Study Of Force/Position Impedance Control Of Industrial Robots

With the continuous expansion of industrial robot applications, the kind of tasks it can do are increasing. For grinding, polishing, assembly, the contact task category, merely position control is incompetent, because in this kind of tasks that have the contact force requiring, tiny position deviation at the end-effector of the robotic manipulator is likely to lead to a huge contact force, so the manipulator and the target will cause a certain degree of damage. In contact type of task, contact force control is inserted in the control system. This dissertation mainly discusses about the industrial robot force/position control issue.In order to improve the performance of force/position tracking control of industrial robots, in this dissertation, the strong feasibility of impedance control method chosen as the main control scheme through several kinds of control strategy that is applied at the present stage are analyzed and compared. The fuzzy inference and neural network inverse system are used in the control system to improve the environment adaptability of the impedance control system and precision of the location tracking, so that the industrial robot control system can achieve better performance of force/position tracking control. The main research work in this dissertation includes:First, by comparing the advantages and disadvantages of four kinds of force control strategies, the characteristics of each control strategy are clarified, and the impedance control method is selected as for its characteristics of robustness.Second, D-H method is employed for the analysis of inverse kinematics of industrial robot. An effective kinematics transform matrix is established, and the reverse solution method is used for the solution of the manipulator motion equation. On the basis of kinematics analysis of the manipulator, Lagrange method is used for dynamics analysis and derivation of the model. After the model of the industrial robot is established, and analysis of traditional position based impedance control, the fuzzy control principle is employed for the regulation of the coefficients of the PD controller designed and self-tuning regulation of the parameters of the impedance model, respectively, and design a set of improved impedance based on position control system, fuzzy impedance control system based on fuzzy PD controller. The impedance of the improved control system possesses good force/position tracking performance and environment adaptability.Then, according to the understanding of the inverse system, a controller is designed for the robotic manipulator as the original system with neural network inverse system, taking the advantages of its own, it is used to the industrial robot impedance control, establishing a fuzzy impedance based on neural network inverse system control system. Due to the introduction of the neural network inverse system impedance control system structure was optimized, and applicability was further improved.Finally, SimMechanics module is used to the simulation modeling of robotic manipulator., fuzzy impedance control system based on fuzzy PD controller and fuzzy impedance based on neural network inverse system control system are established, respectively, under the Matlab/Simulink simulation platform. Under specific experimental setting force/position tracking simulation, the simulation results verify the effectiveness of the proposed control method, the applicability is enhanced, the tracking precision is increased, the convergence speed is faster, and oscillation amplitude is smaller, etc.Through the above research work, with the help of the fuzzy inference system, neural network inverse system, the designed impedance control system possesses good force/position tracking capabilities, and can achieve a good effect of industrial robot force/position control.