The Research On The Robot Networked Control Systems Based On Neural Network

Along with the development of computer and Internet technology, the distance of remote control keeps on increasing. The research of Internet-based remote control has been paid more and more attention. Because the Internet is nonreal-time, the information transmitted through it will inevitably be delayed. Tests show that the delay is uncertain and the delay will also cause data loss. The transmission delay and its uncertainty will reduce the control quality; this will affect the application of the networked control system (NCS). So, we should study the effect of the above factors on the remote control, and find new control algorithm and the compensation measures. In this thesis, by making the AS-R robot, made by Shanghai Grandar Company, as the control object, we analyse the network delay and it's effect on control results, and propose a solution.This thesis is divided into six parts. Chapter 1 briefly introduces the research results of NCS, and the backgrounds, and the significance of the topic. Chapter 2 introduces the GPC and neural network. In chapter 3, aimed to the robot which uses PC as core controller, compared with normal identification method, a new method of parameter identification which based on the angle model of the robot's driving motor is presented. This method avoids the difference operation of the angle data fed back by two wheels'encoder, and effectively improves the accuracy and anti-jamming capability of parameter identification. In chapter 4, according to previous research results, a new neural-predictive control algorithm, whose prediction steps are variable, is presented. By forecasting the fluctuation of the transmission delay, the algorithm can change the prediction steps, which ensures the control effect of the system. Chapter 5 introduces the design and implementation of the NCS. Because the implementation of control algorithm is limited by many factors, this chapter briefly studies the simulation of real networks. Chapter 6 summarizes the whole work, and gives some suggestions for future study.