In light of the problem of trajectory tracking control in vehicle servo systems with system model uncertainty and external time-varying disturbance, an effective trajectory tracking control method that can handle system model uncertainty and external time-varying disturbances is proposed. To achieve this goal, a novel composite robust integral of the sign of the error (RISE) control method is introduced that combines a multi-layer neural network and an extended state observer. Specifically, multi-layer neural networks are utilized to approximate the uncertainty of the system model, while an extended state observer is employed to estimate the fitting near-error and the external time-varying interference, which are used as feedforward compensation. Finally, the RISE controller is implemented as a robust feedback controller. By applying Lyapunov theory for stability analysis and conducting experiments, the results demonstrate that the proposed approach exhibits excellent performance and robustness in addressing the uncertainties and disturbances involved in trajectory tracking control for vehicle servo systems.

中文翻译：

---

未知建模不确定性和未知时变扰动的车载伺服系统复合RISE控制

针对系统模型不确定性和外部时变干扰的车辆伺服系统轨迹跟踪控制问题，提出一种能够处理系统模型不确定性和外部时变干扰的有效轨迹跟踪控制方法。为了实现这一目标，引入了一种新颖的复合鲁棒误差符号积分（RISE）控制方法，该方法结合了多层神经网络和扩展状态观察器。具体来说，利用多层神经网络来近似系统模型的不确定性，同时利用扩展状态观测器来估计拟合近误差和外部时变干扰，并用作前馈补偿。最后，RISE 控制器被实现为鲁棒反馈控制器。通过应用李雅普诺夫理论进行稳定性分析并进行实验，结果表明该方法在解决车辆伺服系统轨迹跟踪控制中涉及的不确定性和扰动方面表现出优异的性能和鲁棒性。