Improved Dung Beetle Optimizer Algorithm With Multi-Strategy for Global Optimization and UAV 3D Path Planning

In high-dimensional scenarios, path planning is a challenging and computationally complex optimization task that requires finding optimal paths within complex domains. Metaheuristic (MH) algorithms offer a practical approach to addressing this issue. The Dung Beetle Optimizer (DBO), categorized as a MH algorithm, takes inspiration from the biological behaviors exhibited by dung beetles. However, DBO exhibits limitations, including inadequate global search capabilities and a tendency to converge on local optima. To address these challenges, this paper proposes a multi-strategy Improved Dung Beetle Optimization algorithm (IDBO) for UAV 3D path planning. Initially, cubic chaos mapping is applied for population initialization, enhancing diversity. Subsequently, a novel global exploration strategy replaces the DBO's original rolling phase, improving information exchange and minimizing parameter dependence. Third, an adaptive t-distribution is introduced to adjust dung beetle positions, balancing exploration and exploitation. Finally, an enhanced population update strategy is proposed, utilizing varied behavioral logic at different algorithm stages to improve solution quality and search efficiency. Additionally, performance comparisons with six advanced algorithms on the CEC2017 test suite, and the validation of IDBO's effectiveness via the Wilcoxon rank-sum and Friedman mean rank test. Meanwhile, in UAV 3D path planning experiment, IDBO achieves the best cost index, which is 1.34% higher than the best cost of original DBO, and is also significantly better than the most advanced algorithms such as WOA, GSA, HHO, COA, and the standard deviation is reduced by 99.93% compared with DBO algorithm, which proves the effectiveness and robustness of IDBO in UAV 3D path planning.