Lateral Trajectory Tracking of Unmanned Vehicles Based on Active Disturbance Rejection Model Predictive Control

In addition to internal state constraints such as lateral velocity and cross-swing angular velocity, external factors, such as a complex time-varying environment, directly impact the tracking performance. For a typical underdriven incomplete constraint system of a vehicle, it is possible to explicitly handle the physical constraints of the system while achieving high accuracy and stable trajectory tracking in the presence of parameter uncertainties and internal/external disturbances. This paper proposes a lateral trajectory tracking control algorithm based on active disturbance rejection model predictive control. By employing the concept of disturbance compensation, a multivariate extended state observer is designed to real time observe and estimate the system state and unknown disturbances, effectively compensating for them to enhance the system's robustness. Furthermore, adopting the principle of optimal control, the optimal control input is determined using a partially known adaptation model, considering the physical constraints of the system, thereby ensuring the stability of vehicle driving. Finally, comparative simulation experiments are conducted to validate the effectiveness of the proposed algorithm.