Research Of Binocular Visual Servo Control System Via 6-DOF Industrial Manipulator

Six degree of freedom industrial manipulator is widely used in manufacturing industry, it is imperative to make manipulator more intelligent. Loading binocular vision system is the basic means to achieve intelligent, thus constituting binocular vision servo control system based on six degrees of freedom industrial manipulator. Therefore, it had a far-reaching on researching 6-DOF industrial manipulator visual servo control system. The research work of this paper is divided into two parts, namely, the visual system and the control system.Two fundamental problems about the visual system has been solved, namely, the extraction of the target information and the camera calibration. Extract target information to rely on image processing technology, target information is extracted by image processing technology, including image enhancement, filtering noising, edge extraction and so on. Target detected based on color and shape are introduced respectively. Camera calibration is mainly divided into direct and indirect calibration. In direct calibration, calculate the geometric model of the camera is required, long time either. Therefore, application of optimized Extreme Learning Machine (ELM) via improved Electromagnetism-Like (EM) in camera calibration is proposed, the mapping between the image coordinates and the spatial coordinates is fitted by the ELM, and the step size and spatial domain of EM are improved. Then input weights and the bias of ELM are optimized via improved EM. The experimental results show that, convergence speed is faster, the generalization ability of ELM is enhanced, precision is enhance as well, suitable for spot welding and other high precision requirements place.On the control of the 6-DOF industrial manipulator. Firstly, the kinematics and dynamics of the manipulator are introduced, and the higher order system is divided into two order systems. Secondly, the global fast terminal sliding mode control (GFTSMC) method is used to control the machine body, and the stability, singularity and convergence of the machine are analyzed. The sliding mode control law has a direct influence on the control effect via the analysis, so the parameters of GFTSMC are optimized by the particle swarm optimization algorithm (PSO). Finally, simulation and experiments show that the response speed and convergence rate of optimization system is faster, and chattering is reduced effectively.Finally, through the integration of the two part, a visual servo control system based on 6-DOF industrial manipulator is formed, and the related experiments show that the binocular vision servo system that proposed is effective and feasible.