The Planning Of Collision Avoidance Based On Human-Robot Cooperation For Remote Manipulator

Robots are stepping out of factory into people's daily life. From small UAV, electric self-balancing scooter and smart sweeping robot, to service robot and medical robot, robots will be much closer to us. The world's largest economy, China, is faced with population aging,rising labor costs and other issues. In order to solve these problems, China has begun to use advanced manufacturing robots in plants to improve the industry. Robots in the manufacturing sector will continue to make vital contribution.Robots are playing such a great role in human production and human life,but imperfect technology of intelligent robot has greatly limited the applying of robots in human society.Therefore according to the current needs, robot-related issues are classified as: collision avoidance, collision detection, tele-operation, man-robot cooperation, multi-robot cooperation,etc. The study of these directions will continuously improve robot technology and intelligence.This paper studies a variety of control modes for robot collision avoidance based on man-machine cooperation. The robot control interface can help switch between various control modes rapidly,and make it easy to implement collision avoidance for manipulator.Using GJK collision detection algorithm, the shortest distance between the robot and the known environment can be calculated in real-time,as well as a set of closest points. Both of them are the basis of collision avoidance planning. Based on the information,it is capable to improve the artificial potential field method and achieve collision avoidance for manipulator.Specific contents are as follows:First, this paper models the collision avoidance environment. It creates models for each object in the environment that the manipulator exists and proposes a representation approach for multi-rigid-body: common tree structure. An object comprises a number of sub-objects, as well as a sub-object contains a number of single rigid bodies or convex polyhedron,and once the main object is created, the relationship among the various levels is confirmed.Second, this paper studies collision detection and collision avoidance. Collision avoidance in the paper, mainly means that the manipulator moves from the current position-orientation to desired position-orientation, at the same time avoiding obstacles in the path under a certain control mode, finally completing the task without collision. This paper also expounds the principles of collision avoidance, related collision detection algorithms, and compares the currently mainstream collision detection algorithms, finally selects the GJK collision detection algorithm.Then, this paper studies the performance of collision avoidance for the semi-autonomous manipulator based on human-robot cooperation. By GJK algorithms, it's able to access collision information of robot in the surrounding environment: whether collision happens, the shortest distance and the multiple groups of point set. In this paper, it constructs the collision avoidance environment by artificial potential field method and leads the manipulator to move towards the direction while the potential declines. Therefore, the manipulator can reach the desired position and orientation with collision-free.Finally, it carries out experiments of collision avoidance for manipulator based on human-robot cooperation. This paper describes the composition of experiment system and collision avoidance control system, as well as functional units in the human-computer control interface. It also provides the experimental data of collision avoidance for the manipulator in different motion patterns.