论文部分内容阅读
提出了一种带有神经网络补偿的机械手PD控制策略,该方法结合PD控制器和神经网络的优势,解决了在机械手工业应用中,常规的控制策略在处理机械手耦合和非线性特性时控制效果差的问题。该方法基于常规的PD控制策略,采用径向基(radial basis function,RBF)神经网络动态补偿机械手系统的非线性,改善系统的控制性能。该文的控制策略是基于离散时间模型的,可以直接应用到控制系统中。为实现该文控制方法,开发了基于半实物仿真技术的开放式机械手平台,并且在该平台上对该方法进行了实验研究,实验结果表明:该文所提的控制策略实现简单,同时具有较高的控制精度。
This paper presents a robot PD control strategy with neural network compensation. This method combines the advantages of PD controller and neural network, and solves the problem that in the robot industrial application, the conventional control strategy can control the coupling effect and nonlinear characteristic of the manipulator Poor problem. Based on the conventional PD control strategy, this method uses radial basis function (RBF) neural network to dynamically compensate the nonlinearity of the manipulator system and improve the control performance of the system. The control strategy of this paper is based on discrete time model and can be directly applied to the control system. In order to realize this control method, an open manipulator platform based on hardware-in-the-loop simulation technology is developed, and the method is experimentally studied on the platform. The experimental results show that the control strategy proposed in this paper is simple, High control accuracy.