VeHICaL: Verified Human Interfaces, Control, and Learning for Semi-Autonomous Systems
The VEHICaL project is developing the foundations of verified co-design of interfaces and control for human cyber-physical systems (h-CPS) --- cyber-physical systems that operate in concert with human operators. VeHICaL aims to bring a formal approach to designing both interfaces and control for h-CPS, with provable guarantees. The project is making contributions along four thrusts: (1) formalisms for modeling h-CPS; (2) computational techniques for learning, verification, and control of h-CPS; (3) design and validation of sensor and human-machine interfaces, and (4) empirical evaluation in the domain of semi-autonomous vehicles. The VeHICaL approach is bringing a conceptual shift of focus away from separately addressing the design of control systems and human-machine interaction and towards the joint co-design of human interfaces and control using common modeling formalisms and requirements on the entire system. This co-design approach is making novel intellectual contributions to the areas of formal methods, control theory, sensing and perception, cognitive science, and human-machine interfaces.
Caltech will participate in research related to specification, design and verification of networked control systems with applications to human-controlled cyberphysical sys- tems (h-CPS). Working jointly with researchers at UC Berkeley, Caltech will extend previous work in synthesis of control protocols for hybrid systems to include interactions with humans and the applications to semi-autonomous vehicles. Caltech will support all program reviews and annual technical reports, in addition to participating in outreach activities.
- Contracts of Reactivity (Tung Phan-Minh and Richard M. Murray, Submitted, Int'l Conf on Formal Modeling and Analysis of Timed Systems (FORMATS) 2019)
- Towards Assume-Guarantee Profiles for Autonomous Vehicles (Tung Phan-Minh, Karena X. Cai, Richard M. Murray, Submitted, 2019 Conference on Decision and Control (CDC))
- Inverse Abstraction of Neural Networks Using Symbolic Interpolation (Sumanth Dathathri, Sicun Gao, Richard M. Murray, To appear, 2019 AAAI Conference on Artificial Intelligence)
- Risk-aware motion planning for automated vehicle among human-driven cars (Jin I. Ge, Bastian Schurmann, Richard M. Murray, and Matthias Althoff, Submitted, 2019 American Control Conference (ACC))
- Voluntary lane-change policy synthesis with reactive control improvisation (Jin I. Ge and Richard M. Murray, To appear, 2018 Conference on Decision and Control (CDC))