1.2, UC2.2, UC3.1, UC3.2
MoSCoW Scale
mus -ha e
Dependencies
-
Inpu Needed
T a ic da a, SUMO calib a ed models
Ve i ica ion Me hod
In eg a ion es ing, unc ional es ing
Commen s
-
ID
TR-FUN-5.4-1
Name (Op ional)
In as uc u e Segmen Risk Calcula ion
Desc ip ion
The sys em shall calcula e he dynamic isk o each in as uc u e
segmen using bo h his o ical and eal- ime da a, applying an
FMEA-based isk assessmen me hodology. The isk calcula ion
mus conside oad cha ac e is ics, a ic low, and s uc u al
ulne abili y in o ma ion. This isk e alua ion enables simula ion
o wha -i scena ios and suppo s bo h he de ini ion o
main enance needs and he planning o op imal main enance
in e en ions.
Ca ego y
Func ional
Rela ed Tasks
T5.4
@iD i ing Conso ium – 101147004
Page 55 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Rela ed Tool
Dynamic Risk Assessmen Tool
Rela ed Use Cases
UC2.1_4 (Ka lo ac), UC2.1_5
MoSCoW Scale
Mus -ha e
Dependencies
Requi es inpu om condi ion moni o ing ools (de ec ed
s uc u al ulne abili ies), a ic models and oad cha ac e is ic
da a. Da a in eg a ion wi h Tool 1.2 (Logis ics Planning Tool)
Inpu Needed
Real- ime oad cha ac e is ics, a ic low da a, ulne abili y da a
and scena io simula ion esul s (op ionally including damage
in o ma ion)
Ve i ica ion Me hod
Valida ion o calcula ed isk me ics agains expe assessmen s
o ield da a; Func ional es s e i ying eal- ime isk upda es
when inpu condi ions (e.g., a ic o ulne abili y le els) change;
Re iew o main enance ecommenda ions and plans gene a ed
based on isk indica o s
Commen s
This equi emen is a key enable o p oac i e main enance
s a egies based on Digi al Twin a chi ec u e. I ensu es isk
calcula ions a e dynamically adap ed o e ol ing oad and a ic
condi ions and p o ides essen ial inpu o logis ics and
main enance planning ools.
ID
TR-FUN-5.4-2
Name (Op ional)
Dynamic calcula ion o Heal h Index / Road Condi ion Index (RCI)
Desc ip ion
The sys em mus dynamically compu e Heal h Index o Road
Condi ion Index alues based on eal- ime ( om condi ion
moni o ing and de ec ing oad de ec s) and o ecas ed da a,
wea he , and a ic.
Ca ego y
Func ional
Rela ed Tasks
T5.4
Rela ed Tool
Heal h and Logis ic Managemen Tool
Rela ed Use Cases
UC2.1_4 (Ka lo ac)
MoSCoW Scale
Mus Ha e
Dependencies
In eg a ion wi h senso -based da a acquisi ion sys ems—
p ima ily imaging de ices— o oad su ace inspec ion; execu ion
o compu e ision algo i hms o de ec and classi y oad su ace
de ec s; and inco po a ion o wea he and a ic o ecas ing
pipelines o enable dynamic heal h index compu a ion
@iD i ing Conso ium – 101147004
Page 56 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Inpu Needed
Real- ime o nea eal- ime moni o ing and p ocessing o oad
condi ion da a o es ima e he in as uc u e's heal h s a us,
complemen ed by me eo ological and a ic da a o enhance he
s a is ical accu acy o he heal h index es ima ion.
Ve i ica ion Me hod
Compa ison agains ac ual condi ion measu emen s in pilo
scena ios o simula ed ou comes o m Digi al Twins model
Commen s
C i ical o enabling p oac i e and cos -e ec i e main enance.
Enables downs eam planning and isk analysis ia connec ed
ools (Module 1.1 and Module 1.3)
ID
TR-FUN-5.4-3
Name (Op ional)
Main enance Scheduling and Op imiza ion
Desc ip ion
The sys em mus gene a e sho - and long- e m main enance plans
based on he dynamic isk, heal h indica o s, a ic low o ecas s,
wea he condi ions, and a ailable esou ces. I mus balance isk
educ ion, main enance cos s, and esou ce op imiza ion by
applying me aheu is ic op imiza ion algo i hms and p ede ined
ope a ional cons ain s.
Ca ego y
Func ional
Rela ed Tasks
T5.4
Rela ed Tool
Main enance Scheduling Tool
Rela ed Use Cases
UC2.1_4 (Ka lo ac), UC2.1_5
MoSCoW Scale
Mus -ha e
Dependencies
Requi es inpu om he Dynamic Risk Assessmen Tool (Module 1.1)
and Heal h and Logis ics Managemen Tool (Module 1.2); Requi es
access o ope a ional main enance plans and cons ain s da abases;
Depends on eal- ime upda es o a ic and wea he o ecas s.
Inpu Needed
Risk me ics, heal h indica o s (RCI/Heal h Index), a ic o ecas s,
wea he o ecas s, p ede ined main enance plans, a ailable
esou ce in o ma ion
Ve i ica ion Me hod
Valida ion h ough simula ion o gene a ed main enance plans
agains expec ed op imiza ion goals (cos s isk ade-o s); Re iew
o sho - e m and long- e m plans o e i y co ec p io i iza ion and
esou ce assignmen ; Tes ing o expo ing unc ionali y (CSV/JSON
o ma s).
Commen s
C i ical o ensu e ha main enance in e en ions a e no only isk-
d i en bu also op imized o cos s and ope a ional easibili y. The
ool mus allow dynamic e-scheduling in case o upda ed isks o
unexpec ed a ic/wea he e en s. I is a key ou pu componen o
@iD i ing Conso ium – 101147004
Page 57 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
suppo ing oad manage s in eal- ime and s a egic decision-
making.
ID
TR-FUN-5.5-1
Name (Op ional)
In e ace o Digi al-Twin based con ol cen e
Desc ip ion
The sys em shall p o ide a use in e ace o he Digi al Twin-based
Con ol Cen e , compa ible wi h bo h desk op applica ions and XR
de ices.
Ca ego y
Func ional, Design, in e ope abili y, secu i y, egula o y/compliance
Rela ed Tasks
T5.5 Digi al Twin-Based Con ol Cen e wi h XR ea u es o
Enhanced Si ua ional Awa eness
Rela ed Tool
Digi al-Twin based con ol cen e
Rela ed Use Cases
UC1.1, UC1.2, UC2.1, UC2.2, UC3.1, UC3.2
MoSCoW Scale
Mus -ha e
Dependencies
In e ace depends on he de ice pe o mance
Da a ecei ed om he o he componen s (cen alized a
se e le el)
Inpu Needed
- Use equi emen s lis
Ve i ica ion Me hod
Usabili y es ing
Commen s
-
ID
TR-FUN-5.5-2
Name (Op ional)
Ale ing and communica ion se ice
Desc ip ion
The sys em shall include a dedica ed se ice o bidi ec ional
communica ion wi h he main se e , handling he sending and
ecei ing o messages, including ale s and no i ica ions.
Ca ego y
Func ional, Pe o mance, in e ope abili y, secu i y,
egula o y/compliance
Rela ed Tasks
T5.5 Digi al Twin-Based Con ol Cen e wi h XR ea u es o
Enhanced Si ua ional Awa eness
Rela ed Tool
Digi al-Twin based con ol cen e
Rela ed Use Cases
UC1.1, UC1.2, UC2.1, UC2.2, UC3.1, UC3.2
MoSCoW Scale
Mus -ha e
Dependencies
In e ne Connec ion, Ex e nal se ices (e.g. wea he in o ma ion)
@iD i ing Conso ium – 101147004
Page 64 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
UR_UC1.2_31
TR-FUN-3.3-2, TR-FUN-4.1-6, TR-FUN-4.2-3, TR-FUN-5.5-2,
TR-FUN-5.3-5
UR_UC1.2_32
TR-FUN-3.3-2, TR-FUN-4.2-4, TR-FUN-5.5-2, TR-FUN-5.3-5
UR_UC2.1_1
TR-FUN-4.3-1, TR-FUN-4.3-2, TR-FUN-5.4-1
UR_UC2.1_2
TR-FUN-5.1-3, TR-FUN-4.3-1, TR-FUN-4.3-2
UR_UC2.1_3
TR-FUN-5.5-3
UR_UC2.1_4
TR-FUN-5.4-1
UR_UC2.1_5
TR-FUN-5.4-3
UR_UC2.1_6
TR-FUN-5.5-3
UR_UC2.1_7
TR-FUN-5.5-1
UR_UC2.1_8
TR-FUN-3.2-2, TR-FUN-5.4-1
UR_UC2.1_9
TR-FUN-5.5-2
UR_UC2.1_10
TR-FUN-5.5-2
UR_UC2.1_11
TR-FUN-5.5-2
UR_UC2.1_12
TR-FUN-3.5-1, TR-FUN-3.5-2, TR-FUN-3.5-3, TR-NFUN-3.5-4,
TR-FUN-4.2-2
UR_UC2.1_13
TR-FUN-3.5-1, TR-FUN-3.5-2, TR-FUN-3.5-3, TR-NFUN-3.5-4,
TR-FUN-4.2-3
UR_UC2.1_14
TR-FUN-3.5-1, TR-FUN-3.5-2, TR-FUN-3.5-3, TR-NFUN-3.5-4
UR_UC2.1_15
TR-FUN-5.5-1
UR_UC2.1_17
TR-FUN-5.5-2
UR_UC2.1_19
TR-FUN-4.5
UR_UC2.2_1
TR-FUN-3.3-2
UR_UC2.2_2
TR-FUN-3.3-2
UR_UC2.2_3
TR-FUN-3.3-2
UR_UC2.2_4
TR-FUN-3.3-2
UR_UC2.2_5
TR-FUN-3.3-2
UR_UC2.2_6
TR-FUN-3.3-1
UR_UC2.2_7
TR-FUN-3.3-1
UR_UC2.2_8
TR-FUN-3.3-1
UR_UC2.2_9
TR-FUN-3.3-1
@iD i ing Conso ium – 101147004
Page 65 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
UR_UC2.2_10
TR-FUN-3.2-1
UR_UC2.2_11
TR-FUN-3.2-1
UR_UC2.2_12
TR-FUN-3.3-2
UR_UC2.2_13
TR-FUN-3.3-1
UR_UC2.2_14
TR-FUN-3.3-2
UR_UC2.2_15
TR-FUN-3.2-1
UR_UC2.2_16
TR-FUN-3.2-1
UR_UC2.2_17
TR-FUN-5.5-3
UR_UC2.2_18
TR-FUN-5.5-1
UR_UC2.2_19
TR-FUN-5.5-1
UR_UC2.2_20
TR-FUN-5.5-3
UR_UC2.2_23
TR-FUN-5.5-3
UR_UC2.2_24
TR-FUN-5.5-3
UR_UC2.2_25
TR-FUN-5.5-3
UR_UC2.2_26
TR-FUN-5.5-3
UR_UC2.2_27
TR-FUN-4.4-1, TR-FUN-4.4-5
UR_UC2.2_28
TR-FUN-4.4-1, TR-FUN-4.4-5
UR_UC2.2_29
TR-FUN-4.4-1, TR-FUN-4.4-5
UR_UC2.2_30
TR-FUN-4.4-1, TR-FUN-4.4-5
UR_UC2.2_31
TR-FUN-4.4-1, TR-FUN-4.4-5
UR_UC2.2_32
TR-FUN-4.4-1, TR-FUN-4.4-5
UR_UC2.2_33
TR-FUN-3.5-1, TR-FUN-3.5-2, TR-FUN-3.5-3, TR-NFUN-3.5-4,
TR-FUN-4.1-7
UR_UC2.2_34
TR-FUN-3.5-1, TR-FUN-3.5-2, TR-FUN-3.5-3, TR-NFUN-3.5-4,
TR-FUN-4.1-7
UR_UC2.2_35
TR-FUN-3.5-1, TR-FUN-3.5-2, TR-FUN-3.5-3, TR-NFUN-3.5-4,
TR-FUN-4.1-7
UR_UC2.2_36
TR-FUN-3.5-1, TR-FUN-3.5-2, TR-FUN-3.5-3, TR-NFUN-3.5-4
UR_UC3.1_1
TR-FUN-4.2-4
UR_UC3.1_2
TR-FUN-4.2-4
UR_UC3.1_3
TR-FUN-4.2-2, TR-FUN-4.2-4
@iD i ing Conso ium – 101147004
Page 66 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
UR_UC3.1_6
TR-FUN-3.3-2
UR_UC3.1_7
TR-FUN-3.3-1, TR-FUN-3.3-2, TR-NFUN-3.4.4, TR-FUN-5.5-2
UR_UC3.1_8
TR-FUN-3.3-1
UR_UC3.1_13
TR-FUN-5.3-5
UR_UC3.1_14
TR-FUN-3.3-1
UR_UC3.1_16
TR-FUN-3.3-2, TR-FUN-5.3-5, TR-FUN-5.5-2
UR_UC3.1_18
TR-FUN-3.3-1, TR-FUN-3.3-2
UR_UC3.1_21
TR-FUN-5.1-3, TR-FUN-5.5-1, TR-FUN-5.5-3
UR_UC3.1_22
TR-FUN-5.1-2, TR-FUN-5.1-3, TR-FUN-5.5-3
UR_UC3.1_23
TR-FUN-3.3-2, TR-NFUN-3.4.4, TR-FUN-5.5-2
UR_UC3.1_26
TR-NFUN-GEN-9
UR_UC3.1_27
TR-FUN-3.3-1
UR_UC3.1_28
TR-FUN-5.3-5
UR_UC3.2_1
TR-FUN-3.2-1, TR-FUN-3.3-1, TR-FUN-3.3-2
UR_UC3.2_2
TR-FUN-3.3-2
UR_UC3.2_3
TR-FUN-3.3-1, TR-FUN-3.3-2
UR_UC3.2_4
TR-FUN-3.3-1
UR_UC3.2_5
TR-FUN-3.2-1, TR-FUN-3.3-1
UR_UC3.2_6
TR-FUN-3.2-1, TR-FUN-3.3-2
UR_UC3.2_7
TR-FUN-4.1-8
UR_UC3.2_8
TR-FUN-4.1-8, TR-FUN-4.2-4
UR_UC3.2_9
TR-FUN-3.5-1, TR-FUN-3.5-2, TR-FUN-3.5-3, TR-NFUN-3.5-4
UR_UC3.2_10
TR-FUN-4.5
UR_UC3.2_11
TR-FUN-5.1-2, TR-FUN-5.1-3, TR-FUN-5.1-4
UR_UC3.2_12
TR-FUN-3.3-1, TR-FUN-3.3-2, TR-FUN-4.1-8
UR_UC3.2_13
TR-FUN-3.3-1
UR_UC3.2_14
TR-FUN-3.3-2
@iD i ing Conso ium – 101147004
Page 67 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5 O e all A chi ec u e & Componen s
5.1 Me hodology o De ining he A chi ec u e
The iD i ing a chi ec u e adop s he C4 Model, a widely ecognized me hodology
o he isualiza ion and communica ion o complex so wa e sys ems. This
app oach enables clea , laye ed ep esen a ions ha add ess di e en s akeholde
conce ns, anging om he o e all business con ex o de ailed echnical designs.
The C4 Model, speci ically selec ed o his deli e able, p o ides he ollowing
a chi ec u al pe spec i es:
• Con ex View: Illus a es he sys em as a “black box” in i s en i onmen ,
iden i ying all majo ex e nal ac o s and hei in e ac ions wi h iD i ing.
• Con aine View: B eaks down iD i ing in o i s main echnology con aine s
(applica ions, da abases, se ices), ocusing on he esponsibili ies and
in eg a ion poin s be ween hem o each use case.
• Componen View: (De ailed pe -con aine ) P o ides a u he b eakdown o
how key con aine s a e in e nally s uc u ed and how hei sub-componen s
collabo a e o ealize unc ional equi emen s.
• Code View: (Omi ed om his deli e able) Rese ed o implemen a ion-
le el de ails.
This laye ed app oach ensu es all audiences— echnical, manage ial, and
ope a ional—can clea ly unde s and he sys em a he igh le el o abs ac ion o
hei needs. Fu he mo e, he model’s modula i y allows us o a icula e bo h
common pa e ns ac oss use cases and bespoke design elemen s ha add ess
unique local challenges.
5.2 iD i ing Con ex View
The Con ex View (see Figu e 2) p esen s a high-le el o e iew o he iD i ing
ecosys em, isualizing i s posi ion wi hin he wide ope a ional and s akeholde
landscape. This diag am iden i ies he main ca ego ies o ex e nal ac o s—T a ic
Au ho i ies / Road Ope a o s, UAV Ope a o s / Main enance Teams, and Road
Use s—and delinea es hei p ima y da a exchanges wi h he iD i ing sys em.
Pu pose:
The Con ex View es ablishes he p ojec ’s sys em bounda ies, cla i ies oles and
esponsibili ies, and se s he s age o unde s anding subsequen a chi ec u al
de ails. I ensu es ha all p ojec s akeholde s— om echnical eams o
managemen and end use s—sha e a mu ual unde s anding o who in e ac s wi h
iD i ing and wha he essen ial in o ma ion lows a e.
@iD i ing Conso ium – 101147004
Page 68 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Figu e 2 - iD i ing Con ex View
5.3 iD i ing Con aine s View
The Con aine View diag ams p o ide a ocused b eakdown o he iD i ing sys em
in o i s majo unc ional and echnological building blocks (“con aine s”). These
diag ams cap u e he co e a chi ec u e o iD i ing, p esen ing a laye ed
ep esen a ion ha o ganizes sys em componen s acco ding o hei oles in he
o e all wo k low:
• Da a Acquisi ion: Edge de ices, senso s, UAVs, and o he sou ces
esponsible o ga he ing aw da a om he ield.
• De ec ion & P edic ion: AI/ML ools and isual analy ics modules ha
ans o m aw da a in o ac ionable insigh s and ea ly wa nings.
• Co e Back-End & Op imiza ion: Cen alized se ices, o ches a ion engines,
and simula ion ools, which p ocess, s o e, and op imize sys em ou pu s.
• Use -Facing & Con ol: Applica ions, dashboa ds, and con ol cen es ha
deli e insigh s and ins uc ions o end use s and ope a o s.
Colou Coding and Laye ing:
Each con aine in he diag am is colo -coded o e lec i s alignmen wi h speci ic
Tasks, ensu ing immedia e isual cla i y ega ding componen owne ship and
unc ional g ouping. Laye ed s uc u ing ( e ical sepa a ion) emphasizes he low
o da a and con ol om acquisi ion h ough o ac ionable ou comes.
@iD i ing Conso ium – 101147004
Page 69 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Pu pose:
The Con aine s View enables all s akeholde s o g asp b ie ly and immedia ely,
which so wa e and ha dwa e modules a e ac i e in a gi en use case, how da a
mo es be ween hem, and how he o e all sys em achie es i s unc ional and non-
unc ional equi emen s. This app oach acili a es aceabili y om use needs o
echnical solu ions, suppo s impac analysis o changes, and p o ides a s ong
basis o u he de ailed design in he Componen s’ View.
5.3.1 Con aine s View – Consolida ed
This sec ion p o ides a uni ied Con aine View diag am ha agg ega es he
common a chi ec u al elemen s ac oss all use cases. I highligh s he sha ed
in as uc u e, co e backend se ices, and c oss-cu ing componen s (e.g.,
Da aspace Connec o , Ka ka b oke ) ha unde pin he iD i ing ecosys em. This
consolida ed iew suppo s sys em-wide in eg a ion and euse.
@iD i ing Conso ium – 101147004
Page 70 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Figu e 3 - Consolida ed iD i ing Con aine s iew
Figu e 4 -
Con aine s View
Colou Legend
@iD i ing Conso ium – 101147004
Page 71 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5.3.2 Con aine s View Pe Use Case
While he consolida ed iew ou lines an o e all logical iew o he sys em, each pilo
deploymen has i s own a chi ec u al nuances. The ollowing sub-sec ions p esen
use-case-speci ic Con aine View diag ams, showcasing localized deploymen
pa e ns, da a lows, and componen s ailo ed o he pa icula con ex and
objec i es o each si e.
5.3.2.1 Use case 1.1 G az, Aus ia
Con aine View diag am o UC 1.1 G az is p esen ed in Figu e 5.
Figu e 5 - Use Case 1.1 Con aine s View
@iD i ing Conso ium – 101147004
Page 72 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5.3.2.2 Use Case 1.2 Ne e s, F ance
Con aine View diag am o UC 1.2 Ne e s is p esen ed in Figu e 6.
Figu e 6 - Use Case 1.2 Con aine s View
@iD i ing Conso ium – 101147004
Page 73 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5.3.2.3 Use Case 2.1 Ka lo ac, C oa ia
Con aine View diag am o UC 2.1 Ka lo ac is p esen ed in Figu e 7.
Figu e 7 - Use Case 2.1 Con aine s View
@iD i ing Conso ium – 101147004
Page 80 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
@iD i ing Conso ium – 101147004
Page 81 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Expec ed TRL
TRL
Technologies
The ollowing echnologies will be used o he implemen a ion o he componen :
• SUMO a ic simula ion so wa e
Deploymen
P emises
The es ing o he a ious scena ios de ined will be done in simula ed en i onmen and/o in eal- ime implemen a ion.
The ou e guidance ool will be an add-on module o SUMO a ic simula o . The sys em will be es ed i ually on
se e s wi hin he pa ne s’ p emises.
In ol ed
Pa ne s
MBL; TUC
Rela ed
Technical
equi emen s
TR-FUN-3.2-1
@iD i ing Conso ium – 101147004
Page 82 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table 11 - Signal Con ol Tool – A chi ec u al Design and Desc ip ion
Signal Con ol Tool: Sa e y-Op imized T a ic Managemen Using In elligen
Algo i hms o Di e se Vehicle Flows
Desc ip ion
The sys em ecei es in o ma ion om di e en sou ces (came as,
induc ion loop de ec o s, magne ome e senso s e c.). The signal
con ol ool p o ides dynamic adjus men o a ic ligh imings o
an iden i ied in e sec ion(s) based on he ac ual a ic in o ma ion
coming om he neighbou ing links o he in e sec ion. A each signal
cycle, measu emen s o es ima es o queue leng hs o accumula ed
ehicles a e eeding he con ol s a egy, whose main goal is o
imp o e e iciency o he ips se ed by he con olled in e sec ion(s).
The op imal signal imings a e compu ed in eal- ime and
b oadcas ed o he a ic ligh s.
@iD i ing Conso ium – 101147004
Page 83 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Expec ed TRL
TRL
Technologies
The ollowing echnologies will be used o he implemen a ion o he
Signal Con ol Componen :
• SUMO a ic simula ion so wa e
@iD i ing Conso ium – 101147004
Page 84 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Deploymen
P emises
The es ing o he a ious scena ios de ined will be done in a
simula ed en i onmen and/o in eal- ime implemen a ion. The
signal con ol ool will be an add-on module o he SUMO a ic
simula o . The sys em will be es ed i ually on se e s wi hin he
pa ne s’ p emises.
In ol ed
Pa ne s
MBL; TUC
Rela ed
Technical
equi emen s
TR-FUN-3.2-2
@iD i ing Conso ium – 101147004
Page 85 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table 12 - Mobile Applica ion – A chi ec u al Design and Desc ip ion
Mobile Applica ion: Mobile and in- ehicle applica ions o ea ly wa nings
Desc ip ion
A use -cen ic so wa e sys em ha se es bo h d i e s and non-d i ing oad use s. I connec s di ec ly o use ’s mobile
de ice (And oid sys em) and he Digi al Twin o e ch eal- ime da a on oad condi ions, a ic, and o he po en ial
haza ds
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
@iD i ing Conso ium – 101147004
Page 86 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Expec ed TRL
TRL6
Technologies
The ollowing echnologies will be used o he implemen a ion o he X Componen :
• Uni y
• C#
• REST API
• Pho oshop
• Blende 3D
Deploymen
P emises
And oid based mobile de ice wi h WiFi connec ion
In ol ed
Pa ne s
TUC
Rela ed
Technical
equi emen s
TR-FUN-3.3-1, TR-FUN-3.3-2
Table 13 - In-Vehicle Applica ion – A chi ec u al Design and Desc ip ion
In-Vehicle Applica ion: Mobile and in- ehicle applica ions o ea ly wa nings
Desc ip ion
A use -cen ic so wa e sys em ha se es d i e s. I connec s di ec ly o he ehicle’s buil in in o ainmen sys em
(And oid sys em) and he Digi al Twin o e ch eal- ime da a on oad condi ions, a ic, and o he po en ial haza ds
@iD i ing Conso ium – 101147004
Page 87 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Expec ed TRL
TRL6
Technologies
The ollowing echnologies will be used o he implemen a ion o he X Componen :
• Uni y
• C#
• REST API
@iD i ing Conso ium – 101147004
Page 88 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
• Pho oshop
• Blende 3D
Deploymen
P emises
In- ehicle And oid sys em wi h WiFi connec ion
In ol ed
Pa ne s
TUC
Rela ed
Technical
equi emen s
TR-FUN-3.3-1, TR-FUN-3.3-2
@iD i ing Conso ium – 101147004
Page 89 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table 14 - Ae ial Su eillance UAV Sys em – A chi ec u al Design and Desc ip ion
Ae ial Su eillance UAV sys em: Ae ial Su eillance in Inciden Managemen
and Main enance Tasks
Desc ip ion
A UAV sys em will be deli e ed, equipped wi h he abili y o hos
di e en senso s and came as (i.e. high- esolu ion came a, he mal
e c.) in o de o ensu e p ecise, e icien , e ec i e da a collec ion om
he a eas o in e es on demand and/o nea eal ime and add ess he
iD i ing speci ic needs in ials.
The UAV will be able o pe o m manual o au onomous ligh s by
p epa ing a ligh plan p io o he mission.
Fo he pilo use cases and in he case o he need o execu e he AI
Algo i hms in nea eal- ime, he ollowing wo app oaches a e
a ailable:
1. A NVIDIA AI-o ien ed compu e will be a ailable o moun ing
on he UAV whe e he payloads will in e ac wi h he onboa d
compu e o achie e he objec i es o each speci ic mission
acco ding o PUC equi emen s.
2. T ans e he ideo s eam in eal ime in he g ound s a ion
compu e whe e he AI algo i hms will be ins alled, execu ed,
and hen ans e he equi ed da a o any cloud/se e
needed.
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
@iD i ing Conso ium – 101147004
Page 96 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Technologies
The ollowing echnologies will be used o he implemen a ion o he
X Componen :
• Py hon
• Deep lea ning de ec ion models
Deploymen
P emises
Embedded on edge de ice
In ol ed
Pa ne s
MBL, UNI.EIFFEL
Rela ed
Technical
equi emen s
TR-FUN-4.1-5
Table 18 - License Pla e De ec ion Tool – A chi ec u al Design and Desc ip ion
License Pla e De ec ion Tool: Real-Time Edge Compu ing o Visual Moni o ing
Desc ip ion
This ool uns on embedded de ices and uses compu e ision
algo i hms o au oma ically iden i y license pla es om ideo oo age
once a a ic iola ion is de ec ed. A e ecognizing o enses such as
sea bel non-compliance o cell phone usage while d i ing, he ool
accu a ely cap u es he license pla e o he o ending ehicle.
@iD i ing Conso ium – 101147004
Page 97 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Expec ed TRL
TRL 6
Technologies
The ollowing echnologies will be used o he implemen a ion o he
X Componen :
• Py hon
• Deep lea ning de ec ion models
Deploymen
P emises
The ool will un on an embedded de ice such as he NVIDIA Je son
In ol ed
Pa ne s
SIMAVI, TEKNIKER, UNI.EIFFEL
Rela ed
Technical
equi emen s
TR-FUN-4.1-4,
@iD i ing Conso ium – 101147004
Page 98 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table 19 - Helme De ec ion Tool – A chi ec u al Design and Desc ip ion
Helme De ec ion Tool: Real-Time Edge Compu ing o Visual Moni o ing
Desc ip ion
This ool le e ages ad anced compu e ision algo i hms o
au oma ically de ec helme usage by mo o cyclis s o cyclis s in eal-
ime. I can un on embedded de ices connec ed o oad came as o
UAVs (d ones), enabling e ec i e moni o ing o helme compliance in
a ious en i onmen s. By ensu ing ide s wea helme s, he ool
p omo es sa e y and suppo s a ic egula ion en o cemen .
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Expec ed TRL
TRL 6
Technologies
The ollowing echnologies will be used o he implemen a ion o he
X Componen :
• Py hon
• Deep lea ning de ec ion models
@iD i ing Conso ium – 101147004
Page 99 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Deploymen
P emises
The ool will un on an embedded de ice such as he NVIDIA Je son
In ol ed
Pa ne s
SIMAVI, TEKNIKER, UNI.EIFFEL
Rela ed
Technical
equi emen s
TR-FUN-4.1-3
Table 20 - Zeb a C ossing De ec ion Tool – A chi ec u al Design and Desc ip ion
Zeb a C ossing De ec ion Tool: Real-Time Edge Compu ing o Visual
Moni o ing
Desc ip ion
The sys em shall de ec he p esence o zeb a c ossing in ideo ames
cap u ed by oadside came as o UAVs
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Expec ed TRL
TRL 6
@iD i ing Conso ium – 101147004
Page 100 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Technologies
The ollowing echnologies will be used o he implemen a ion o
he X Componen :
• Py hon
• Deep lea ning de ec ion models
Deploymen
P emises
The ool will un on an embedded de ice such as he NVIDIA Je son
In ol ed
Pa ne s
TEKNIKER
Rela ed
Technical
equi emen s
TR-FUN-4.1-6
Table 21 - Fallen T ee and Rockslide De ec ion Tool – A chi ec u al Design and Desc ip ion
Fallen ee and Rockslide Pla e De ec ion Tool: Real-Time Edge Compu ing o
Visual Moni o ing
Desc ip ion
This ool u ilizes compu e ision algo i hms o moni o landscapes
and oadways o allen ees and ockslides using ideo eeds om
UAVs (d ones). I analyses ae ial image y o quickly iden i y and assess
po en ial haza ds, enabling p omp epo ing and esponse.
@iD i ing Conso ium – 101147004
Page 101 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Expec ed TRL
TRL 6
Technologies
The ollowing echnologies will be used o he implemen a ion o he
X Componen :
• Py hon
• Deep lea ning de ec ion models
Deploymen
P emises
The ool will un on an embedded de ice such as he NVIDIA Je son
In ol ed
Pa ne s
MBL, UNI.EIFFEL, ACCELI
Rela ed
Technical
equi emen s
TR-FUN- 4.1-7
@iD i ing Conso ium – 101147004
Page 102 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table 22 - C ashed Vehicle De ec ion Tool – A chi ec u al Design and Desc ip ion
C ashed ehicle De ec ion Tool: Real-Time Edge Compu ing o Visual
Moni o ing
Desc ip ion
This ool employs compu e ision algo i hms o iden i y c ashed
ehicles in eal- ime using ideo eeds om CCTV came as o UAVs
(d ones). I analyses he image y o de ec unusual s a iona y ehicles
o collisions on he oad, p omp ly ale ing eme gency se ices.
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Expec ed TRL
TRL 6
Technologies
The ollowing echnologies will be used o he implemen a ion o he
X Componen :
• Py hon
• Deep lea ning de ec ion models
Deploymen
P emises
The ool will un on an embedded de ice such as he NVIDIA Je son
In ol ed
Pa ne s
ACCELI, UNI.EIFFEL, MBL, TEKNIKER
@iD i ing Conso ium – 101147004
Page 103 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Rela ed
Technical
equi emen s
TR-FUN-4.1-8
Table 23 - Red Ligh Viola ion De ec o – A chi ec u al Design and Desc ip ion
Red ligh iola ion de ec o : AI-Powe ed Beha iou al analysis o objec s o
in e es
Desc ip ion
An in elligen ision-based module ha moni o s ehicle beha iou a
in e sec ions. I de ec s i a ehicle has c ossed a i ual s op line
du ing a ed a ic ligh phase by combining eal- ime ehicle
de ec ion and acking, a ic ligh da a and ule-based iola ion
de ec ion logic. This componen suppo s e idence gene a ion o
en o cemen and eal- ime ale s o a ic managemen sys ems.
A chi ec u al
Diag am and
Subcomponen
s (componen s
View)
Subcomponen s:
- De ec o : Loca es ehicles in each ide ame.
- T acke : assigns unique IDs o each ehicle and main ains i s
ajec o y
- Viola ion Engine: e i ies i ual line c ossing, e alua es
whe he he c ossing occu ed du ing a ed a ic ligh phase,
and igge s an ale .
- Expo e : sa es e idence (image + imes amp + ID + a ic ligh
s a us)
Expec ed TRL
TRL 5-6
@iD i ing Conso ium – 101147004
Page 104 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Technologies
The ollowing echnologies will be used o he implemen a ion o he
Componen :
- The so wa e shall be implemen ed p ima ily in Py hon
- SQLi e, Pos g eSQL, o MongoDB o expo
Deploymen
P emises
On-p emises se ice
In ol ed
Pa ne s
CERTH, TEKNIKER, SIMAVI, UNIV-EIFFEL
Rela ed
Technical
equi emen s
TR-FUN-4.2-1
Table 24 - Imp ope Lane Usage De ec o – A chi ec u al Design and Desc ip ion
Imp ope line usage de ec o : AI-Powe ed Beha iou al analysis o objec s o
in e es
Desc ip ion
This module de ec s imp ope lane usage by ehicles in u ban
segmen s. The sys em analyses he ajec o y o each ehicle in eal-
ime o de e mine whe he i de ia es om i s assigned lane o en e s
es ic ed zones. I uses objec de ec ion, acking, and spa ial
easoning o iden i y iola ions.
@iD i ing Conso ium – 101147004
Page 105 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Subcomponen s:
- De ec o : Loca es ehicles in each ide ame.
- T acke : assigns unique IDs o each ehicle and main ains i s
ajec o y
- Viola ion Engine: alida es ehicle ajec o ies based on he
land zone. I compa es he ehicle’s ajec o y wi h he allowed
lane zones.
- Expo e : sa es e idence (image + imes amp + ID + lane in o)
Expec ed TRL
TRL 5-6
Technologies
The ollowing echnologies will be used o he implemen a ion o he
componen :
• Py hon
Deploymen
P emises
On-p emises
In ol ed
Pa ne s
CERTH, TEKNIKER, SIMAVI, UNIV-EIFFEL
Rela ed
Technical
equi emen s
TR-FUN-4.2-2
Table 25 - Zeb a C ossing Viola ion De ec o – A chi ec u al Design and Desc ip ion
Zeb a c ossing iola ion de ec o : AI-Powe ed Beha iou al analysis o objec s o
in e es
@iD i ing Conso ium – 101147004
Page 112 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
In ol ed
Pa ne s
DREVEN
Rela ed
Technical
equi emen s
TR-FUN-4.4-1/TR-FUN-4.4-2/TR-FUN-4.4-3
@iD i ing Conso ium – 101147004
Page 113 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table 29 - AI-Based Real-Time Wea he Ale Sys em – A chi ec u al Design and Desc ip ion
AI-Based Real-Time Wea he Ale Sys em: Sma En i onmen al Condi ion Moni o ing o P oac i e Road Sa e y Measu es
Desc ip ion
This sys em enables eal- ime and o ecas -based en i onmen al moni o ing by in eg a ing s a ion obse a ions and
wea he model p edic ions. Da a is made accessible o speci ic geog aphic a eas, and an au oma ed p ocess (scheduled ia
c on) checks i p ede ined haza d h esholds a e me . The ou comes a e p ocessed by an AI engine (LLM), which gene a es
ale s en iched wi h isk le els and wea he - ela ed da a. These ale s a e hen se ed h ough an API, making hem a ailable
o he iD i ing pla o m o enhanced decision-making and oad sa e y measu es.
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
@iD i ing Conso ium – 101147004
Page 114 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Expec ed TRL
TRL 6
Technologies
• Py hon
• Docke
• Gi
• Django Res F amewo k
• Rela ional Da abase (e.g., Pos g eSQL)
• Open Sou ce o Cloud-Based LLM ( ia API access)
Deploymen
P emises
DREVEN in-house se e s o Cloud in as uc u e
In ol ed
Pa ne s
DREVEN
Rela ed
Technical
equi emen s
TR-FUN-4.4-4
@iD i ing Conso ium – 101147004
Page 115 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table 30 - 3D-Sma Tool – A chi ec u al Design and Desc ip ion
3D-Sma Tool: AI-d i en 3D Scene Gene a ion o Sa e y & Main enance
Moni o ing
Desc ip ion
The ool aims o ans o m isual da a om oad en i onmen s in o
3D ep esen a ions, in eg a ing i in o he iD i ing ecosys em o
imp o ed sa e y, main enance insigh s and inciden /p og ess
moni o ing. This in ol es acqui ing da a using UAVs, p ocessing i
using Neu al Radiance Fields o Gaussian Spla ing, and enhancing
iD i ing's abili y o de ec and assess oad haza ds, imp o ing
si ua ional awa eness and emo e esponses o oad manage s.
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Expec ed TRL
TRL 6
Technologies
The ollowing echnologies will be used o he implemen a ion o he
3D-Sma Tool:
• 3D Gaussian Spla ing
• Neu al Radiance Field
@iD i ing Conso ium – 101147004
Page 116 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Deploymen
P emises
-
In ol ed
Pa ne s
ACCELIGENCE, TEKNIKER, SIMAVI
Rela ed
Technical
equi emen s
TR-FUN-4-5
@iD i ing Conso ium – 101147004
Page 117 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Table 31 - Dynamic Moni o ing Pla o m – A chi ec u al Design and Desc ip ion
Dynamic Moni o ing Pla o m: Dynamic Upda e o C i e ia Ca alogue Th ough Con inuous Lea ning
Desc ip ion
The ool aims o analyse and p edic he s a e o mul imodal anspo ne wo ks by in eg a ing eal- ime and heo e ical
da a. Using a gene ic and mul imodal da a model, i enables ne wo k moni o ing, diagnosis, simula ion, and o ecas ing
based on key pe o mance indica o s like conges ion, esilience, and speed. Wi h he abili y o in e ac wi h ex e nal ools ia
APIs and web se ices, i p o ides dynamic mapping, inciden diagnosis, and pe o mance isualiza ion, ensu ing enhanced
sa e y and main enance insigh s.
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
@iD i ing Conso ium – 101147004
Page 118 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Expec ed TRL
TRL 6
Technologies
The ollowing echnologies will be used o he implemen a ion o he Dynamic Upda e o C i e ia Ca alogue Componen :
1. Py hon
Lea ning models
Deploymen
P emises
-
In ol ed
Pa ne s
-
Rela ed
Technical
equi emen s
TR-FUN-5.1-1 TR-FUN-5.1-2, TR-FUN-5.1-3,
Table 32 - SUMO: Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning Sys ems – A chi ec u al Design and Desc ip ion
SUMO: Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning Sys ems
Desc ip ion
The ool aims o simula e and analyse in e modal a ic sys ems, in eg a ing oad ehicles, public anspo , and pedes ians
o enhanced a ic managemen . By u ilizing eal-wo ld a ic da a, oad opology, and SUMO add-ons like VANETT o
elecommunica ions, i enables de ailed a ic low simula ion, isk p edic ion, and haza d assessmen . The pla o m
suppo s eal- ime and la ge-scale simula ions, equi ing compu a ional esou ces based on se ice le els, making i a
aluable ool o op imizing u ban mobili y and sa e y.
@iD i ing Conso ium – 101147004
Page 119 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Expec ed TRL
TRL 6
Technologies
The ollowing echnologies will be used o he implemen a ion o he SUMO: Digi al Twin Componen :
2. Sumo o a ic simula ion o use case ne wo ks
3. Py hon o Sa e y assessmen
4. Lea ning algo i hms (Deep lea ning/ s a is ical me hods)
Geojson o sa e y ale s
Deploymen
P emises
-
@iD i ing Conso ium – 101147004
Page 120 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
In ol ed
Pa ne s
TUC
Rela ed
Technical
equi emen s
TR-FUN-5.3-*
Table 33 - CARLA: Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning Sys ems – A chi ec u al Design and Desc ip ion
CARLA: Digi al Twin Powe ed P edic i e Sa e y Measu es and Wa ning Sys ems
Desc ip ion
The ool aims o simula e and analyse a ic en i onmen s in 3D, p o iding insigh s o au onomous d i ing sys ems. By
in eg a ing eal-wo ld a ic da a, oad ypes, and obs acles, i enables isk p edic ion and haza d isualiza ion.
@iD i ing Conso ium – 101147004
Page 121 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
A chi ec u al
Diag am and
Subcomponen s
(componen s
View)
Expec ed TRL
TRL 6
Technologies
The ollowing echnologies will be used o he implemen a ion o he CARLA: Digi al Twin Componen :
• CARLA o simula ion o moni o ed a eas
• De elopmen s in Py hon o Sa e y assessmen
T ajec o y ecogni ion and p ocessing
T aining Deep lea ning/ s a is ical models
E alua ion o he models
Deploymen
P emises
-
@iD i ing Conso ium – 101147004
Page 128 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
A chi ec u al
Diag am and
Subcomponen
s (componen s
View)
Expec ed TRL
TRL6
Technologies
The ollowing echnologies will be used o he implemen a ion o he
X Componen :
• Uni y
• C#
• REST API
• Pho oshop
• Blende 3D
Deploymen
P emises
Lap op (Windows), XR glasses
In ol ed
Pa ne s
INTRA, INFRA PLAN, ALP.LAB, AIM, DREVEN, THESSALONIKI
Rela ed
Technical
equi emen s
TR-FUN-5.5-1, TR-FUN-5.5-2, TR-FUN-5.5-3
5.5 iDRIVING P ocess Views
The ollowing sec ion p esen s componen -le el low diag ams in he o m o
sys em sequence diag ams, con ibu ed by all pa ne s. Each diag am illus a es
he un ime in e ac ion be ween key elemen s o he espec i e componen . A
b ie desc ip ion accompanies each diag am o cla i y i s pu pose and in eg a ion
wi hin he iDRIVING a chi ec u e.
@iD i ing Conso ium – 101147004
Page 129 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5.5.1 Teknike Da aspace Connec o
A modula solu ion ha allows o ganiza ions o es ablish a single poin o en y o
accessing and exchanging da a wi hin a da a space. I ensu es in e ope abili y in
da a sha ing, os e s us among pa icipan s, and gua an ees da a so e eign y
h oughou i s en i e li ecycle.
Figu e 11 - Teknike da aspace connec o sequence diag am
Sequence diag am s eps, p esen ed in Figu e 11:
1. The Da a Owne de ines hei da a ca alog:
a. B ie ca alog desc ip ion
b. Ca alog loca ion (Da aSe ice)
c. Lis o da ase s:
i. Da ase loca ion (Da aAdd ess)
ii. Da a ype (Dis ibu ion)
iii. Access policies (O e )
2. The Da a Use eques s he ca alog
3. The Da a Use ’s connec o ini ia es he Ca alog P o ocol
4. The Da a Use ecei es and accep s he ca alog's access policies
(Con ac Nego ia ion P o ocol)
a. Bo h pa ies can moni o he con ac nego ia ion p ocess
5. The Da a Use eques s a da ase → he T ans e P ocess P o ocol s a s
6. Da a ans e depends on he chosen me hod (PUSH o PULL):
a. PULL: P ocessing ools access he da a di ec ly
b. PUSH: Da a is deli e ed o a de ined loca ion ia he p ocessing
ools
@iD i ing Conso ium – 101147004
Page 130 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5.5.2 Rou e Guidance ool
Desc ip ion o he sequence diag am: The p ocess s a s wi h senso s (came as,
GPS, oad senso s, e c.) de ec ing inciden s, conges ion, o haza dous condi ions.
This da a is analysed by he ela i e iD i ing componen , which ex ac s key
a ic pa ame e s. The esul s a e communica ed o he ITMS which hen
e alua es he si ua ion and de e mines op imal a ic managemen s a egies
(based on he simula ed scena ios), such as dynamic signal adjus men s and
ou e ecommenda ions based on ehicle ype. The sys em communica es wi h
a ic signals o implemen changes and p o ides guidance o oad use s
(d i e s, uck d i e s, mo o cyclis s, cyclis s) h ough he iD i ing use
communica ion channels (app, in- ehicle e c.). In c i ical si ua ions, an
Eme gency Response Sys em is also igge ed o eme gency ehicle d i e s o
ensu e imely in e en ion.
Figu e 12 - Rou e Guidance ool sequence diag am
Sequence diag am s eps, p esen ed in Figu e 12:
1. Recei e inpu : Da a om a ious sou ces is collec ed and p ocessed by he
ela i e iD i ing componen s. Use ul in o ma ion ega ding ac ual o
imminen a ic conges ion is ex ac ed and p o ided o he ou e guidance
algo i hm as inpu .
2. His o ical da a o a el demand, a ic conges ion, a ic signal se ings and
o he ne wo k- ela ed in o ma ion will be used o conduc ing SUMO
mic oscopic simula ion es s
@iD i ing Conso ium – 101147004
Page 131 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
3. Ale s and wa nings abou iden i ied inciden s (e.g. acciden s, loopholes o
in as uc u e ailu e, allen ees, ex eme wea he condi ions) a speci ic
loca ion and ime in he ne wo k ( oad, lane), which a e p o ided by he
ela i e iD i ing ools, a e gi en as inpu o he ou ing algo i hm. The ype,
loca ion and ime o he inciden a e communica ed.
4. All inpu da a a e in oduced in he SUMO simula ion en i onmen o he
espec i e a ic ne wo k ( own/ci y) unde conside a ion. Inciden loca ion
is used o de ine he a ec ed a ea. T a ic conges ion le els, po en ially
haza dous wea he condi ions (e.g. wind, looding, s o m, in as uc u e
ailu e e c.) a e used o es ima e he alue o a isk unc ion o e e y link in
he a ec ed a ea pe ehicle ype.
5. Pa hs o all ips ha a e c ossing he a ec ed a ea a e ecalcula ed by he
Rou e Guidance Tool’s algo i hms, based on he cu en ehicle loca ion and
i s des ina ion. Al e na i e pa hs a e decided so ha he o al isk le el o all
used pa hs is minimized among al e na i e op ions.
6. Al e na i e pa hs a e communica ed o he espec i e ehicles, h ough he
ela i e iD i ing communica ion componen s.
5.5.3 Signal Con ol Tool
Figu e 13 - Signal Con ol ool sequence diag am
@iD i ing Conso ium – 101147004
Page 132 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Sequence diag am s eps, p esen ed in Figu e 13:
1. Recei e inpu : Da a om a ious sou ces (e.g. came as, GPS, and loop
de ec o s) is collec ed and p ocessed by he ela i e iD i ing componen s.
2. De ec ion: Use ul in o ma ion ega ding de ec ed ehicles a in e sec ions
and/o queue leng h es ima ion a in e sec ions pe link, is p o ided o he
signal con ol algo i hm as inpu .
3. Signal Con ol Module: The module is in oduced in he SUMO simula ion
en i onmen wi h he app op ia e his o ical da a o a el demand, a ic
conges ion, a ic signal se ings and o he ne wo k- ela ed in o ma ion, o
be used o conduc ing SUMO mic oscopic simula ion es s. Acco ding o
he numbe o ehicles wai ing in he queues, a ic ligh spli s a e
compu ed and b oadcas ed o he a ic ligh s.
4. Expo : Finally, du ing he expo , a ic ligh spli s a e s o ed in app op ia e
ou pu iles.
5.5.4 Mobile Applica ion & In-Vehicle Applica ion
Figu e 14 - Mobile Applica ion & In-Vehicle Applica ion sequence diag am
Sequence diag am s eps, p esen ed in Figu e 14:
1. P epa a ion: Mobile/in- ehicle applica ion is deployed on he mobile de ice
and unning p ope ly.
@iD i ing Conso ium – 101147004
Page 133 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
2. Connec ion: Mobile/in- ehicle applica ion is connec ed o he in e ne and
o he se e .
3. Send/ ecei e: Mobile/in- ehicle applica ion sends/ ecei es messages o
di e en si ua ions.
4. Se e /Message bus: Se e /Message bus sends/ ecei es in o ma ion.
5.5.5 Ae ial Su eillance UAV Sys em
Deli e a econ igu able UAV sys em equipped wi h di e en senso s and came as
o enable p ecise, e icien da a collec ion om a eas o in e es on demand o in
eal ime. This ask aims o de elop a UAV capable o au onomous na iga ion and
eal- ime decision-making, pa icula ly in complex en i onmen s such as
main enance ope a ions and inciden moni o ing. By le e aging AI, ad anced
algo i hms, and he unique capabili ies o each UAV, he sys em will ensu e a swi ,
accu a e, and e ec i e esponse o a ious oad si ua ions.
Figu e 15 - Ae ial Su eillance UAV Sys em sequence diag am
@iD i ing Conso ium – 101147004
Page 134 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Sequence diag am s eps, p esen ed in Figu e 15:
1. P epa e UAV Au onomous Fligh : The UAV (Unmanned Ae ial Vehicle)
sys em is p epa ed o deploymen , ensu ing ha he d one is ope a ional,
and all equi ed pa ame e s a e se .
2. D one Ready o Au onomous Fligh : The UAV unde goes inal sys em
checks, including ba e y le el, GPS calib a ion, and communica ion wi h he
con ol pla o m.
3. Deploy UAV: The UAV akes o and begins pa olling he de ined a ea.
4. The UAV Cap u es he equi ed in o ma ion such as main enance ope a ions
and inciden moni o ing based on he selec ed senso s (RGB / The mal
came as).
5. Image / Video P ocessing Using he De eloped AI Algo i hms
6. The pla o m ecei es in o ma ion ega ding any main enance ope a ions
and inciden moni o ing.
5.5.6 Au onomous UAV Deploymen o A ea Co e age
This componen p o ides an AI-powe ed se ice o gene a ing op imal ligh
pa hs o a single o mul iple UAVs o coope a i ely co e la ge o sho e a eas. I
akes as inpu a geospa ial a ea (in WGS84 o ma ) and UAV pa ame e s, and
gene a es ene gy-awa e, o e lap- ee ajec o ies. The pa hs accoun o
obs acles, no- ly zones, and indi idual UAV cons ain s (e.g., ba e y, max ange).
The se ice is in eg a ed in o an end- o-end pla o m, whe e he on end
collec s mission de ini ions and he backend p ocesses hem o deli e mission-
eady ligh plans in s anda d o ma s.
Figu e 16 - Au onomous UAV Deploymen o A ea Co e age sequence diag am
@iD i ing Conso ium – 101147004
Page 135 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Sequence diag am s eps, p esen ed in Figu e 16:
1. Inpu Sou ce
The inpu o he ool is a use -de ined mission a ea, ep esen ed as a polygon in
WGS84 coo dina es, op ionally including no- ly zones and s a ic obs acles.
Addi ionally, UAV speci ica ions such as ba e y le el, ligh ime, al i ude limi s,
speed, and came a ield-o - iew (FoV) a e equi ed.
2. Da a P ocessing
The sys em pa ses and alida es he geospa ial inpu and UAV pa ame e s.
I hen compu es an op imal co e age pa h o ei he a single UAV o mul iple UAVs
based on he mission de ini ion. Obs acles and no- ly zones a e ac o ed in o he
pa h planning algo i hm o ensu e legal and sa e ligh ope a ions.
In he case o mul i-UAV ope a ions, he sys em dis ibu es sub-a eas and
compu es coo dina ed co e age pa hs o ensu e comple e and e icien mission
execu ion.
3. Ou pu Sou ce
A s anda dized .JSON ile is gene a ed ha includes:
- The compu ed waypoin s o each UAV (i mul i-UAV)
- A ea and obs acles/No-Fly zones (i any) de ini ions
- Mission pa ame e s
5.5.7 Sea Bel and Cell Phone De ec ion Tool
Figu e 17 - Sea Bel and Cell Phone De ec ion Tool sequence diag am
Sequence diag am s eps, p esen ed in Figu e 17:
1. Inpu Sou ce
- The module ecei es da a om s ee o d one came as.
- Inpu can be ei he indi idual ames o ideo s eams.
2. Da a P ocessing
@iD i ing Conso ium – 101147004
Page 136 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
- A deep lea ning model de ec s a ic iola ions, including:
o D i e s no wea ing sea bel s
o D i e s using mobile phones
o Bike s no wea ing helme s
- Each ame is p ocessed o de ec ele an objec s, wi h ou pu including
o Objec class
o Con idence sco e
o Bounding box coo dina es
3. Me ada a Gene a ion
A JSON ile is c ea ed con aining me ada a o each de ec ed objec .
The sys em also coun s he numbe o de ec ed ca s pe ame and expo s
his in o ma ion.
4. Ou pu Gene a ion
The p ocessed ames o ideo a e o wa ded o he iD i ing UI.
De ec ion esul s (objec s and me ada a) a e also sen o TEKNIKER o
u he p ocessing in he ac i i y ecogni ion ask.
5.5.8 License Pla e De ec ion Tool
Figu e 18 - License Pla e De ec ion Tool sequence diag am
Sequence diag am s eps, p esen ed in Figu e 18:
1. Inpu Sou ce
The inpu sou ce o his ool is he same as he de ec ion ool— ames om
s ee came as o d one came as.
2. Da a P ocessing
The ool p ocesses he ame simul aneously wi h he de ec ion module.
I ocuses on iden i ying ehicle license pla es and ex ac ing hei numbe s.
Once a license pla e is de ec ed, i links his in o ma ion o he
@iD i ing Conso ium – 101147004
Page 137 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
co esponding o ending ehicle (e.g., no wea ing a sea bel , using a phone,
e c.).
3. Me ada a Gene a ion
The ou pu is a JSON ile con aining he license pla e numbe s o he
o ending ehicles, along wi h ele an me ada a.
4. Ou pu
A JSON ile is gene a ed ha includes he license pla es o all iden i ied
o ending ehicles in he p ocessed ame.
5.5.9 Helme De ec ion Tool
Figu e 19 - Helme De ec ion Tool sequence diag am
Sequence diag am s eps, p esen ed in Figu e 19:
1. Inpu Sou ce
- The module ecei es da a om s ee o d one came as.
- Inpu can be ei he indi idual ames o ideo s eams.
2. Da a P ocessing
- A deep lea ning model de ec s a ic iola ion (bike s no wea ing
helme s)
- Simul aneously we de ec he license pla e o each o ende
- We combine he in o ma ion o he ehicle and he license pla e
3. Me ada a Gene a ion
- A JSON ile is c ea ed con aining me ada a o each de ec ed objec .
4. Ou pu
The p ocessed ames o ideo a e o wa ded o he iD i ing UI.
De ec ion esul s (objec s and me ada a) a e also sen o TEKNIKER o
u he p ocessing in he ac i i y ecogni ion ask.
@iD i ing Conso ium – 101147004
Page 144 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5.5.17 Po hole De ec ion and Se e i y Assessmen Tool
Figu e 27 - Po hole De ec ion and Se e i y Assessmen Tool sequence diag am
Sequence diag am s eps, p esen ed in Figu e 27:
1. Accep Inpu Sou ces
- The module ecei es inpu om a ious sou ces, including d one came as,
ixed a ic came as, and dashboa d came as in ehicles.
2. Selec and P ocess Inpu
- A e selec ing he app op ia e sou ce, he module eeds he da a in o a
deep lea ning model designed o de ec oad de ec s such as po holes
and c acks.
3. Gene a e Ou pu
- The model p ocesses he inpu images and ou pu s anno a ed images,
ma king de ec loca ions wi h bounding boxes.
- Simul aneously, he same in o ma ion is o ma ed as JSON ex , which
includes me ada a like GPS coo dina es (i p o ided by he o iginal
sou ce).
4. Assess De ec Se e i y
- The p ocessed da a is sen o he se e i y assessmen module, whe e each
de ec ’s se e i y is es ima ed.
5. Fo wa d Resul s o UI
- Finally, all esul s, along wi h se e i y assessmen s, a e sen o subsequen
modules and in eg a ed in o he p ojec 's UI, whe e hey a e p esen ed.
@iD i ing Conso ium – 101147004
Page 145 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5.5.18 Wea he P edic ion Tool (WRF and Da a Assimila ion)
Figu e 28 - Wea he P edic ion Tool sequence diag am
Sequence diag am s eps, p esen ed in Figu e 28:
1. Se Up En i onmen
Deploy WRF and WRFDA wi hin a Docke con aine o easy managemen
and po abili y.
2. Da a P epa a ion
Ga he me eo ological da a (e.g., GFS, obse a ional da a) equi ed o ini ial
condi ions and bounda y condi ions.
3. Da a Assimila ion
Use WRFDA o in eg a e eal- ime obse a ional da a in o he model,
imp o ing ini ial condi ions o be e o ecas accu acy.
4. Run Wea he P edic ion
Execu e WRF model using he p epa ed inpu da a and assimila ed ields o
wea he p edic ion.
5. Pos -P ocessing
Gene a e o ecas s and ele an me eo ological ou pu s (e.g., p ecipi a ion,
empe a u e) o analysis and decision-making.
6. Ve i ica ion
Compa e model ou pu s wi h eal- ime da a o ensu e o ecas quali y and
accu acy.
@iD i ing Conso ium – 101147004
Page 146 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5.5.19 AI-Based Real-Time Wea he Ale Sys em
Figu e 29 - AI-Based Real-Time Wea he Ale Sys em sequence diag am
Sequence diag am s eps, p esen ed in Figu e 29:
1. Fo ecas & Nowcas Re ie al
Collec o ecas da a om ex e nal sou ces and gene a e cu en (nowcas )
wea he da a.
2. Loca ion-Based Fil e ing
Ex ac and o ganize da a o p ede ined disc e e loca ions o in e es (e.g.,
ci ies, oad segmen s).
3. Haza d Iden i ica ion
Apply h eshold-based ules o de ec po en ial haza ds based on o ecas
and eal- ime da a.
4. AI-D i en Ale Gene a ion
I h esholds a e exceeded, eed he e en in o an AI model (LLM) which
c ea es a con ex -awa e haza d ale .
5. Ale Deli e y
Ale s a e exposed ia a REST API and can be accessed in eal ime by he
iD i ing pla o m o o he sys ems.
@iD i ing Conso ium – 101147004
Page 147 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
5.5.20 3D-SMART Tool
Figu e 30 - 3D-SMART Tool sequence diag am
Sequence diag am s eps, p esen ed in Figu e 30:
1. The ool p oposes he UAV’s pa h based on he a ge a ea’s ex en ,
ensu ing ha he scene will be cap u ed om all possible iewpoin s.
2. Raw da a acqui ed om he UAV is ex ac ed, p ocessed, and o ma ed
be o e being ed in o he model.
3. The model ecei es he p ocessed da a, op imizes i s pa ame e s, and
ans o ms he 2D da a in o a highly de ailed 3D en i onmen
4. Realis ic 3D ende s o gaussian spla s in ply o ma can be ex ac ed.
5.5.21 Clai eSITI Pla o m
Figu e 31 - Clai eSITI Pla o m sequence diag am
Sequence diag am s eps, p esen ed in Figu e 31:
1. Access da a om use cases
@iD i ing Conso ium – 101147004
Page 148 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
- Collec da a om each use cases (UAVs, wea he s a ions, AI came a
ou pu s) based on SCC equi emen s
2. P ocess o he compu a ion o KPI based on new da a collec ed
3. Upda e SCC o mula/pa ame e s h ough con inuous lea ning o KPI
4. s o e KPI send upda es o all pa ne s o oad manage s h ough
on end dashboa d o geojson
5.5.22 SUMO (Simula ion o U ban Mobili y)
P o ide ale s and sa e y measu es based on a digi al win buil using he SUMO
simula o . The ool is designed o simula e a ious anspo scena ios using
his o ical da a and eal- ime ield da a (UAVs, came as, senso s) o an icipa e isks
o di e en oad use s and deli e ale s o sa e y measu es o use s and ne wo k
manage s.
Figu e 32 - SUMO sequence diag am
Sequence diag am s eps, p esen ed in Figu e 32:
1. Access da a om moni o ed a eas
a. Collec da a om edge senso s (UAVs, wea he s a ions, AI came a
ou pu s) o adap a ic scena ios based on eal- ime da a.
2. Access his o ical da a o a ic calib a ion
a. Use his o ical da a o de e mine ini ial a ic demand ac oss he
ne wo k, including a eas ou side moni o ed zones.
3. Simula e a ic scena ios in SUMO
a. Pe o m a simula ion o oad use s in SUMO, in eg a ing bo h
his o ical demand and eal- ime da a. SUMO also in eg a es
@iD i ing Conso ium – 101147004
Page 149 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
communica ion ools o simula e a ious scena ios acco ding o oad
use compliances o ID i ing sys ems and communica ion delays
4. P ocess sa e y assessmen on simula ed da a
a. Ex ac sa e y ea u es om simula ed da a—such as use ajec o ies
and beha iou s— o compu e sa e y measu es o ale s.
5. P o ide geoloca ed ale s and measu es
a. Send sa e y measu es, ale s, o isk assessmen s ac oss he ne wo k
based on in as uc u e ype, o ma ed as GeoJSON.
5.5.23 CARLA (Au onomous D i ing Simula ion Pla o m)
P o ide ale s and sa e y measu es based on a digi al win buil using CARLA and
SUMO simula o s o ensu e a 3D ep esen a ion on moni o ed a ea. The ool is
designed o simula e a ious anspo scena ios using his o ical da a and eal- ime
ield da a (UAVs, came as, senso s) o an icipa e isks o mul iple oad use s (ca ,
pedes ian, cyclis ) and deli e ale s o sa e y measu es o use s and ne wo k
manage s.
Figu e 33 - CARLA sequence diag am
Sequence diag am s eps, p esen ed in Figu e 33:
1. Access da a om moni o ed a eas
a. Ge da a om on edge senso s (UAV, wea he s a ion, IA came a
ou pu s) o de ec incoming oad use on he moni o ed a ea
2. Access his o ical da a o a ic calib a ion
a. His o ical da a p o ides ini ial demands o he whole ne wo k ou side
moni o ed a eas
3. Simula e in 3D oad use beha iou s on CARLA and SUMO
@iD i ing Conso ium – 101147004
Page 150 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
a. Realise a 3D simula ion o he moni o ed a eas in use cases based on
incoming oad use s and his o ical demands. Va ious oad use
cha ac e is ics (agg essi e, isky) can be p o ided o simula ion.
4. P ocess sa e y assessmen on simula ed da a
a. Ex ac sa e y ea u es om simula ed da a – use ajec o ies,
beha iou s – o compu e sa e y measu es o ale s
5. P o ide ale s and measu es
a. In e ac wi h oads use s on si e wi h sa e y measu es by p o iding
po en ial isks
5.5.24 AI-Op imized Main enance h ough Digi al Twin
Consis s o :
• Module 1.1: Dynamic Risk Assessmen
• Module 1.2: Heal h and Logis ics Managemen
• Module 1.3: Main enance Scheduling
The sys em elies on a sha ed in as uc u e and da a pipelines o suppo decision-
making in nea eal- ime.
Global Componen Sequence O e iew
Figu e 34 - AI-Op imized Main enance h ough Digi al Twin - Global Componen sequence
diag am
Sequence diag am s eps, p esen ed in Figu e 34
Da a low and Module in e ac ions:
@iD i ing Conso ium – 101147004
Page 151 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
1. Senso Ne wo k collec s eal- ime da a like condi ion, wea he , and a ic,
and sends i o Module 1.2 (Heal h & Logis ics Managemen ).
2. Module 1.2 pe o ms da a p ocessing (ETL) and compu es heal h indices (like
RCI and HI), hen displays his in o on he Dashboa d.
3. Module 1.2 also sends p ocessed heal h and ulne abili y da a o Module 1.1
(Dynamic Risk Assessmen ), which:
• Calcula es isks using me hods like FMEA.
• Runs simula ions (e.g., “wha i " scena ios).
• Ou pu s esul o he Dashboa d (e.g., isk maps).
4. Module 1.1 passes isk and scena io da a o Module 1.3 (Main enance
Scheduling), while Module 1.2 also p o ides logis ics in o.
5. Module 1.3 combines inpu s o gene a e sho - and long- e m main enance
plans, cos es ima es, and p io i y lis s—also isualized on he Dashboa d.
Module 1.1 Risk Assessmen
Figu e 35 - AI-Op imized Main enance h ough Digi al Twin – Module 1.1 Risk Assessmen
sequence diag am
@iD i ing Conso ium – 101147004
Page 152 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
Sequence diag am s eps, p esen ed in Figu e 35:
1. Recei e da a om Module 1.2: Inpu s include he Heal h Index, a ic, oad,
and ulne abili y in o ma ion.
2. Compu e Risk using FMEA Me hodology: Module 1.1 uses Failu e Modes and
E ec s Analysis (FMEA) o assess po en ial isks and hei impac .
3. Run Scena io Simula ion – Wha -i Engine: I simula es a ious hypo he ical
condi ions o es sys em esilience and esponse unde di e en s esso s.
4. Gene a e Risk Me ics: Ou pu s include quan i iable isk alues, ankings, o
ca ego ies ha can in o m decision-making.
5. Expose REST API o Dashboa d: Module 1.1 sha es esul s ia an API, enabling
in eg a ion and eal- ime access.
6. Dashboa d displays isk maps and simula ions: The on end/dashboa d
isualizes hese insigh s o use s—e.g., isk hea maps, simula ion ou comes.
Module 1.2 Heal h & Logis ics Managemen
Figu e 36 - AI-Op imized Main enance h ough Digi al Twin – Module 1.2 Heal h & Logis ics
Managemen sequence diag am
Sequence diag am s eps, p esen ed in Figu e 36:
1. Ge Ex e nal Fo ecas s: Module 1.2 ecei es wea he and a ic o ecas s om
ex e nal APIs.
2. S ep 2: S eam Senso Da a: I simul aneously inges s eal- ime condi ion
da a om on- ield senso s (e.g., pa emen RCI).
@iD i ing Conso ium – 101147004
Page 153 o 157
In elligen & Digi al Roadway In as uc u e o Vehicles
In eg a ed wi h Nex -Gen Technologies
3. ETL P ocessing: Ex ac ion o aw da a, T ans o ma ion (cleaning,
no maliza ion), Loading and in eg a ion in o i s sys em
4. P edic i e Modelling: I uns machine lea ning models o compu e a Heal h
Index o each oad segmen .
5. Send o Dashboa d: The inal ou pu — oad condi ion o ecas s and Heal h
Index—is pushed o he dashboa d o end-use isualiza ion.
Module 1.3 Main enance Scheduling
Figu e 37 - AI-Op imized Main enance h ough Digi al Twin - Module 1.3 Main enance
Scheduling sequence diag am
Sequence diag am s eps, p esen ed in Figu e 37:
1. S ep 1: Inpu om Module 1.1
Recei es isk le els and scena io ou pu s ha indica e po en ial ailu e zones
and isk-c i ical a eas.
2. S ep 2: Inpu om Module 1.2
Ge s logis ics and condi ion me ics, including oad deg ada ion,
accessibili y, and a ailable esou ces.
3. S ep 3: Op imiza ion Engine
