MSRIME: a multi-strategy RIME optimization algorithm for three-dimensional UAV path planning

Three-dimensional (3D) unmanned aerial vehicle (UAV) path planning is critical for ensuring efficient mission execution. However, existing algorithms commonly suffer from issues such as slow convergence, premature convergence to local optima, and an imbalance between global exploration and local exploitation. To overcome these limitations, this paper proposes a multi-strategy RIME algorithm (MSRIME) for 3D UAV path planning. First, a hybrid initialization strategy that combines chaotic mapping with latin hypercube sampling is employed to enhance the diversity of the initial population, thereby improving solution quality during the early search phase. Second, the soft-rime search mechanism is refined by incorporating Lévy flight and a hierarchical, population-based position update strategy, which jointly balance exploration and exploitation and accelerate convergence. Third, a periodic perturbation mechanism, constructed by integrating a polynomial mutation operator with sine-cosine functions, is introduced to increase search randomness and expand solution space coverage. Experimental results demonstrate that the proposed MSRIME algorithm exhibits superior global optimization capability, robustness, and adaptability compared to several state-of-the-art benchmark algorithms.