Robot path planning based on multi-strategy improved artificial rabbits optimization

To address the imbalance between global exploration and local exploitation as well as stagnation in late-stage optimization during robot path planning, a multi-strategy improved artificial rabbits optimization (IARO) is proposed. A stochastic centroid dynamic opposition-based learning strategy is incorporated to enhance the diversity of the initial population, thereby improving the quality of initial solutions. During the detour foraging phase, a Lévy flight strategy is introduced to strengthen the algorithm’s exploration capability, effectively mitigating local optima traps and premature convergence. Furthermore, the energy factor is dynamically adjusted using a hyperbolic cosine-sine function to balance the exploitation and exploration phases, facilitating more effective information exchange within the population. Ablation experiments validate that IARO demonstrates superior stability and convergence accuracy. When applied to three mobile robot path planning models with varying complexities, IARO exhibits significantly enhanced pathfinding efficiency and robustness.