Trust-Based Shared Steering Control With Game Theory Using Event-Triggered ADP for Human-Vehicle Cooperative Driving

The trust level of drivers in automated driving systems (ADS) significantly impacts human-vehicle cooperation in terms of driving authority allocation and dynamic interactions during collaborative driving. However, trust dynamics for human-vehicle shared steering control (SSC) has rarely been researched. This paper presents a quantitative trust model and then proposes a trust-based cooperative driving strategy. Three contributions have been made in this paper: 1) An enhanced quantitative trust dynamics model is introduced, integrating real-time driving risk estimation and the deviation of driving operation expectations between human and vehicle; 2) A dynamic authority allocation strategy is proposed using fuzzy inference, providing real-time control authority based on trust and its changing rate; 3) A non- zero-sum (NZS) differential game framework is constructed, and an event-triggered adaptive dynamic programming (ET-ADP) method is adopted for Nash equilibrium control strategy derivation. The effectiveness and rationality are validated through high-fidelity Carsim-Simulink co-simulations and driver-in-the-loop (DIL) experiments, demonstrating the proposed model can achieve optimal control while reducing the driver operational workload and improving trust levels.