The time-impact optimal trajectory planning for multi-joint robot based on GA-PSO

In order to simplify robot control and reduce the programming difficulty of impact-optimized trajectories, this paper studies the application of gesture recognition in industrial robot trajectory planning. Meanwhile for reducing the vibration and mechanical wear caused by excessive robot impact, it is proposed to use an improved GA-PSO algorithm for time-impact optimal trajectory planning, that is, to use adaptive inertia weight and crossover mutation probability. The improved algorithm has the characteristics of fast convergence in the early stage of iteration and not easy to fall into local optimal solution in the later stage of iteration. In the trajectory planning, a 5-5-5 interpolation polynomial is used to ensure the continuity of the trajectory planning speed, acceleration, and impact. Under the control of the set fitness function, the search of the improved GA-PSO algorithm is used to simulate the results. The experimental results show that the optimized trajectory significantly reduces the impact simultaneously.