Research On Collision Detection And Handling Of Industrial Robots

With the proposal of "Industry 4.0" and "Made in China 2025",more and more attention will be paid to the development of industry in the future,and industrial robots have become the priority development area of technological innovation.Robots are gradually replacing human beings with their advantages.However,due to the complexity of some tasks,robots cannot completely replace human.Therefore,the concept of human-computer cooperation is proposed.In the process of man-machine cooperation,collisions often occur.Robot safety has become an urgent problem to be solved.This paper mainly studies the collision detection of robots,and puts forward different collision handling methods according to different collision situations to ensure the safety of people and robots.In this paper,firstly,the D-H parameter method is used to model the kinematics of the robot body,and Lagrange method is used to model the dynamics of the robot body.In addition,considering that the robot will have different loads in its work,the six-axis force sensor is used to measure the mechanical parameters of the robot's load and to carry out dynamic modeling.Secondly,collision detection of the robot body is realized based on the generalized momentum theory of the robot.Combined with the dynamic model of the robot end load,a collision detection algorithm with the robot load is proposed.The simulation experiment of robot collision detection is carried out by Matlab,and the feasibility of the collision detection algorithm was verified.Then,the moment of each joint of the robot obtained by the robot collision detection algorithm is analyzed to identify which link on the robot has collided and the direction and magnitude of the collision force.The collision force identification theory in this paper is verified by changing the direction and magnitude of the collision force in the Matlab simulation experiment.Then the robot collision situations are classified according to the identified collision information,and corresponding collision response processing methods are proposed for different collision situations.Finally,using the existing STEP robot in the laboratory,model SD700 E,a robot collision detection test bed is built to carry out collision detection experiments to verify the feasibility of the collision detection algorithm proposed in this paper in reality.In addition,collisions of different directions and sizes are applied to the working robot,which verifies the feasibility of the robot collision handling method proposed in this paper.