SFSAE: An Improved Alpha Evolution Algorithm for Cooperative Multi-UAV Path Planning in Complex 3D Mountainous Environments
Multi-unmanned aerial vehicle (UAV) path planning in complex three-dimensional mountainous environments poses significant challenges. Existing methods suffer from difficulties in maintaining population diversity, susceptibility to local optima, and imbalance between exploration and exploitation. To solve these problems, a multi-constraint comprehensive cost model is constructed that simultaneously considers path length, obstacle threat, relative flight altitude, path smoothness, and inter-UAV safety distance. Moreover, an adaptive feedback search alpha evolutionary algorithm (SFSAE) is proposed to effectively handle multi-UAV path planning in complex mountainous environments. First, a randomly shifted Sobol low-discrepancy sequence is introduced to improve population quality by increasing the spatial coverage uniformity of the initial population. Second, a diversity-aware adaptive spiral escape mechanism is designed to dynamically enhance the capability of escaping from local optima according to the population aggregation state. Finally, a stage-adaptive sine-cosine trajectory correction mechanism is introduced, in which non-monotonic weights are adopted to enlarge the mid-stage search span and improve the exploration-exploitation balance of the algorithm. The effectiveness of the proposed method is validated on benchmark functions and multiple representative scenarios. Compared with five representative metaheuristic optimization algorithms, SFSAE achieves a Friedman average rank of 2.065 with the first overall rank; the average path cost is reduced by 15.33–26.21%. The results show that SFSAE has favorable optimization accuracy, stability, and adaptability for constrained path planning in complex environments.
Authors
- Quanbo Ge (ORCID: https://orcid.org/0000-0002-6907-7837)
- Jiao Han (ORCID: https://orcid.org/0000-0001-9305-3918)
- Erdong Yuan (ORCID: https://orcid.org/0000-0001-8789-4470)
- Xiaoming Tao (ORCID: https://orcid.org/0000-0002-8763-9338)
- Junlei Chen (ORCID: https://orcid.org/0009-0005-8163-5247)
Institutions
- Zhejiang Normal University (CN)
- Xinjiang University (CN)
- Tsinghua University (CN)
Publication Details
- Journal
- Algorithms
- Published
- 2026-11-16
- DOI
- https://doi.org/10.3390/a19111008
- Primary Topic
- Metaheuristic Optimization Algorithms Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00