Research On Trajectory Planning Method Of Manipulator Based On Multi-objective Optimization

With the wide application of the manipulator,the trajectory planning of the manipulator becomes more and more important in the research of the motion control technology.This paper takes SR1-540 industrial robot arm as the research object,and takes reducing the running time,energy loss and impact of the robot arm as the goal of trajectory planning,and carries out trajectory planning on the obstacle avoidance path.Under the constraints of kinematics and dynamics,a multi-objective particle swarm optimization algorithm(NSDS-MOPSO)was proposed,which combined the natural selection mechanism of genetic algorithm and dynamic step size position updating,to carry out trajectory planning for the manipulator and solve the optimal trajectory.The main research contents of this paper are as follows:1)By analyzing the structure of the manipulator,MDH method was used to establish the mathematical model of the manipulator,analyze the inverse kinematics,and the forward and inverse kinematics simulation in MATLAB was carried out.Newton-Euler method was used to model the dynamics of the manipulator,and Adams was used to establish the dynamics simulation model,and the simulation and verification of the dynamics model were carried out.2)In order to improve the limitation of inverse kinematics by traditional methods,the differential evolution(TSDE)algorithm of mixed Tent and Sine chaotic initialization was adopted to solve the inverse kinematics of the manipulator.TSDE algorithm has faster convergence speed,better balance between global and local exploration capabilities,reduces the ability to fall into the local optimal solution,and has higher convergence accuracy and fewer iterations.In order to solve the obstacle avoidance path planning problem of SR1-540 manipulator,an obstacle avoidance model of the manipulator was established by using multi-sphere envelope method.The improved RRT algorithm is used to solve the path planning problem,and the results are optimized by path node and smoothed.The path curve obtained by this algorithm is smoother,the number of effective path points is less,the path cost is less,and it has better performance.3)NSDS-MOPSO algorithm was proposed based on quintic nonuniform rational Bspline curve(NURBS)to interpolate the joint trajectory.Under multiple constraints,the multi-objective comprehensive optimization of the robot trajectory is realized by the algorithm.The time performance index of the optimization results reached 44.90%,the energy consumption performance index reached 1.90%,and the impact performance index reached 68.57%.The trajectory of the manipulator is relatively optimal in terms of time,energy and impact.The optimal solution has good diversity and distribution uniformity.4)The experimental platform and simulation platform of SR1-540 manipulator were built to implement multi-objective trajectory optimization.The real-time speed and torque curves of the joint motor were obtained through the input of the time-positionvelocity sequence and the output of the simulation and experiment platform.By analyzing its characteristics,the velocity and acceleration curves of each joint are smooth and conform to the constraint range.The experimental data effectively verifies the feasibility of the optimal trajectory obtained by the algorithm.