Kinematics Analysis And Trajectory Planning Of A Three-arm Robot

With the development of space and robot technology, space robots will play an important role in space exploration in the future. A small robot which crawling on the surface of the space station, can be used for the inspection and maintenance of the station’s external structure, and it can solve some task flexibly than large arm located on the outside of the space station. The subject of this paper is to study the three arm robot which can use the truss structure that lying on the surface of space station for crawling. And this paper mainly studies the transition process of the robot in the space truss structure.According to the requirements of the three-arm robot on orbit mission, firstly, make a description of the robot’s climbing transition based on coordinate transformation. Then, use the improved D-H method to establish the link coordinate parameters of the robot, and obtain its forward kinematics. The calculation of the robot’s inverse kinematics can be classified by its movement mode and condition. When the robot moves along a straight line in the plane,the transition of the robot is simplified as 4 DOF manipulator, and obtain its inverse kinematics by geometric method. When the robot moves in three dimensional space,there are two cases.If the robot’s moving environment is uncomplicated,the transition of the robot can be simplified as 6 DOF manipulator, and calculate its inverse kinematics by normal algorithm. If there are obstacles between the robot and the truss, analyze this process by 8DOF mechanical arm,and use some ways to establish constraint equations, which usually utilize the performance requirement of the redundant DOF robot, after that solve its inverse kinematics based on projection gradient method.In view of the fact that the truss structure lying on the surface of space station is square-shaped, the problem of calculating graspable regions can be simplified as the intersection between square bar’s axis and robot’s workspace. After research and analysis ofthe intersection between square bar’s axis and robot’s workspace, and in line with the geometrical characteristics of the graspable regions, the planning method of graspable point based on the classification of graspable regions is proposed.By studying the trajectory planning algorithm of point-to-point in the joint space, determining using the B-spline as the interpolation function of robot trajectory planning, and using genetic algorithms to solve the problem of trajectories’ optimizing time.