Intelligent Predictive Control of a Six-Joint Robotic Manipulator Towards a Vision Based Application

Recently, the automation systems with vision based robots have taken a great deal of attention because of the advances in computer technology and expanding demand for such techniques in the industry. In this study, an intelligent robot control system was designed based on using Generalized Predictive Control (GPC) and Elman neural network to reduce the process time towards a vision based study. A six-joint robotic manipulator was firstly controlled by using GPC. A prepared simulation software was used in the implementation of GPC towards vision based application. Meanwhile, the control parameters are recorded in a file. Then, using this data set an Elman neural network was trained to control the robotic manipulator. On the other hand, an industrial vision system was designed and the control simulations are done according to the designed vision system. Image processing algorithms and GPC algorithm take too much process time that is why to reduce the process time of the control algorithm will be useful for an industrial application. The results of the robot control simulations and implemented image processing algorithms were given.