Improved Model-Free Adaptive Control and Line-of-Sight Guidance for Autonomous Vehicle Path Following

Path tracking control is a fundamental technology in autonomous vehicle applications, but it faces significant challenges related to vehicle modeling and external disturbances. In this paper, an improved model-free adaptive control strategy, leveraging data-driven approaches and gradient descent parameter optimization, is proposed. First, an enhanced lineofsight method is used to generate the desired heading angle, simplifying the multi-dimensional path tracking problem into a more manageable heading angle tracking task. Then, the gradient descent model-free adaptive control strategy is applied, with the core idea of incorporating the gradient descent algorithm in the controller's time-varying parameter update process to achieve optimal control. The effectiveness of the proposed algorithm is validated through simulations, demonstrating its robustness in nonlinear systems and path tracking control scenarios. Comparative analysis with model-based methods highlights the superior stability and adaptability of the approach.