Copy igh © 2025 1
Eu opean Con e ence on EDGE AI
Technologies and Applica ions - EEAI
20-22 Oc obe 2025, Naples, I aly
The in e sec ion o imagina ion and
execu ion, whe e edge AI lea ns o
c ea e, eason, and ac .
Copy igh © 2025
Eu opean Con e ence on EDGE AI
Technologies and Applica ions - EEAI
Milan, I a20-22 Oc obe 2025 Naples, I aly
2
Copy igh © 2025
O idiu Ve mesan, SINTEF AS, No way
End- o-End Wo k low Au oma ion o Robus Edge AI Sys ems
Copy igh © 2025
P esen a ion Ou line
•Edge AI Landscape E olu ion
•Eme gence o Agen ic AI
•End- o-End Wo k low Au oma ion in Edge AI
•Agen ic Fea u es in Edge AI
•Co e Agen ic Capabili ies - Au onomous Edge AI Sys ems
•F amewo ks - Da a Inges ion, Model Deploymen , O ches a ion
•Go e nance, Compliance, and Risk Managemen
•Technical and Ope a ional Challenges
•Fu u e Di ec ions and Inno a ion Oppo uni ies
•Conclusions
3
Copy igh © 2025
Edge AI Landscape E olu ion
4
Eme gence o agen ic AI oSys ems ha gene a e insigh s and au onomously execu e mul isys em asks
and o ches a e en i e wo k lows wi h minimal human in e en ion.
He e ogeneous edge de ices oSys ems equi ing in e ope abili y, scalabili y, minimising la ency, enhancing
secu i y, and ensu ing eal- ime esponsi eness.
E2E wo k low au oma ion
oSys ems equi ing end- o-end wo k low au oma ion amewo ks ha
in eg a e da a inges ion, model aining, eal- ime in e encing, and con inuous
lea ning ac oss dis ibu ed de ices.
❖Technical dimensions (e.g., da a pipelines, model deploymen on de ices wi h a ying compu e capabili ies,
he e ogeneous sys em o ches a ion, and con inuous moni o ing)
❖Eme gen agen ic ea u es (e.g., in en unde s anding, con ex -awa e lea ning, au onomous easoning, and adap i e
beha iou )
❖Suppo o he new wa e o au onomous sys ems.
Copy igh © 2025
Eme gence o Agen ic AI
5
Agen ic AI ma ks a pa adigm shi om pu ely gene a i e o p edic i e
sys ems o hose ha inco po a e au onomous easoning.
De ined by hei abili y o eason, plan, and ac independen ly in
complex, dynamic en i onmen s, hese sys ems can p ocess inpu s om
mul iple sou ces, de e mine app op ia e ac ions, and econ igu e
wo k lows in si u.
Agen ic sys ems o igina e om a usion o p obabilis ic easoning and
de e minis ic execu ion, a syn hesis ha enables hem o moni o eal-
ime condi ions and ake p oac i e measu es wi hou cons an human
supe ision.
•The expansion o IoT de ices and he need o low-la ency decision-making has
os e ed he de elopmen o edge AI.
•By p ocessing da a locally on a mul i ude o he e ogeneous de ices, edge AI
sys ems can deli e nea -ins an aneous esponses, educe ne wo k bandwid h
usage, and p ese e da a p i acy.
•The in eg a ion o hese wo pa adigms, agen ic beha iou and edge deploymen ,
equi e he de elopmen o new wo k low au oma ion amewo ks o
o ches a ing asks ac oss dispe sed loca ed nodes.
Copy igh © 2025
End- o-End Wo k low Au oma ion in Edge AI
6
Deploying AI models on edge
de ices, which ha e limi ed
compu e powe necessi a es
echniques such as model
quan iza ion, p uning, and
knowledge dis illa ion o
op imise models o educed
la ency and minimal esou ce
usage. As edge AI in ol es he
eal- ime execu ion o agen ic
unc ions, ensu ing ha hese
comp essed models can s ill
pe o m asks au onomously
and accu a ely is c i ical.
He e ogeneous edge en i onmen s
ha e unique challenges in da a
inges ion due o a iable connec i i y,
di e se da a o ma s, and esou ce-
cons ained de ices. New
amewo ks employ a combina ion
o senso usion, eal- ime da a
p e-p ocessing, and dis ibu ed
s o age s a egies o ensu e ha
da a is eliably acqui ed and
s anda dised. Da a no maliza ion
p ocesses, including il e ing, noise
educ ion, and ea u e ex ac ion,
applied be o e in e ence o lea ning
algo i hm is key o p o iding edge AI
E2E wo k low au oma ion.
Da a Inges ion and P ep ocessing a he Edge Da a Model Deploymen on He e ogeneous De ices
Copy igh © 2025
End- o-End Wo k low Au oma ion in Edge AI
7
Moni o ing amewo ks
ou pu p ecise me ics
ela ed o sys em
pe o mance, e o a es,
and p ocessing la encies.
Cen al o end- o-end
wo k low au oma ion is he
o ches a ion laye , a
managemen sys em ha
coo dina es ac oss
nume ous agen s and
de ices.
O ches a ion and Wo k low Au oma ion Con inuous Moni o ing and Feedback Loops
The laye enables:
•Dynamic Task Scheduling: Au oma ically dis ibu ing
asks among a ailable esou ces.
•In e -Agen Communica ion: Ensu ing agen s sha e
con ex , da a, and ins uc ions seamlessly.
•In eg a ed Sys em Moni o ing: Con inuously acking
ask p og ess, sys em heal h, and deploymen e icacy.
•Au oma ed eedback loops allow he sys ems o i e a e
in nea eal- ime, upda ing ope a ional pa ame e s
based on pe o mance da a.
•The eedback mechanisms acili a e sel -co ec ion,
enabling he edge sys em o adap i s wo k lows o call
o human in e en ion when anomalies a e de ec ed.
Copy igh © 2025
Agen ic Fea u es in Edge AI
8
•Au onomous Reasoning and
Decision-Making.
•In en Unde s anding and P oac i e
Ac ion.
•Con ex -Awa e Lea ning and
Adap i e Beha iou .
•Syne gy Be ween Agen ic Beha iou
and Edge AI.
Copy igh © 2025
Co e Agen ic Capabili ies - Au onomous Edge AI Sys ems
9
Fea u e Desc ip ion
Au onomous
Reasoning
Independen
mul i-
s ep
planning
and execu ion.
In en
Unde s anding
In e p e a ion
o human
and
machine
commands
and
p oac i e
adjus men s.
Con ex
-Awa e Lea ning
Adap i e
modi ica ions
based
on his o ical and eal
-
ime
da a.
Copy igh © 2025
O idiu.Ve mesan@sin e .no
Fo you a en ion
Copy igh © 2025
Thank You
Copy igh © 2025
E en O ganise s
17
SMARTY - Scalable and Quan um Resilien He e ogeneous Edge Compu ing enabling
T us wo hy, ocuses on cloud-edge con inuum o he e ogeneous sys ems, ha p o ec s
da a-in- ansi and da a-in-p ocess, employing no el accele a o s o quan um esilien
communica ions, con iden ial compu ing, and so wa e de ined pe ime e s.
h ps://www.sma y-p ojec .eu/
EdgeAI (Edge AI Technologies o Op imised Pe o mance Embedded P ocessing)
de elops new elec onic componen s and sys ems, p ocessing a chi ec u es, connec i i y,
so wa e, algo i hms, and middlewa e h ough he combina ion o mic oelec onics,
edge AI, embedded sys ems, and edge compu ing. www.edge-ai- ech.eu
EdgeAI
NEUROKIT2E
SMARTY
dAIEDGE
dAIEDGE is he Eu opean Ne wo k o Excellence o dis ibu ed, us wo hy, e icien ,
and scalable AI a he Edge and p omo es he applica ion, de elopmen , and
deploymen o A i icial In elligence (AI) on edge compu ing pla o ms.
h ps://daiedge.eu/
NEUROKIT2E (Open sou ce deep lea ning pla o m dedica ed o Embedded ha dwa e
and Eu ope) p oposes a Deep Lea ning Pla o m o Embedded Ha dwa e a ound an
es ablished Eu opean alue chain (AI HW/SW). The solu ions de eloped suppo neu al
ne wo k design, op imisa ion, and implemen a ion on cons ained HW.
h ps://www.neu oki 2e.eu/
NEUROKIT2E
Copy igh © 2025
E en O ganise s
18
REBECCA (Recon igu able He e ogeneous Highly Pa allel P ocessing Pla o m o sa e
and secu e AI) aims o democ a ize he de elopmen o edge AI sys ems and c ea e a
comple e ha dwa e and so wa e s ack cen e ed a ound a RISC-V CPU, which o e
highe pe o mance, ene gy e iciency, sa e y, and secu i y han exis ing sys ems.
www. ebecca-chip.eu
TRISTAN aims o expand, ma u e, and indus ialize he Eu opean RISC-V ecosys em o
compe e wi h comme cial op ions by le e aging he Open-Sou ce communi y o gain
p oduc i i y and quali y. A Eu opean s a egy o RISC-V designs will be de ined,
c ea ing a eposi o y o indus ial a e building blocks o SoC designs in a ious
applica ion domains. www. is an-p ojec .eu
CLEVER (Collabo a i e edge-cLoud con inuum and Embedded AI o a Visiona y indus y
o hE u uRe) p oposes inno a ions in ha dwa e accele a o s, design s ack, and
middlewa e so wa e ha e olu ionize he abili y o edge compu ing pla o ms o
ope a e ede a ed, le e aging spa se esou ces ha a e coo dina ed o c ea e a
powe ul swa m o esou ces. www.cle e p ojec .eu
SMARTEDGE
TRISTAN
CLEVER
REBECCA
The Sma Edge p ojec aims o achie e dynamic in eg a ion o decen alized edge
in elligence while p io i izing eliabili y, secu i y, p i acy, and scalabili y. The Sma Edge
solu ion includes a low-code ool p og amming en i onmen wi h h ee main ools:
Con inuous Seman ic In eg a ion, Dynamic Swa m Ne wo k, and Low-code Toolchain o
Edge In elligence. h ps://www.sma -edge.eu/
Copy igh © 2025
E en O ganise s
19
The objec i es o LoLiPoP IoT (Long Li e Powe Pla o ms o In e ne o Things) a e o
de elop ene gy ha es ing-based inno a i e Long Li e Powe Pla o ms ha enable
e o i o wi eless senso ne wo k edge de ices o asse acking, condi ion and
pe o mance moni o ing. www.lolipop-io .eu
LoLiPoP IoT
EdgeAI-T us
AIMS5.0
SC4EU
SC4EU is a unique Chips JU "Inno a ion Ac ion" p ojec o ake he supply chain
managemen o semiconduc o p oduc ion in Eu ope o a new le el. A ue demand
pla o m, along wi h i s on ology as a o mal desc ip ion o all in o ma ion wi hin he
chain, acili a es close in e ac ion and smoo h, anspa en collabo a ion, making e en
highly complex supply chains esilien , lexible, and agile. h ps://sc4.eu/
EdgeAI-T us add esses an ad anced, us wo hy edge AI ecosys em h ough cu ing-
edge ha dwa e, so wa e, and ools. The p ojec aims o enhance decen alized EdgeAI
ope a ions ha a e secu e, eliable, and sus ainable. By in eg a ing AI-based
algo i hms, de ices, and APIs, EdgeAI-T us os e s in e ope abili y and secu e da a
exchange ac oss di e se pla o ms. om senso -ac ua ed de ices o cloud sys ems, all
wi hin a dynamic ze o us en i onmen . h ps://www.edgeai- us .eu/
AIMS5.0 aims o boos he economy by adop ing, ex ending, and implemen ing AI-
enabled HW and SW componen s and sys ems ac oss he en i e indus ial alue chain.
New echnologies om IoT and based on Seman ic Web on ologies, ML and AI help
Eu opean manu ac u e s o shi om Indus y 4.0 o Indus y 5.0, c ea ing human-
cen ic wo kplace condi ions and a clima e- iendly p oduc ion. h ps://aims50.eu/
Copy igh © 2025
Suppo ing O ganiza ions
20
The Eu opean Technology Pla o m on Sma Sys ems In eg a ion is an indus y-
d i en policy ini ia i e, de ining esea ch, de elopmen and inno a ion needs as
well as policy equi emen s ela ed o Sma Sys ems In eg a ion and in eg a ed
Mic o- and Nanosys ems. The main objec i e is o de elop a ision and o se up a
S a egic Resea ch Agenda. www.sma -sys ems-in eg a ion.o g
Inside Indus y Associa ion is he Eu opean Technology Pla o m o esea ch,
design and inno a ion on In elligen Digi al Sys ems and hei applica ions. The
Associa ion is a membe ship o ganisa ion o he Eu opean esea ch and
inno a ion ac o s wi h mo e han 200 membe s and associa es om all o e
Eu ope. www.inside-associa ion.eu
Chips Join Unde aking suppo s esea ch, de elopmen , inno a ion, and u u e
manu ac u ing capaci ies in he Eu opean semiconduc o ecosys em. Launched as
pa o he Chips o Eu ope Ini ia i e, i con on s semiconduc o sho ages and
s eng hens Eu ope's digi al au onomy, engaging a signi ican EU,
na ional/ egional and p i a e indus y unding o nea ly €11 billion.
h ps://po al.chips-ju.eu opa.eu/
EU
AENEAS
EPoSS
INSIDE
Chips JU
AENEAS s anding o Associa ion o Eu opean NanoElec onics Ac i i ieS, is an
indus ial Associa ion, es ablished in 2006, p o iding unpa alleled ne wo king
oppo uni ies, policy in luence & suppo ed access o unding o all ypes RD&I
pa icipan s in he ield o mic o and nanoelec onics enabled componen s and
sys ems. h ps://aeneas-o ice.o g/
