Research On Intelligent PID Control Method For Robot Manipulators

With the rapid development of science and technology,as well as the continuous improvement of our country industrialization level,industrial robot installation quantity increased year by year in our country.Industrial robot is indispensable digital industrial equipment in Industrial production process which has become a supporting technology of advanced manufacturing industry.Now the installation quantity of industrial robot has become an important standard to judge a country industrialization level.With China’s demographic dividend dropped year by year and the rising labor costs,the robot industry in China has entered the important opportunity of development.But our country industrial robots rely on a large number of imports,it is necessary to realize the development of domestic robot industry.At this time,this research has good prospects for development and important research significance.This article takes a boiler water wall welding manipulator for thermal power plant in Liaoyang city as the background.By the research status at home and abroad for manipulator control and in view of the robot manipulator trajectory precision tracking control problem,intelligent control combined with PID control method is studied.Industrial robot is given priority to with single joint PID control.With the continuous development of intelligent control and the requirement of control accuracy in the manufacturing industry,first of all,using neural network to PID parameters setting,to improve the performance of PID controller.Because of the single joint control dissever the integrity of the system,the single joint PID control cannot solve the problem of the strong coupling of system.The PID neural network method is used on the manipulator trajectory tracking control to solve this problem.The strong coupling of system can be solved by using the PIDNN control method,but external interference and itself modeling uncertainty of system is ignored.The RBF neural network adaptive method based on the model is used to solve the problem.The RBFNN is used to approximate model of the system.This method is used on the trajectory tracking control of robot arm and some improvement of algorithm is made.Finally on the basis of the simulation results,the various algorithms are compared and analyzed.