多策略改进蜣螂优化算法无人机三维路径规划

Multi-Strategy Improved Dung Beetle Optimization Algorithm for UAV 3D Path Planning

摘要: 无人机三维路径规划具有高维优化特性，需要强大的全局搜索能力。为了解决传统DBO算法的过早收敛问题，提高局部搜索效率，本研究提出了一种多策略融合算法IDBO。该方法首先利用邻域拓扑机制来增强种群间的信息交流，保持多样性，并结合三重自适应权重、振荡增强的社会认知动态平衡机制。其次，引入PSO分阶段引导滚球机制，以达到更好的收敛精度。然后通过精英–差分协同觅食机制显著提高全局探索能力。通过CEC2017的29个基准函验证，IDBO表现出优越的收敛精度和鲁棒性。在UAV3D路径规划实验进一步证明了IDBO生成更平滑路径的能力，这些发现证实了IDBO通过协调多策略优化解决复杂场景下无人机路径规划挑战的有效性。

Abstract: UAV 3D path planning exhibits high-dimensional optimization characteristics, requiring robust global search capabilities. To address the premature convergence problem of traditional Direct Optimization (DBO) algorithms and improve local search efficiency, this study proposes a multi-strategy fusion algorithm, IDBO. This method first utilizes a neighborhood topology mechanism to enhance inter-population communication and maintain diversity, combined with a triple adaptive weighting and oscillatory enhancement social cognitive dynamic balance mechanism. Secondly, a phased guided rolling ball mechanism (PSO) is introduced to achieve better convergence accuracy. Then, an elite-differential cooperative foraging mechanism significantly improves global exploration capabilities. Validated by 29 benchmark functions at CEC2017, IDBO demonstrates superior convergence accuracy and robustness. Further experiments in UAV 3D path planning demonstrate IDBO’s ability to generate smoother paths. These findings confirm the effectiveness of IDBO in addressing the challenges of UAV path planning in complex scenarios through coordinated multi-strategy optimization.