Background
Current Level 3 Autonomous systems such as Tesla FSD or Hybrid Closed Loop systems such as Medtronic 770 have both humans in charge of key decisions as well as are part of the system leading to critical need for new modeling and simulation techniques which takes into account this unique characteristics of these systems: Human in the Loop (HIL) and Human in the Plant (HIP) – generically termed as HIL-HIP systems – for such important tasks as Safety evaluation/certification, Testing (test case generation), Personalization within safety envelop, and trustworthy system design.
Simulating complex systems (e.g. AI-enabled cyber-physical systems) that involve human interaction, such HIL-HIP systems, presents significant challenges. These systems are categorized by the continuous integration of human decisions and actions within automated processes, making them inherently dynamic and difficult to accurately model. Traditional simulation methods often struggle to balance speed and accuracy when dealing with time-varying components and real-time human intervention. There is a need for a more efficient and accurate approach for advancing technologies like Level-3 autonomous systems, hybrid close-loop medical devices and wireless mobile networks.
Invention Description
Researchers at Arizona State University have developed an innovative Piecewise Linear Time-Invariant Simulation (PLIS) framework, which significantly improves speed and accuracy in systems that integrate human actions within wireless close-loop mobile networks. PLIS addresses the complexities in simulating the Human in the Loop Human in the Plant (HIL-HIP) systems. By discretizing time-varying components into linear segments, the PLIS approach significantly enhances simulation speed, while maintaining accuracy as demonstrated in a case study involving an artificial pancreas for Type 1 diabetes patients and Braking in Level 3 autonomous vehicle.
Potential Applications:
- Trustworthy Healthcare systems designed to manage chronic conditions like free-living or in-hospital diabetes care
- Robust Infrastructure planning and testing automation software engineering tool
Benefits and Advantages:
- Achieves more than double the simulation speed, with the potential for an 8x increase, all while preserving high fidelity
- Utilizes systematic error bounding methods to ensure accuracy
- Expediting certification and time-to-market of AI-enabled systems
Related Publication: High Fidelity Fast Simulation of Human in the Loop Human in the Plant (HIL-HIP) Systems