TIME-OPTIMAL TRAJECTORY GENERATION FOR INDUSTRIAL ROBOTS BASED ON ELITE MUTATION SPARROW SEARCH ALGORITHM

To improve the efficiency and stability of industrial robots, a time-optimal trajectory planning method based on an elite mutation sparrow search algorithm (EMSSA) is proposed. First, a non-uniform septic B-spline interpolation trajectory function is constructed, which overcomes the shortcoming of unsmooth joint acceleration or jerk in low-order interpolation and assigns kinematic parameters at the starting and stopping points. Second, the fitness function is constructed. It minimizes the sum of time intervals between two adjacent knots in B-spline trajectory considering kinematic constraints. An EMSSA is proposed to schedule the time intervals and generate the time-optimal septic B-spline trajectory. Elite reverse learning strategy is used to optimize the initial population and accelerate the convergence speed of the algorithm. Besides, to enhance the solution quality and avoid falling into local optimization, the algorithm is improved by cosine-descending search step and normal-Cauchy mutation strategies. Furthermore, an example is given to verify that the proposed algorithm is effective in solving the time-optimal trajectory planning problem with multi-constraint and has the advantages of fast solving speed, high precision, and good robustness. © 2023 Acta Press. All rights reserved.