【摘要】 针对轮式移动操作机器人的非完整性和不确定性,为其设计了鲁棒跟踪控制器.即采用分段运动学到动力学方法为移动平台设计动力学跟踪控制器,并分别为移动平台和机械手设计神经网络控制器,利用遗传算法来搜索神经网络的优化权系值,从而补偿平台与机械手间的不确定量.通过Matlab的Simulink以及CMEX文件对所设计的控制器仿真,结果表明此控制器具有较强的鲁棒性.更多还原

【Abstract】 Focusing on the nonholonomy and uncertainty of wheeled mobile manipulators, a robust following controller was presented, the dynamic following controller was designedby the method of backstepping kinematics into dynamics, and the neural network controllers for the mobile platform and manipulator by the genetic algorithm to search for the neural network’s weights. The uncertainty terms were compensated. The robust controller was simulated using the simulink of Matlab and C MEX-file. The results show the controller has better robustness.更多还原