Robust path-tracking control for intelligent vehicles considering human–vehicle multi-factors

Human–machine co-driving refers to a technological architecture in which human driver and automated system control vehicle sharing and cooperate to accomplish driving tasks. The architecture could solve the safety and ethical dilemmas of autonomous vehicles, which is an innovative method to improve vehicle intelligence. To realize high-precision and safe path-tracking control, this paper proposes a parallel steering sharing human–machine co-driving framework based on constraint-following controller and driving weight allocation module. First, a parallel human–machine co-driving control framework is established, which uses a two-point preview driver model and constraint-following controller to represent the human driver and automated system, respectively. Second, the underactuated characteristics of front steering vehicles are analyzed. A multiple-constraint path-tracking robust controller is designed based on constraint-following control approach, which considers the multi-source uncertainty and multiple constraints of real road conditions. Third, a multi-factor driving weight allocation module is constructed, which considers human–machine characteristics and cooperation performance. Finally, the effectiveness of the proposed method is verified by CarSim–MATLAB/Simulink joint simulation under three different case studies. This paper creatively synthesizes the vehicle uncertainty, multiple path-tracking constraints, and human–machine cooperation performance, and provides a new idea for human–machine co-driving vehicles to realize robust and safe path-tracking control.