Abstract: A multi-strategy improved pelican optimization algorithm was proposed to address the issues of easily falling into local optima and slow convergence speed in mobile robot path planning. Firstly, the pelican population was initialized using Logistic chaotic mapping to enhance population diversity and distribution uniformity, thereby improving global search capability. Secondly, the sine optimization algorithm and a nonlinear inertia weight coefficient were incorporated to increase search diversity and convergence speed. Then, the Lévy flight strategy was introduced to strengthen the ability to escape local optima and maintain sufficient global search performance in the later iterations. Simulation results demonstrate that, based on six benchmark test functions, the proposed algorithm significantly improves both global search capability and convergence accuracy. In path planning on 20×20 and 40×40 grid maps, the average path length is shortened by more than 7%−10% compared with the original pelican optimization algorithm, sparrow search algorithm, and grey wolf optimizer, while the computation time is also reduced. These results verify that the proposed algorithm possesses superior path planning performance and robustness in complex environments.