Research Publications
Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence
Arslan Chaudry*, Puneet K Dokania*, Ajanthan Thalaiyasingam* and Philip H S Torr
European Conference on Computer Vision (ECCV), pages 532-547, September 2018
Research into safety in autonomous and semiautonomous 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 paper studies the application of formal methods to the verification of a human driver model built using the cognitive architecture ACT-R, and to the design of correct-by-construction Advanced Driver Assistance Systems (ADAS). The novelty lies in the integration of ACT-R in the formal analysis and an abstraction technique that enables finite representation of a large dimensional, continuous system in the form of a Markov process. The situation considered is a multi-lane highway driving scenario and the interactions that arise. The efficacy of the method is illustrated in two case studies with various driving conditions.
Download the full paper