Research on the Nash Game based Shared Control and its Critical Testing Scenario Generation Method

Abstract

The shared control system is aiming at improving both driving safety and comfort. The testing scenario generation method, especially for critical scenarios, is the vital step for development and deployment of shared control vehicles. The research presented in this paper focuses on a Nash game based shared control strategy and its critical testing scenario generation method. The driving authority allocation strategy basing on the traffic situation and driving capability is established firstly. In this part, both the traffic situation and driving capability are quantified and a fuzzy rule is proposed to allocate the driving authority. The knowledge of non-cooperative game and Nash equilibrium theory are applied to obtain the optimal shared control strategy. In the given operational design domain (ODD), testing scenarios can be evaluated by a novel proposed indicator, scenario importance, which can be computed as the combination of risk degree and occurrence probability. Test results show that the proposed shared control strategy can intervene drivers’ operation in different scenario risk degrees to ensure safety and improve driving acceptability and the scenario generated method can validate the proposed share control system reasonably and effectively as well.