Multi-strategy secretary bird optimization algorithm for UAV path planning in complex environment.

This paper proposes a UAV path planning method based on a Multi-strategy Secretary Bird Optimization Algorithm (MSBOA) to address the challenges of navigating complex terrain. First, a pooling mechanism is introduced to enhance population diversity and improve the algorithm's optimization capabilities, balancing global exploration and local exploitation. Second, a dynamic fitness distance balance technique is incorporated to balance exploration and exploitation, preventing the population from becoming trapped in local optima while improving convergence accuracy. Finally, a greedy selection-based centroid reverse learning approach is used to update the population, enhancing the algorithm's exploratory performance. To validate the effectiveness of the proposed improved algorithm, the proposed MSBOA is compared with classical and advanced intelligent algorithms by solving the CEC2017 benchmark test functions and a designed UAV environment model. Comparative analysis of simulation results indicates that the proposed MSBOA converges faster and achieves higher accuracy than the traditional SBOA. It effectively handles complex UAV path planning problems, enabling the design of faster, shorter and safer flight paths. This further demonstrates the excellent performance of the multi-strategy SBOA in UAV path planning, highlighting its broad application prospects.