An intelligent human-machine interaction-based longitudinal control strategy for autonomous vehicles

In the foreseeable future, the anticipation is that intelligent vehicles will transition to a mode where the intelligent driving system collaborates seamlessly with the human driver. This harmonious integration between the driver and the intelligent control system holds paramount significance for the successful execution of driving tasks, ultimately contributing to the development of more advanced and user-friendly automobiles. A pivotal element in advancing from assisted to autonomous driving lies in the establishment of a human-machine co-driving mode. This research delineates a longitudinal control strategy tailored for intelligent vehicles featuring human-machine interaction. The approach involves the creation of a personalized safe distance model for car-following by collecting driver characteristic parameters. Focused on the car-following methodology, this study formulates the kinematics state space equation, performance index function, and constraint conditions governing car-following dynamics. Subsequently, a car-following control strategy is devised based on model predictive control (MPC), which is addressed through rolling optimization techniques. Building upon this foundation, a human-machine driving control strategy is proposed to dynamically allocate driving authorities in real-time. This strategy takes into account speed and vehicle distance risk as two-dimensional inputs, employing a cooperative driving control strategy within the dual-drive dual-control system. The proposed method was validated in a simulated environment.