Research on multi-factor optimization-based authority allocation strategy for human-machine shared steering control

To ensure the safety and stability of vehicles during human-machine collaborative driving, this paper proposes a multi-factor optimization-based authority allocation strategy to address the distribution of control between the driver and the automation system. First, to minimize the impact of human-machine conflicts on the vehicle, we design a driving intention consistency allocation strategy model (DICS) using a cosine similarity algorithm. Next, we develop a driver’s driving ability-based authority allocation strategy (DDAS) using the Hamiltonian Monte Carlo algorithm, which allocates control based on an evaluation of the driver’s driving ability. Additionally, a spatial collision risk-based authority allocation model (SCRM) is established using Artificial Potential Field (APF) and Dynamic Potential Field (DPF) methods to assess the impact of environmental conditions on authority allocation. Finally, a fuzzy inference algorithm is adopted to dynamically perform comprehensive internal allocation, with its scope determined by the allocation values corresponding to DICS, DDAS, and SCRM. Simulation results demonstrate that the proposed strategy effectively coordinates human-machine conflict, driver capability, and environmental risk, enhancing vehicle safety and stability during collaborative driving. The crucial code can be obtained from the provided link: https://github.com/gyhhq/human-machine-cooperative-driving .