An Efficient Artificial Bee Colony Algorithm for Analytical Inverse Kinematics of Redundant Manipulator

In this paper, a Cauchy refraction-based Artificial Bee Colony algorithm (CRABC) is proposed for analytical inverse kinematics of redundant manipulators. Based on ABC, refracted opposition-based learning strategy and Cauchy operator are combined to design a food source update strategy, which can enhance the convergence accuracy and speed. In addition, in the scout bee phase, a new individual is generated by the proposed update strategy rather than random generation. To verify the effectiveness of CRABC, a 6-DOF manipulator was used to explore its inverse kinematics problem whose solutions need to meet the requirements of accuracy, continuity and real-time performance. A fitness function consisting of position accuracy and joint variation is proposed for solving the designed minimization optimization problem, where joint variations are set as a penalty term for the constraint processing. We compared CRABC with classical ABC and other intelligent optimization algorithms on a single target point, a straight trajectory and a circular trajectory. The results show that CRABC has high position accuracy and smooth joint movement in solving the inverse kinematics problem, and is very suitable for the operation tasks that require high accuracy.