| With the approaching time of "Made in China 2025",the application of robots is becoming more and more extensive,and robots can be seen everywhere in production and life.In the production,the environment around the robotic arm is not only complex and diverse,but also faces increasing work efficiency and precision requirements.In view of this,this paper firstly studies the obstacle avoidance path planning with the shortest path as the performance index,and then studies the trajectory planning with the time-impact optimal.The main research contents are as follows: In this paper,the kinematics analysis of the manipulator from the forward and inverse aspects is firstly solved by the geometric method and the analytical method.It lays a theoretical foundation for the following path planning and trajectory planning research.Considering the complexity of the working environment of the manipulator,various obstacles are often encountered,and the obstacle avoidance path planning algorithm is studied for this purpose.Select the bounding sphere method to envelop the obstacles,and select the cylinder method to simplify the manipulator.Taking the shortest path as the performance index and avoiding obstacles as the constraint condition,the particle swarm algorithm with adaptive inertia weight is used to solve the problem,and the numerical simulation is carried out.The simulation results show that the improved PSO algorithm can effectively help the manipulator find the optimal path.Considering the working efficiency and service life of the manipulator,the trajectory planning of the manipulator is studied with time-impact as the performance index.The "4-5-4" piecewise polynomial interpolation is used for trajectory planning in the joint space,so as to ensure that the manipulator meets the requirements of smooth position and continuous speed and acceleration.The Q-learning particle swarm algorithm is used to solve the time-shock optimal trajectory of the manipulator under the constraints of position,velocity,acceleration and jerk.The algorithm regards particles as agents,and uses the update strategy of the Q-learning algorithm to help the PSO algorithm to adjust the parameters,thus effectively combining the advantages of the two algorithms.Trajectory optimization by Q-learning particle swarm algorithm can avoid the limitation of convexity of the solution problem.Finally,taking the PUMA560 manipulator as the simulation object,the Q-learning particle swarm algorithm is used for simulation analysis.The simulation results show that the algorithm can effectively help the manipulator to achieve comprehensive optimal performance,it can save maintenance costs while improving industrial production efficiency. |
