Integrating elite opposition-based learning and Cauchy–Gaussian mutation into sparrow search algorithm for time–impact collaborative trajectory optimization of robotic manipulators

Abstract

Aiming at the problems such as low convergence efficiency, local optimization traps, and insufficient multi-objective cooperative optimization existing in the multi-objective trajectory planning of industrial robotic arms, this study proposes a trajectory optimization method based on a new improved sparrow search algorithm (NISSA). Firstly, by integrating elite reverse learning and the Cauchy–Gaussian mutation strategy, the NISSA algorithm is constructed to enhance the global search ability and convergence efficiency. Secondly, the 3–5–3 polynomial interpolation method is adopted to establish a continuous and smooth joint spatial trajectory model to ensure the continuity of position, velocity, and acceleration. Finally, a multi-objective optimization function integrating time and mechanical shock is constructed, and the collaborative optimization of efficiency and stability is achieved through dynamic weight allocation. The simulation experiments based on the IRB4600 six-axis robotic arm show that compared with the traditional sparrow algorithm (SSA) and multi-strategy improved particle swarm optimization (MIPSO), NISSA shortens the trajectory planning time by 19.6 %, reduces path redundancy by 25.7 %, increases the iterative convergence speed by 68.75 %, and reduces the standard deviation of joint acceleration to 28.5 % of the original value. The research results provide theoretical support and technical implementation paths for the high-precision and efficient operation of robotic arms in complex industrial scenarios.