A multi-mechanism balanced advanced learning sparrow search algorithm for UAV path planning

Abstract

Unmanned aerial vehicle (UAV) is highly flexible and versatile, ranging from monitoring and surveying to rescue and military applications, but finding the best path requires a large amount of computing resources. Through intelligent path planning algorithms, UAV can find the best path according to task requirements and environmental conditions, avoid obstacles, and bypass dangerous areas, thereby effectively reducing the risks and errors of task execution. However, ordinary heuristic algorithms often do not achieve satisfactory results. To address this problem, A multi-mechanism balanced advanced sparrow search algorithm (BALSSA) is proposed. To achieve a balance between exploration and exploitation in the algorithm, we introduce two innovative techniques: an adaptive weight jumping mechanism and a suicide mutation perturbation. Then a balanced advanced learning is proposed. this approach enhances the evolutionary learning capabilities of SSA and aids the algorithm in timely escaping from local optima, and then propose a spiral factor improved progressive learning to improve the mutual learning performance between individuals, tested on 23 benchmark test functions, CEC2017 and CEC2022 The set is compared with algorithms in recent years and proposed algorithm variants, and the results show that BALSSA has better optimization and robustness. Finally, the proposed BALSSA is applied to UAV path planning. Compared with the variant SSA in recent years, BALSSA shows more stable and accurate optimization performance to verify the practicability of the improved algorithm.