Research on Adaptive Particle Swarm Optimization Algorithm for Manipulator Trajectory Planning

This is the authors' abstract. We don't add key points for this paper.

In addressing the efficiency and stability challenges associated with manipulator operations, this paper proposes a time-optimal trajectory planning method utilizing an Adaptive Particle Swarm Optimization algorithm. The approach is grounded in a 3-5-3 composite piecewise polynomial interpolation function, with the optimization target set as the time within each segment. The improved Adaptive Particle Swarm Optimization algorithm is employed for the optimization process. The algorithm initializes the population through an optimal point set, introducing dynamic factors and mutation strategies to enhance optimization capabilities. The study employs a six-degree-of-freedom manipulator model, comparing the proposed method with the standard Particle Swarm Optimization approach. The results demonstrate significant improvements in convergence speed and global search capabilities. Importantly, the algorithm avoids local convergence, ultimately reducing the optimization time by 15.58%. The generated trajectories, satisfying the kinematic constraints of the manipulator, exhibit smooth joint angle, velocity and acceleration curves, ensuring stable manipulator motion. The proposed method demonstrates feasibility in addressing efficiency and stability concerns in manipulator operations.