Time-optimal trajectory planning of collaborative robots based on improved exponential optimization algorithm

Abstract

To improve the operational efficiency of collaborative robots and ensure trajectory smoothness, a time-optimal trajectory planning method based on improved exponential optimization algorithm is proposed. Strategies including opposition based learning, dynamic population reduction, hybrid adaptive search, adaptive weight, elite preservation and boundary constraint are introduced to enhance the algorithm performance. Taking Diana7 7-DOF manipulator as the object, kinematic model is established via D-H parameter method, and fifth-order polynomial interpolation is adopted for trajectory planning. Simulation is carried out with MATLAB Robotics Toolbox. Results show that the manipulator model is accurate and the algorithm is reliable. Compared with the original scheme, trajectory time is shortened by about 65%, with smooth and oscillation-free motion, which effectively improves the efficiency of physical therapy and massage.