基于RBF神经网络的多移动机械臂同步控制

Synchronous control of multiple mobile robotic manipulators based on RBF neural network

以多固定基机械臂同步控制为基础,应用RBF神经网络方法,并结合领导者-追随者(主从控制)通信拓扑方法,针对存在不确定性的多移动机械臂系统,设计出一种新的分布式主从同步控制器,并使得控制器的适应性得以提高。从动移动机械臂和领导移动机械臂之间的同步误差由主从式通信拓扑的网络结构来定义。本研究对从动移动机械臂的建模误差等不确定性进行了逼近和补偿,并对领导移动机械臂的控制力矩进行了估计,RBF神经网络的权值根据自适应律在线进行更新。通过Lyapunov方法验证了所设计控制算法的稳定性,最后通过Matlab仿真验证了其可行性。

In this paper,considering the uncertainty of multiple mobile manipulators,a new distributed leader-follower synchronous controller was designed to improve its adaptability by using RBF neural network and leader-follower communication topology based on the synchronous control of multiple fixed base manipulators. The synchronization error between the leading and follower mobile manipulator was initially established using the leader-follower communication topology network structure. Afterwards,by updating the weights of RBF neural network online with adaptive law,the systematic uncertainties of mobile manipulator such as model errors were approached and compensated,and the leading mobile manipulator control torque was estimated. By constructing Lyapunov function,the stability of the system was analyzed. Finally,the effectiveness of the designed controller was verified using Matlab simulation.