Multi-Strategy Whale Optimization for Open Complete-Coverage Path Planning of Aircraft-Skin Wall-Climbing Robots

Aircraft-skin maintenance requires complete surface-cell visitation, admissible cross-component transfers and local exclusion bypasses. This study develops an offline open complete-coverage planner on the NASA Common Research Model using a multi-strategy whale optimization algorithm (MSWOA). The method integrates aircraft-surface continuity preference with transition consistency (ASCP-TC) for topology-aware seeding, permutation-aware rank rejuvenation (PARR) for discrete route updates, and elite local refinement (ELR) with exact-delta evaluation. Across three maintenance transfer-cost settings on the 116-cell benchmark, MSWOA achieves the lowest mean values in all 18 setting–metric combinations when compared with six other population methods, deterministic nearest-neighbor construction (NN), and NN with local search (NN + LS). Mean task energy is 12.55–13.32% lower than WOA and 8.44–8.97% lower than converged NN + LS. Sequential ablation shows additional improvements at each of the three mechanism additions. Independent runs with 12,000 function evaluations retain the lowest MSWOA mean energy among the seven population methods. Parameter sensitivity analyses and 60–232-cell CAD configurations further support its performance across the tested conditions. These results demonstrate an effective and computationally efficient approach to energy-aware offline aircraft-skin coverage planning.