基于神经网络的移动机器人动态面控制OA
A Dynamic Surface Control Based on Neural Network for Mobile Robots
针对具有高度时变参数和未知扰动的移动机器人运动控制问题,以及在其运动过程中的部分子系统间摩擦和耦合问题,文中提出了一种基于非线性扰动观测器(Disturbance Observer,DOB)和径向基函数神经网络(Radial Ba-sis Function Neural Network,RBFNN)的新型动态表面控制(Dynamic Surface Control,DSC)方法.通过神经网络逼近系统动力学中的摩擦等非线性函数.结合扰动观测器处理移动机器系统模型的不确定性和外部未知扰动降低系统的不确定性,实现更精确和稳定的运动控制.采用李雅普诺夫稳定性定理验证闭环系统的稳定性,确保控制方案的可靠性.在MATLAB/Simulink平台对所提控制方案进行仿真,验证了该控制器在各种干扰和参数不确定的情况下能够有效保证移动机器人系统跟踪期望轨迹的鲁棒性和准确性.
In view of the motion control problem of mobile robots with highly time-varying parameters and un-known disturbances,as well as the friction and coupling problems among some subsystems during their motion pro-cess,a novel DSC(Dynamic Surface Control)method based on nonlinear DOB(Disturbance Observer)and RBFNN(Radial Basis Function Neural Network)is proposed.Approximate nonlinear functions such as friction in system dy-namics through neural networks.Combining the disturbance observer to handle the uncertainty of the mobile ma-chine system model and the external unknown disturbance to reduce the uncertainty of the system,more precise and stable motion control can be achieved.The stability of the closed-loop system is verified by using Lyapunov's stabil-ity theorem to ensure the reliability of the control scheme.The proposed control scheme is simulated on the MAT-LAB/Simulink platform,verifying that the controller can effectively ensure the robustness and accuracy of the mo-bile robot system in tracking the expected trajectory under various interferences and parameter uncertainties.
薛浩然;鲁守银;杨钊懿;王荣业;陈洪健
山东建筑大学 信息与电气工程学院,山东 济南 250100山东建筑大学 信息与电气工程学院,山东 济南 250100山东建筑大学 信息与电气工程学院,山东 济南 250100山东建筑大学 信息与电气工程学院,山东 济南 250100山东建筑大学 信息与电气工程学院,山东 济南 250100
信息技术与安全科学
移动机器人动态面控制运动控制扰动观测器神经网络轨迹跟踪未知扰动非线性函数
mobile robotsdynamic surface controlmotion controldisturbance observerneural networktrajectory trackingunknown disturbancenonlinear function
《电子科技》 2026 (9)
19-25,7
山东省自然科学基金(2019JZZY010435,ZR2022MF309)Natural Science Foundation of Shandong(2019JZZY010435,ZR2022MF309)
评论