Impedance Control of Industrial Robot Based on Multi-Directional Particle Swarm Optimization

The effectiveness of impedance control is significantly influenced by the appropriateness of the settings for impedance parameters. In This paper, a new multi-directional particle swarm optimization (MDPSO) algorithm is proposed to obtain the optimal impedance parameters for precise force tracking during the movement of industrial robot. Firstly, the position-based impedance controller between the industrial robot and its environment is established, and the difference of the contact force is used as input to the controller to obtain the desired trajectory. Then, enhancements to the particle swarm optimization (PSO) algorithm are introduced by adding additional optimization directions, thus developing multi-directional particle swarm optimization algorithm. The MDPSO algorithm determines the best individuals in terms of response speed, steady state error, oscillations and overshoot as additional directional data to improve the efficiency during the iterative process. Furthermore, the algorithm incorporates mutation operation on the particles to avoid falling into local optimum. The simulation results demonstrate that the MDPSO algorithm surpasses the traditional PSO algorithm in both convergence speed and accuracy. Additionally, experimental results show that excellent force tracking performance can be observed when the parameters derived from the MDPSO algorithm are applied for impedance controller.