Real - Time Obstacle Avoidance Algorithm for Fruit and Vegetable Picking Manipulator

  • Rui Tang

Abstract

Research on obstacle avoidance planning algorithm of picking manipulator is the key to improve the efficiency of fruit and vegetable picking. Based on the structure and parameter design of the picking manipulator, the kinematics characteristics of the manipulator are analyzed, the forward kinematics equation of the 6-DOF picking manipulator is derived, and the obstacle avoidance path planning method in the joint configuration space of the picking manipulator is proposed to avoid the complex calculation of the inverse kinematics solution when the workspace of the manipulator is mapped to the joint configuration space. The cylindrical envelope model of picking manipulator and the envelope model of obstacle's axially bounding box are established, the collision detection method of manipulator is improved, and the collision database of manipulator is constructed to provide data support for three-dimensional path planning of obstacle avoidance. Based on safety criterion and shortest path criterion, heuristic function is introduced, pheromone updating strategy is optimized, ant colony algorithm is improved to ameliorate its convergence speed, make it more problem-oriented, and solve the real-time problem of three-dimensional path planning algorithm for obstacle avoidance of picking manipulator. By comparing the computational results of the improved ant colony algorithm with the fast random tree algorithm, the superiority of the improved adaptive group algorithm in the three-dimensional path planning of picking manipulator obstacle avoidance is proved.

 

Published
2019-08-25