Adaptive authority dynamic game for human-machine cooperative control

Abstract

An adaptive dynamic allocation strategy for driving authority based on the non-cooperative game was designed to solve the decision conflict problem between the driver and automatic system in human-machine driving. The coupling dynamics model of human-machine interaction was established according to the coupling mechanism between the driver and the automatic system. Considering the driver’s driving status and the driving risk of vehicles, the driving authority dynamic allocation strategy was designed. On this basis, the human-machine cooperative multi-objective optimal control problem is constructed, and the dynamic allocation control system was designed based on the non-cooperative game theory. The optimal control method was used to solve the Nash equilibrium solution and achieve the optimal authority allocation. The effectiveness of the proposed adaptive authority dynamic allocation strategy was verified under normal and emergency conditions. Results showed that the proposed strategy can effectively reduce human-machine decision conflicts and achieve harmonious human–machine cooperative driving.