Forward Kinematics of Parallel Manipulators Based on Differential Evolution

The forward kinematics problem of a parallel manipulator was transformed into an optimization problem which is easy to obtain its inverse kinematics.A differential evolution(DE) algorithm was used to directly obtain a globally optimal solution of forward kinematics by minimizing the difference between the computed and the given link length.The computed length of each link can be obtained by solving the inverse kinematics.The numerical simula-tion results of a 6-SPS parallel manipulator show that compared with genetic algorithm,the DE algorithm performs well in terms of the quality of the solution and the speed of convergence.The errors of the position and pose are smaller than 0.000 1 mm and 0.000 10 after 508 generations.The proposed method can not only avoid the complicated mathematical derivation and the selection of initial iterative values,but also get high accuracy of forward kinematics,which provides a new way for solving the forward kinematics of parallel manipulators.