论文部分内容阅读
实现了一种对球形移动机器人的滚动速度进行控制的方法.球形移动机器人的控制输入和状态输出间存在难以精确数学描述的非线性关系,本文采用径向基函数神经网络,以在线训练的方式建立了球形机器人输入与输出的非线性映射;然后采用反馈线性化方法,设计了球形机器人的速度控制器,该控制器由反馈线性化控制器和减小神经网络逼近误差的补偿控制器构成;给出了该控制器的实现步骤.多次实验结果表明,该方法可以实现球形移动机器人稳定的速度控制.
A method of controlling the rolling speed of a spherical mobile robot is achieved.However, there is a nonlinear relationship between the control input and the state output of a spherical mobile robot that can not be accurately described mathematically. In this paper, a radial basis function neural network The nonlinear mapping of spherical robot input and output is established. Then the feedback linearization method is used to design the speed controller of spherical robot. The controller is composed of feedback linearization controller and compensation controller to reduce the approximation error of neural network. The implementation steps of the controller are given.Many experiments show that this method can achieve stable velocity control of spherical mobile robot.