A Motion Planning Of Manipulator In Limited Space

Based on 8-DOF redundant manipulator, this paper studies the problem of collision detection and motion planning of redundant manipulator in limited space. Redundant 8-DOF robots could adjust joints position to avoid the singular configuration and optimize the robot arm trajectory with the redundant joints. In this paper, we focus on the motion planning in limited space. The main research contents are as follows:Firstly, established the linkage coordinate system of the manipulator in the limited space with the D-H coordinate method, and analyzed the basic theory of the manipulator. The kinematic model of the manipulator is established according to the DH parameter in the limited space. Define limited space and acquire the limited space boundary parameters of the manipulator. Then the limited space is analyzed to obtain the working space of the manipulator.Secondly, the collision detection algorithm of the manipulator is studied, and the conditions of the collision detection between the arm and the obstacle are analyzed. The algorithm of the AABB bounding box algorithm, the AABB hierarchical bounding box tree algorithm and the disadvantage of AABB in constructing the bounding box are introduced,and propose an improved approach. construct the hierarchical bounding box tree of the robot arm and the obstacle, including the selection of the hierarchy bounding box tree, the determination of the construction method of the bounding box tree, and the update of hierarchy bounding box tree. Finally, the feasibility and efficiency of the algorithm are verified by experiments.Thirdly, introduce the principle of the original RRT algorithm, the extension of the RRT algorithm and the characteristics of the other algorithms that the original RRT algorithm do not have. The Bidirectional fast random-expansion tree algorithm and its advantages and disadvantages are compared with the RRT algorithm. The result shows that the bidirectional fast random expansion tree algorithm can improve the planning efficiency better, but can not get the optimal path. Therefore,the RRT * algorithm is studied on this basis. In order to verify the effectiveness of the RRT * algorithm, the RRT algorithm and the RRT *algorithm are simulated in different environments under the same conditions, and the RRR and RRT * algorithm in different environments are tested and the RRT * algorithm test trend of the optimal value over time.Finally, the simulation experiment of 8-DOF manipulator is built with MATLAB, and the result indicate that there is no collision between the manipulator and the obstacle from the specified starting position to the target pose. It is proved that the RRT * algorithm is feasible for the path planning and could find optimum trajectory over time.