Research into safety in autonomous and semi-autonomous vehicles has, so far, largely been focused on testing and validation through simulation. Due to the fact that failure of these autonomous systems is potentially life-endangering, formal methods arise as a complementary approach. This project studies the application of formal methods to the verification of a human driver model built using the cognitive architecture, and to the design of correct-by-construction Advanced Driver Assistance Systems (ADAS).
The aim is to create a rigorous framework for the design of ADAS in semi-autonomous vehicles that accounts for the congestive process of the driver. To illustrate the validity of the approach, various driving scenarios will be studied, including multi-lane highway driving scenarios, in which interactions with other vehicles arise.
F. Eiras and M. Lahijanian, “Towards Provably Correct Driver Assistance Systems through Stochastic Cognitive Modeling,” in Robotics: Science and Systems Workshop on Robust Autonomy: Tools for Safety in Real-World Uncertain Environments, 2019.
M. Wu et al., “Gaze-based Intention Anticipation over Driving Manoeuvres in Semi-Autonomous Vehicles,” in Proceedings of the IEEE International Conference on Intelligent Robots and Systems, 2019, pp. 6210–6216.
F. Eiras, M. Lahijanian, and M. Kwiatkowska, “Correct-by-Construction Advanced Driver Assistance Systems based on a Cognitive Architecture,” in 2019 IEEE 2nd Connected and Automated Vehicles Symposium, 2019, pp. 1–7.