A Hybrid Spiral-Biased Particle Swarm Optimization with Black-winged Kite Algorithm for Mobile Robot Path Planning

Abstract

To enhance the global search and local optimization capabilities of swarm intelligence algorithms in robot path planning, this paper proposes an improved particle swarm optimization algorithm (HSB-PSO) that incorporates a dynamic strategy. This algorithm features three key innovations: First, a dynamic adaptive spiral strategy is introduced to adaptively guide particles toward the optimal solution by adjusting spiral parameters at different iteration stages, enhancing the particles’ global exploration capability and convergence accuracy. Second, a probability-decayed black kite behavior mechanism is designed to simulate the perturbation behavior of black kites during predation, and a probabilistic control factor is introduced to dynamically adjust its influence, effectively improving search diversity and avoiding local optima. Finally, an elite-guided on-demand reverse learning strategy is combined to selectively perform reverse learning on elite particles based on the current state of the swarm, further enhancing local search and convergence speed. Simulation experiments demonstrate that the HSB-PSO algorithm demonstrates superior optimization capability and path quality in multiple typical path planning test scenarios, validating the effectiveness and practical value of the proposed strategy.