Suitability assessment method for safe robot tooling designin Human RobotCollaborative applications

ScienceDirect Available online at www.sciencedirect.com Procedia CIRP 128 (2024) 770–775 2212-8271 © 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the 34th CIRP Design Conference 10.1016/j.procir.2024.07.063 Keywords: Human –Robot Collaboration, Workspace design; Safety; 1. Introduction Flexibility and adaptability in contemporary production systems is of high importance due to the need to cope with variable productivity in terms of size and variability [1]. In this context, along with the ideas related to Industry 4.0 and 5.0 paradigms, HRC which is studied thoroughly through the last decade seems to be capable of supporting flexible manufacturing by combining the positive aspects of both words, the strength, accuracy and repeatability of robots along with the dexterity, cognition and adaptability of humans [2][3]. Collaborative robots (cobots) are increasingly adopted in the market, supporting the production of small and mediumsized enterprises (SMEs) that benefit from their costeffectiveness, ease of use, and deployment capabilities; apart from SMEs, also other large manufacturing industries are also utilizing cobots with increasingly higher rates [4]. 1.1. Human –Robot collaboration classification The broad spectrum of the HRC concept has led to the distinction of four categories of human-robot relationships based on their interaction levels. Coexistence involves humans and robots sharing a physical space but working independently without overlapping workspaces or direct contact. Interaction is characterized by communication or control dynamics within a shared workspace, where tasks are completed sequentially with potential physical contact. Cooperation scenario includes autonomous agents, human and robot, collaborating for mutual benefit, sharing resources in partially overlapping workspaces, often without direct contact. Lastly, collaboration involves the highest relationship Suitability assessment method for safe robot tooling design in HumanRobot Collaborative applications Konstantinos Katsampiris-Salgadoa, Nikos Dimitropoulosa, George Michalosa, Sotiris Makrisa* aLaboratory for Manufacturing Systems and Automation, Department of Mechanical Engineering and Aeronautics, University of Patras, Patras, 26504, Greece * Corresponding author. Tel.: +30-2610-910160; fax: +30-2610-997314.E-mail address:[email protected] Abstract State of the art applications in human-robot collaborative work cells, are, sometimes, highly compromised by strict directives to avoid potential hazards for operators. Robot motion (e.g., speed, momentum, power etc.) is strictly defined according to ISO/TS15066 to guarantee that potential pressures and forces due to collisions do not exceed predefined limits for different parts of the human body. In this direction the authors suggest a robotic tool design methodology that could optimize Human –Robot Collaborative (HRC) applications by preventing vast reduction of robot velocity during the design phase, while eliminating the need for redesign during the validation stages of the cell. Also, actuation mechanisms and their components are also presented to uphold human safety standards. To demonstrate the effectiveness of the proposed method, a real-life industrial use case has been materialized, involving ahigh-payload collaborative robot performing assembly operations. The results underscore the effectiveness of these design practices in enhancing the efficiency and safety of human-robot collaboration. 34th CIRP Design Conference © 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the 34th CIRP Design Conference Konstantinos Katsampiris-Salgado et al. / Procedia CIRP 128 (2024) 770–775 771 level of humans and robots, where they work jointly in a shared space, in coordinated, synchronous activities to achieve common goals, with physical contact being a possibility [2]. 1.2. Human –Robot collaboration high level challenges Deploying HRC systems presents several challenges that need to be addressed to ensure both efficiency and safety. A major concern is guaranteeing safety between humans and robots, particularly in shared workspaces [5]. This requires sophisticated safe sensing and safe control systems that can ensure that the appropriate safety functions are performed without posing anyrisk for human health. Another challenge lies in the ergonomic integration of robots into human-centric environments, where robots adapt to human needs without causing any ergonomic hazards neither in short nor long term. Additionally, it is important to establish robust means of intuitive interaction between humans and robots not only to prevent potential threats but also to enhance the symbiotic experience of humans [7].A few more challenges, in terms of productivity, is that for safety precautions cobots run in minimized speed while also have a high initial investment cost. Lastly, there are regulatory and standardization issues that present many challenges in HRC systems [5], as the rapid evolution of HRC technologies and the need for their deployment in more and various applications often outpaces the development of appropriate safety guidelines and sensorial safety systems [8]. 1.3. Contribution of work and document structure This paper addresses the identified problems in the literature for comprehensive and practical safety assessment tools in HRC systems, by proposing a novel suitability assessment method for safe robot tooling design that bridges the gap between advanced safety methodologies and their application in industry [16]. For the methodology of the assessment tool, the review of HRC classification (Section 1.1) and challenges (Section 1.2) preceded. Αn in depth study of the relevant standards was performed subsequently, to conclude with the methodology requirements both from regulatory and application aspects that are outlinedin Section 2. The approach towards the method development culminates in with the definition of the assessment topics and scores (Section 3), leading to a more effective design of robotic tools, preventing design flaws that would fail safety validation tests during the deployment of the HRC system, as it is demonstrated in Section 4. 2. Design requirements for HRC and regulatory framework 2.1. General Requirements Design requirements for HRC are based on numerous aspects, with a major direction towards human safety and compliance with legal and regulatory standards, due to the inherent danger of humans working together with robotic systems. However, safety is not the only aspect that should be taken into allowance during the deployment of the HRC systems. The design of robots should also consider human ergonomics to reduce physical strain and increase comfort during collaboration, especially in cases where humans and robots share the same task and workpiece (e.g., case where the robot acts as workpiece holding device), or in cases human operators directly physically interact and control robots (e.g., impedance/force control) [8]. 2.2. Requirements from standards Safety requirements in Human-Robot Collaboration (HRC) design, as outlined in ISO 10218-1, ISO 10218-2, and ISO/TS 15066, are crucial for ensuring a secure interaction between humans and robots in a shared workspace. The primary goal is the risks mitigation that are associated with robot operation and human interaction in a systematic sense. The process begins with a detailed risk assessment, identifying potential hazards related to the robot's functionality, system design, and specific application use. Based on this, the design should include strategies for intrinsically safe design (e.g., not sharp tooling’s), safeguarding measures (e.g., sensor monitoring and physical barriers), and additional protective measures, when necessary, e.g., setting limits on robot dynamics (speed and force) thus preventing injury severity and also incorporating ergonomic design to minimize human strain [10], [11], [12]. In workspace design, safety is addressed by configuring the environment to prevent risks such as trapping or crushing hazards. This involves careful placement of machinery and ensuring the collaborative space is designed to reduce the possibility of human-robot collisions. Techniques like safetyrated soft axis and space limiting are employed to minimize the operational space of the robot, reducing the risk of accidental contact [10], [11], [12]. For the operation of collaborative robots, safety measures include emergency stop functions, controlled stopping, and hand-guiding mechanisms. These features allow for a quick response in case of a hazard and facilitate safe interaction when humans and robots are in close proximity. Additionally, robots should be designed to seamlessly switch between collaborative and non-collaborative modes, ensuring safety across different operational scenarios. [10], [11], [12]. Along with the safety measures, training operators is also among the safety practices to ensure awareness of the risks and precautions when working with robots In Table 1the main standards that were utilized for the analysis included in this work are summarized. Table 1. Main directives for HRC applications. ISO Name ISO 10218-1 Robots and robotic devices -Safety requirements for industrial robots -Part 1: Robots ISO 10218-2 Robots and robotic devices -Safety requirements for industrial robots -Part 2: Robot systems and integration ISO/TS 15066 Robots and robotic devices -Collaborative robots ISO 13849-1 Safety of machinery -Safety-related parts of control systems - Part 1: General principles for design ISO 13849-2 Safety of machinery -Safety-related parts of control 772 Konstantinos Katsampiris-Salgado et al. / Procedia CIRP 128 (2024) 770–775 systems - Part 2: Validation ISO 13855 Safety of machinery -Positioning of safeguards with respect to the approach speeds of parts of the human body ISO 13857 Safety of machinery -Safety distances to prevent hazard zones being reached by upper and lower limbs 2.3. Power and Force Limiting HRC Power and force limiting (PFL) in human-robot collaboration involves a method of operation where physical contact between the robot system (including the workpiece) and an operator may occur [12]. This type of operation requires robot systems that are specifically designed for this type of HRC, aiming to limit to a safe level the contact forces and pressure between human operators and robots during a contact event. Robot movement is closely monitored with high precision, and any small deviation from the planned position can be quickly identified. The robot uses precise and robust encoders with high resolution to accurately track its speed and position. To measure forces and torques applied in a collision event, numerous approaches exist; robots may utilize the current values for each joint, or tactile sensors may cover the tools and the robot itself for increase protection [6]. The benefits of this type of collaboration are many, as for example reduced need for safety barriers, meaning that more room on the shop floor is available. What is more, the lack of safety monitoring systems such as laser scanners and safety camera systems, allow also for better exploitation of the shop floor. Protective stops can be reduced due to violation of the safety zones (e.g., from laser scanners) leading to lower production times. The solutions that incorporate PFL, may also allow increased levels of flexibility and reconfigurability, as the integrated safety system, does not require sophisticated layouts for additional monitoring devices to be installed, adapting to a variety of work environments and tasks [8]. 3. Robotic tool design for HRC Approaching the appropriate design method for robotic tools that are aimed to be applied for HRC applications is a complex and multidisciplinary effort. The design assessment method that the authors propose is based on a scoring system for different aspects of the robotic tool system. The aspects that have been chosen are an outcome of the design requirements for HRC that can be found by the standards from Table 1. 3.1. HRC Tool Suitability Assessment (HRC-TSA) HRC Tool Suitability Assessment (HRC-TSA) scoring system is an approach for validating the suitability of a designed tool protype prior to materializing it. HRC-TSA is based on three main pillars of assessment inspired by the standards special from ISO/TS 15066. The three pillars are: (i) Functional Safety, (ii) Collaborative Safety and (iii) Ergonomic Safety and will be further analyzed in sections 3.1.1. 3.1.2. and 3.1.3.The overall HRC-TSA is presented as a whole in Fig.1. For each pillar and sub-category, a number of points are assigned. If a sub-category is fully satisfied, then the maximum points are scored. If the measures are not totally satisfied, then depending on whether they are partially or not a low or middle score is given. To prevent arbitrages, the scoring should be given by an expert designer, as inexperience engineers will not be capable of performing an objective assessment. Fig. 1. HRC –TSA method summary 3.1.1. Functional Safety 3.1.1.1. System Performance Level and Category Performance Level (PL) refers to the reliability required for a safety function, indicating its effectiveness in reducing or eliminating harm. The ISO 13849-1 standard outlines five levels: PL a, PL b, PL c, PL d, and PL e, with PL e indicating the highest level of safety is required. Factors such as component lifespan, diagnostic coverage, common cause failure, system structure, fault behavior, safety software, systematic failure, and environmental conditions are vital in estimating the PL. Just 3 parameters need to be examined to assess the performance level (PL), as indicated in Table 2.: Table 2Parameters to determinate PLr in accordance with EN ISO 13849-1 Severity of injury S Slight (normally reversible injury) S1 Serious (normally irreversible injury including death) S2 Frequency and/or exposure to a hazard F Seldom to less often and/or exposure time is short F1 Frequent to continuous and/or exposure time is long F2 Possibility of avoiding the hazard P Possible under specific conditions P1 Scarcely possible P2 The required performance level PLr is calculated utilizing the graph in Fig.2. Konstantinos Katsampiris-Salgado et al. / Procedia CIRP 128 (2024) 770–775 773 Fig. 2. Risk graph in accordance with EN ISO 13849-1 Category (Cat.), in ISO 13849-1, relates to the architectural structure and fault tolerance of safety-related control system parts. It shows a system's resilience to failures while maintaining safety functions. An overview of the categories based on ISO 13849-1: •Cat. B: Basic circuits without diagnostics. •Cat. 1: Similar to Cat. B, with minimal fault detection but no guarantee against hazards. •Cat. 2: Incorporates fault detection, signaling faults but may not prevent hazards. •Cat. 3: Features redundant channels for improved fault tolerance, though not all faults are detected. •Cat. 4: Has redundant channels for continuous monitoring and halts operation upon detecting a single fault. The architectural category is based on the specific PLr and the readers can refer to ISO 13849-1:2023 for more information on the calculations and their correlation. It is important for a robotic tool to comply with the safety requirements related to Performance Level, which should be PL“d” and the structural system Category that should be “3” for the HRC application, as this is instructed by ISO/TS 15066. If during the robotic tool design the components used (e.g., pneumatic safety valves, drive systems, switches etc.) do comply with the PLd and Category 3 requirements, the tool scores 20 points, otherwise automatically the tool cannot pass the implementation stage. 3.1.1.2. Fault detection/response sensing systems Based on 3.1.1.1, the subcomponents of the robotic tool, due to their high-PLr required and architectural category, will be fail-safe, ensuring that safety functions are maintained even during a failure. To guarantee optimal performance of the robotic tool and prevent uncontrolled incidents, points in the scoring system are allocated for integrating monitoring devices. For instance, a well-designed tool with a pneumatic circuit should include pressure sensors within the pipes, providing feedback to a safety PLC that oversees all safety functions of the robot, including light and sound awareness indicators. Other similar enhancements include feedback signals from electrovalves, proximity/optical sensors in the gripper's fingers for alignment and positioning accuracy, pneumatic cylinder sensors for position feedback to control systems to avoid motion during incorrect mechanism actuation, and safety switches ensuring all moving parts of the robotic tool are correctly positioned before any motion execution.The maximum points assigned for this sub-category is 10. 3.1.1.3. Emergency Stop functionality and enabling selector. The collaborative gripper tool should include at least one emergency stop button, allowing the operator to halt the robot instantly in an emergency. This button must be easily accessible and visible. For larger robotic tools, multiple buttons might be necessary. The emergency stop button(s) should be easily seen, and it would be beneficial to be accompanied also with a reset button, to clear errorsand allow the robot to continue with its next task. For applications involving a hand-guiding system, the handle used for Force/Torque (F/T) control should feature a three-stage enabling selector (deadman switch). This switch is off when not pressed or lightly pressed, on under medium pressure, and off again if pressed too hard, sending a feedback signal to the PLC. An emergency stop button should also be conveniently located near, if not on, the enabling selector for quick access during emergency events. The maximum points assigned for this sub-category are 10. 3.1.2. Collaborative safety 3.1.2.1. Passive design to eliminate high pressure. It is of high importance during the design phase, to integrate safety-focused design features that minimize injury risk upon contact between the robotic tool and the human operator. Thus, the cobot’s tool design should include comprehensive safety features as for example fully rounded edges, extensive padding to cover hard/rigid surfaces, and compliant materials over all contact surfaces. The goal for this design approach, is to limit the excreted forces and especially pressure values that come up during a collision or contact event among humans and the moving robot. A compliant tool, without sharp edges, along with the incorporation of a robust collision detection system, during the validation phase led to higher speeds without exceeding the biomechanical limits set in ISO/TS15066. The maximum points assigned for this sub-category are 20. 3.1.2.2. Sensor system for collision detection. In case the robot cannot provide collision detection on to the tool, as the collision detection system of the arm is done by covering the robot with safety sensorized skins (e.g., COMAU AURA), the proper passive design is not adequate. Two approaches that might be followed, is the covering of the robotic tool with protective skins, (e.g., AIRSKINS by BLUE DANUBE ROBOTICS), or to add safety rated according to 3.1.1.1. F/T sensors that can detect safely collisions [13] and share the collision detection results with the safety control unit which is responsible for the execution of the appropriate safety functions. It is also important to mention that even in cases where the robotic system might contain a robust collision detection system, the gripper’s geometry might include restrictions that could not make the gripper valid for PFL. In this scenario, that geometry, is advised to be also covered by sensorized skins to mitigate risk and allow higher robot speeds as per the safety validation testing. The maximum points assigned for this subcategory are 20. 774 Konstantinos Katsampiris-Salgado et al. / Procedia CIRP 128 (2024) 770–775 3.1.2.3. Operator notification. An effective gripper design should include devices that inform the operator about the robot's operational status. For instance, a light indicator on the tool could signal its different states; whether the robot is in PFL collaborative mode, if it is operating at full speed with safety dynamics disabled, or in an emergency state. Additionally, sound awareness devices could be beneficial, especially in scenarios like pressure loss. They could alert the operator to maintain a safe distance from the robot, thereby preventing injuries that might occur if parts fall due to a failure of the pneumatic circuit. The maximum points assigned for this sub-category are 10. 3.1.3. Ergonomic safety The design of the robotic tool should also include considerations related to ergonomics. As mentioned in 3.1.1.3. the switches for the emergency stop functions and the hand guiding devices should be easily accessible and do not cause any ergonomic concerns. The design and position of any manually controlled robot guiding device should be such that will be convenient for extended use, with clarity in control preventing loss of attention, improper operation, or possible confusion. During collaborative operations the robotic tool should prevent any possible stress, fatigue, or lack of concentration (e.g., unclear light indications, too much noise from sound awareness devices etc.). For the ergonomics related considerations, 10 points is the highest score. 3.1.4. Final assessment After the initial design is completed, a senior engineer takes on the role of assessing the tool's design, acting as a referee. Using the summarized chart in Fig.1., the referee evaluates each category and sub-category. The design is rated based on its compliance with each sub-category. Low ratings indicate zero or minimal compliance with the practices described in this paper. Medium scores are given when requirements are partially or largely met, but some aspects still need addressing to ensure successful validation tests by the safety expert. The highest rating in each category is scored only if the referee finds no issues and all sub-category requirements are fully satisfied. Given the strict nature of HRC safety precautions, a similar approach should be adopted in HRC-TSA. More rigorous checks lead to the application of better design practices, preventing unnecessary costs and time wastage before the actual materialization of the tool. Once scores are assigned to each subcategory, they are scaled to a 0-10 range. Based on Fig.3., five potential outcomes are identified. Scores of 1-2 suggest a radical design change is necessary, along with a thorough review of standards and design guidelines. Scores of 3-4 indicate a major redesign is required, as the design significantly deviates from desired requirements, and a re-evaluation of inherent process hazards is advised. Scores of 5-6 reveal several safety gaps; however, a robust redesign could achieve the desired outcome. For scores of 7-8, minor modifications are needed, and the tool design is close to realization, provided certain points are addressed. Lastly, scores of 9-10 imply the robotic tool is highly suitable for HRC applications, meeting all key requirements effectively. Fig. 3. Acceptability scoring from HRC-TSA method. 4. Renewable energy sector use case –High payload robot application. 4.1.1. Scenario description The proposed approach was evaluated through a practical example from the renewable energy sector. The first commercial High Payload Collaborative Robot (HPCR), the COMAU AURA -170kg payload, was employed to pick and manipulate a solar thermal collector, weighing ~40kg and measuring ~2200x1200mm, depending to the variant. The robot presents the collector to the operator in a convenient position, to install small components (like rivets, caps, stickers, etc.) necessary for completing the assembly. This robot is equipped with a protective safety skin that includes capacitive and proximity sensors to detect collisions with its surroundings. For instance, if a person accidentally touches theprotective skin while the robot is moving, it triggers a safety function causing the robot to immediately stop. However, it doesn't have a safety-rated collision detection system that evaluates motor torque values, meaning collisions involving the tool are not detected. To enable PFL for HRC, it's necessary to integrate a collision detection system with the gripper. This can be done through additional protective skins or by evaluating Force/Torque on the flange. 4.1.2. HRC-TSA application. The design of the gripper, based on the HRC-TSA, is presented below. The gripper's end effector is made from a frame of aluminum profiles. Suction caps, equipped with level compensators for picking up the solar collector, are attached to this frame. The vacuum system incorporates multiple pneumatic ejectors, each controlling a pair of a total of eight suction caps. This ensures that the parts will not drop in the event of a suction cap or ejector failure. The ejectors can maintain vacuum during a power failure or if the air supply is interrupted. Also, the ejectors send feedback signals to the safety PLC, alerting it if the vacuum is lost. If pressure drops in any ejector, the PLC activates a siren to warn the operator. To avoid interference with other mechanical parts and ensure the reliability of the pneumatic system, the pneumatic pipes are concealed within the aluminum profile's slots. All sharp edges on the gripper are covered with soft padding for safety. This padding also covers the level compensators on the suction cups. Furthermore, the gripper includes a handguiding device with a dead man switch and an emergency stop button. A F/T sensor, placed between the gripper and the robot's flange, along with motor current data evaluation, forms a collision detection system. This adheres to the dual- Konstantinos Katsampiris-Salgado et al. / Procedia CIRP 128 (2024) 770–775 775 channel approach of architectural Cat3, covering sensor input, processing, and control output. It executes safety functions in the event of a collision. To address trapping hazards, the system combines PFL with Speed and Separation Monitoring (SSM) for HRC. To mitigate collision risks with the solar collector, safety measures like speed reductions are implemented, ensuring the system passes safety validation. A collision detection event is shown in Fig.4.: Fig. 4. During a process cycle, if the operator contacts the robot, the indicator turns from green (a) to red (b) for an emergency stop. 5. Discussion The proposed method for evaluating the design of robotic tools shows significant promise, particularly in terms of cost savings. It has the potential to reduce expenses related to the manufacturing of parts and the acquisition of components that fail safety validation tests. Beyond cost savings, this method could also save considerable time, as the production and ordering of components for robotic tools often lead to delays. Fig. 5Cost and influence level during design (left) [14]. Cost vs design changes graph (right) [14]. As evidenced in various studies and examples in the literature [14][14] it is recognized that making corrections in the early stages of the design process is a best practice. The reason for this is that the potential for influencing product design is greater in the early stages, and the cost associated with changes is lower. This concept is illustrated in Fig.5. 6. Conclusions and future work This paper presents a design approach for robotic tools intended for PFL HRC. This methodology aims to facilitate the safe and effective integration of these tools in collaborative environments. It offers the potential for substantial cost savings by ensuring tool designs are finalized before production, eliminating unnecessary manufacturing. Additionally, it can reduce the time needed for ordering and assembling tools that fully meet HRC standards. For future work, automating this methodology through a software tool would be advantageous. Basic descriptionsof application requirements, tool characteristics and the CAD product design could act as inputs for the software. By incorporating a Large Language Model for better description, the software could check for potential flaws and recommend modifications to make the tools fully compliant for HRC. Acknowledgements This study was partially funded by the H2020 Project “SHERLOCK Seamless and safe human centered robotic applications for novel collaborative workshops” (GA: 820689) (www.sherlock-project.eu) and the HEU Project “CONVERGING Social industrial collaborative environments integrating AI, Big Data and Robotics for smart manufacturing” (GA: 101058521) (www.convergingproject.eu), funded by the EC. References [1] Chryssolouris, G. (2013). Manufacturing systems: theory and practice. Springer Science & Business Media. [2] Wang, L., Gao, R., Váncza, J., Krüger, J., Wang, X. V., Makris, S., & Chryssolouris, G. (2019). Symbiotic human-robot collaborative assembly. CIRP annals, 68(2), 701-726. [3] Makris, S. (2021). Cooperating robots for flexible manufacturing (pp. 123-132). Cham: Springer. [4] Liu, L., Guo, F., Zou, Z., & Duffy, V. G. (2022). Application, development and future opportunities of collaborative robots (cobots) in manufacturing: A literature review. International Journal of Human– Computer Interaction, 1-18. [5] Michalos, G., Makris, S., Tsarouchi, P., Guasch, T., Kontovrakis, D., & Chryssolouris, G. (2015). Design considerations for safe human-robot collaborative workplaces. Procedia CIrP, 37, 248-253. [6] Vysocky, A., & Novak, P. (2016). Human-robot collaboration in industry. MM Science Journal, 9(2), 903-906. [7] Dimitropoulos, N., Michalos, G., & Makris, S. (2021). An outlook on future hybrid assembly systems-the Sherlock approach. Procedia Cirp, 97, 441-446. [8] Michalos, G., Karagiannis, P., Dimitropoulos, N., Andronas, D., & Makris, S. (2022). Human robot collaboration in industrial environments. The 21st century industrial robot: when tools become collaborators, 17-39. [9] Ferreira, M. I. A., & Fletcher, S. R. (Eds.). (2022). The 21st Century Industrial Robot: When Tools Become Collaborators. Springer. [10] ISO 10218-1:2011 Robots and Robotic Devices -Safety requirements for industrial robots -Part 1:Robots, 2011. [11] ISO 10218-2:2011 Robots and Robotic Devices -Safety requirements for industrial robots -Part 2:Robot systems and integration, 2011. [12] ISO/TS 15066:2016 Robots and Robotic Devices -Collaborative Robots,2016. [13] Katsampiris-Salgado, K., Haninger, K., Gkrizis, C., Dimitropoulos, N., Krüger, J., Michalos, G., & Makris, S. (2024). Collision detection for collaborative assembly operations on high-payload robots. Robotics and Computer-Integrated Manufacturing, 87, 102708 [14] Matayoshi, C. T., & Batalha, G. F. (2007). Concurrent engineering– process requirements and life cycle influences on the development of automobile stamped parts. COBEF. [15] Mayhew, D. J., & Bias, R. G. (Eds.). (2005). Cost-justifying usability: An update for an internet age. Morgan Kaufman. [16] Huck, T. P., Münch, N., Hornung, L., Ledermann, C., & Wurll, C. (2021). Risk assessment tools for industrial human-robot collaboration: Novel approaches and practical needs. Safety Science, 141, 105288.