Game Theory, Proxemics and Trust for Self-Driving Car Social Navigation

To navigate in human social spaces, self-driving cars and other robots must show social intelligence. This involves predicting and planning around pedestrians, understanding their personal space, and establishing trust with them. The present paper gives an overview of our ongoing work on modelling and controlling human–self-driving car interactions using game theory, proxemics and trust, and unifying these fields via quantitative models and robot controllers