Game theoretic approaches for the mathematical control, modelling and optimization of interactions between autonomous vehicles and other road users

The rapid advance in the development of autonomous vehicle (AV) technology has brought their widespread deployment to the near horizon. As well as their use in personal transportation, small robots such as the Starship have been designed to provide small goods deliveries such as groceries. However, deploying AVs in dense urban areas is a major unresolved challenge, given their need to interact safely and efficiently with regular vehicles, pedestrians and cyclists. The purpose of the PhD study is to explore the application of mathematical game theory to represent, understand and design such interactions. Such techniques have been found to be previously useful in designing safe transportation systems (Elvik R, 2014: A review of game-theoretic models of road user behaviour. Accident Analysis & Prevention 62: 388-396). Building on research in the on-going research projects CityMobil2 and interAct, and in the expertise of the supervisors in modelling and optimization of transport systems, the research will involve formulating alternative mathematical models, implementing and testing them computationally, and devising appropriate optimization/control mechanisms for both the AVs and the infrastructure, in order to ensure safe and efficient movement for all road users