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 neural network compensator based PD control strategy to control robotic manipulators. This control scheme can deal with problems that traditional control methods can hardly achieve satisfied control performance due to coupling effect and strong nonlinearity of robotic manipulators. In this scheme, radial basis function (RBF) network is integrated to compensate the coupling effect and nonlinearity in addition to traditional PD control. Furthermore, the proposed control strategy is based on discrete time and can easily be applied to industrial application. Finally, a hardware-in- the-loop simulation technique based control system was developed and the proposed control scheme was applied to the robotic manipulator. Experimental results demonstrate that the addressed control strategy can easily be implemented and achieve perfect control precision.