Adaptive control for trajectory tracking of autonomous vehicles with state constraints

This paper presents an adaptive control method for trajectory tracking in autonomous vehicles. The method handles state constraints and improves tracking accuracy under compound perturbations. The approach involves the design of a time-varying error boundary based on the trajectory tracking system’s error trend, along with an enhancement of the tangent-type barrier Lyapunov function. A path lateral tracking controller is developed through the backstepping control method, incorporating a time-varying tangent-type barrier to maintain the combined error within predetermined boundaries. Adaptive laws for neural network weight coefficients are introduced to approximate both the composite perturbation and virtual control law derivatives. System stability over finite time is demonstrated through Lyapunov stability analysis. The effectiveness of the proposed method in improving tracking accuracy and stability is confirmed through comprehensive simulations, hardware-in-the-loop (HIL) testing, and real-vehicle experiments.