Grey wolf optimization enhanced collaborative path planning for UUV swarms

Abstract

Path planning is critical for enabling efficient and collaborative operations in unmanned underwater vehicle (UUV) swarms, particularly in complex and dynamic underwater environments. This study proposes a novel three-dimensional (3D) collaborative path planning framework for UUV swarms, grounded in an advanced grey wolf optimization (GWO) algorithm. The framework integrates a comprehensive 3D model that incorporates kinematic constraints, threat avoidance, swarm collaboration, and path smoothness, tailored specifically for the complex dynamics of UUV swarms. By combining the global search mechanism of Cuckoo Search (CS) and the local refinement mechanism of a multi-population (MP) strategy, the proposed method achieves a robust balance between exploration and exploitation. Additionally, a nonlinear search strategy dynamically adjusts the convergence factor, further enhancing performance in complex 3D environments. Extensive experiments demonstrate that the proposed algorithm significantly improves convergence speed, solution quality, task completion time, and travel distance, highlighting its efficiency and practical applicability in UUV swarm applications.