ACP-Incorporated Perturbation-Resistant Neural Dynamics Controller for Autonomous Vehicles
Ying Liufu, Long Jin, Mingsheng Shang, Xingxia Wang, Fei–Yue Wang
Abstract
Autonomous vehicle control systems are unavoidably influenced by diverse noise perturbations from the unpredictable external environment and internal system. In this consideration, based on the model predictive control (MPC) strategy, a perturbation-resistant neural dynamics (PRND) controller equipped with the noise-suppression ability for the path-tracking control of autonomous vehicles is newly designed in this paper, under the framework of artificial systems, computational experiments, and parallel execution (ACP). In addition, theoretical analyses show that the proposed ACP-incorporated PRND controller can behave with exponential convergence and strong robustness under different noise scenarios. Lastly, computational experiments are conducted and parallelly executed on the CarSim-Simulink platform and E-Car physical platform to demonstrate the effectiveness and superiority of the proposed controller. Overall, this paper provides a new perspective for designing neural-dynamics-based controllers for autonomous vehicles, thereby guaranteeing reliable control performance and effectively resisting noise perturbations.