Research On Obstacle Avoidance Path Planning Of 6-DOF Manipulator Based On Improved RRT Algorithm

With the continuous advancement of science and technology,and the proposal of "Made in China 2025" plan,manipulator,as an essential division of robotics,have been broadly applied in various fields of social production,and its working environment has gradually become more complex with the continuous expansion of applications.Therefore,it is more and more important to quickly plan a good path that does not collide with the surrounding obstacles in the working space of the manipulator.To this end,this paper conducts a study on obstacle avoidance path planning with PUMA560,a 6-DOF manipulator.The main research contents are as follows:(1)Kinematic modeling and analysis of the manipulator.The kinematics model of PUMA560 is established according to the standard D-H parameter method,and on this basis,the forward and inverse kinematics equations are derived by analytical method.Aiming at the problem that traditional inverse kinematics solving methods are slow and prone to multiple solutions,an inverse kinematics solving model using Particle Swarm Optimization(PSO)to optimize the number of hidden layer neural units and learning rate in the Long Short-Term Memory(LSTM)network is proposed,and the performance of model is verified by simulation experiments.(2)Design of collision detection of manipulator and obstacles.The advantages and disadvantages of each of several common hierarchical envelope box collision detection methods are briefly analyzed and pointed out.Combined with the experimental scenario in which the obstacles in this study are static obstacles whose specific information has been calibrated in advance and the position of the robot arm base is fixed,the collision detection design is carried out by using the cylindrical envelope method and the enclosing sphere method to envelop the robot arm linkage and the obstacles in its workspace as cylinders and spheres,respectively.(3)Improving the Rapidly-exploring Random Tree(RRT)algorithm.A multi-strategy fusion improvement of the traditional RRT algorithm is carried out for the factors affecting its path planning efficiency.A heuristic target bias strategy and a gravitational potential field are used to expand the random tree in the direction of the target point,and an adaptive step size strategy is proposed to accelerate the expansion speed and optimize the planned paths.Simulation and comparison experiments between the improved algorithm and related algorithms are conducted in two-dimensional space and three-dimensional space respectively to verify the effect of the improved algorithm on path planning.(4)Simulation experiments of obstacle avoidance path planning for manipulator.Based on the improved algorithm in this paper and the related algorithm,the simulation and comparison experiments of obstacle avoidance path planning for several groups of PUMA560 manipulator are carried out in obstacle Spaces of different complexity,and the process indexes are analyzed to verify the effect of the improved algorithm in the field of obstacle avoidance path planning for manipulator.