Real-Time Inverse Kinematics Using Dual Particle Swarm Optimization DPSO of 6-DOF Robot for Nuclear Plant Dismantling

Abstract

Robotic manipulator inverse kinematics (IK) solution has a significant role in robotics. In this paper, a new paradigm of particle swarm optimization (PSO) called dual particle swarm optimization (DPSO) for the inverse kinematics problem of a robotic manipulator used in the nuclear plant decommissioning process is proposed. The introduced approach of particle swarm optimization relies on the approach of dividing one particle swarm optimization algorithm into two such that each algorithm optimizes separately the problem for position and orientation, in this way the system will achieve faster convergence toward desired results with fewer iterations. The proposed algorithm is designed and implemented in MATLAB/Simulink to solve the inverse kinematics problem for a 6 degree of freedom selective compliance articulated robot arm (SCARA) type robot with joint space constraints. For a real-time experiment, a pair of the joysticks was integrated with MATLAB/Simulink. Using a dual joystick, the position and orientation for the robot were set. Furthermore, the experimental results demonstrate the effectiveness of DPSO on realtime within given constraints.