Project Details
Projekt
Context-Aware Design of Human-Cyber-Physical Systems
Subject Area
Automation, Mechatronics, Control Systems, Intelligent Technical Systems, Robotics
Human Factors, Ergonomics, Human-Machine Systems
Human Factors, Ergonomics, Human-Machine Systems
Term
from 2020 to 2024
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 433524434
Cyber-Physical-Systems steadily gain in autonomy while interacting and cooperating with humans, which increases the potential for conflicts. For systematically designing such systems, an appropriate design process is needed that can take into account the impact on and interrelation with humans. To this end, questions for mechanisms arise to implement human users/operators, social, ethical, and legal requirements for on-line conflict resolution in distributed systems to achieve user/operator acceptance, (safe) usage, and provable adherence to domain-relevant legal regulations.We will develop an innovative and explorative method to integrate users and operators into the HCPS and generate an efficient and effective human-technology system. Hereby, we draw on the lab facilities of the applicants, specifically, on facilities from two domains: energy and transportation. All investigations are done along two use cases in order to ensure coverage of a broad range of interactions and cooperation scenarios between humans and diverse Cyber-Physical-Systems to ensure generalizability.Specifically, we will develop objective, reliable, and valid methods suitable for investigating, how justifications of Human- Cyber-Physical-Systems (HCPS) behavior as generated by SP2 can be communicated to humans interacting and cooperating with the system. This communication should ensure an optimal level of trust, user acceptance, and usage.We will develop a flexible, general design process for new HCPS that can be individually instantiated for different contexts. This design process incorporates soft (e. g. usability studies) as well as quantitative methods (measurements-based) and is parametrizable with different models e. g. for user states, mental/moral constitution, temporal awareness, situational operation state, or national/regional characteristics.In order to achieve this main goal, we will setup and demonstrate two laboratory configurations for analyzing and training humans in HCPS on-line conflict resolution settings in diverse scenarios highlighting different domain-dependent social, ethical and legal requirements. Knowledge about users’ mental or motivational state (from SP1) is integrated. We instantiate a formal game-theoretic model (from SP2) in a multi-domain co-simulation architecture (with human operators in the loop) and empirically investigate, how justifications of HCPS behavior generated from this model are communicated to operators cooperating with HCPS so as to improve trust, usage and acceptance. Domain-specific reasoning aspects (justifications, explanations) are modeled as domain ontologies for automated processing. We will develop reliable and valid methods suitable for assessing and investigating how justifications within HCPS are communicated and perceived, understood and accepted by humans in situationally constrained (stress, conflicts of interest, intellectually overstraining) situations with model parameters from SP4.
DFG Programme
Research Grants
