基于改进蜣螂算法的车机协同巡检路径规划方法
Vehicle-Drone Collaborative Inspection Path Planning Method Based on Improved Dung Beetle Algorithm

针对电网巡检任务中巡检车与无人机协同作业的问题，本文提出了一种基于改进蜣螂算法的路径规划方法，并构建了巡检车与无人机动态协同作业的路径规划模型。为提升蜣螂算法的搜索效率和收敛性能，本文做出了如下改进：首先，采用Logistic混沌映射优化初始解的分布，从而提高搜索的多样性；其次，在蜣螂滚动阶段引入鱼鹰算法的全局搜索策略，增强算法的全局搜索能力，提升收敛速度和稳定性；最后，在蜣螂觅食阶段引入自适应t分布策略，增强局部寻优能力，避免陷入局部最优解。实验结果表明，改进后的蜣螂算法在CEC2021标准函数集测试中，相较于基础蜣螂算法和其他对比算法，表现出更好的求解精度和收敛速度。在实际路径规划算例中，提出的方法在巡检车车速降低至30%、50%的情况下仍能有效求解，并在正常场景下节省了6.38分钟的计算时间，表明该方法具有较好的实用性和适应性。
To address the collaborative operation problem between inspection vehicles and drones in power grid inspection tasks, this paper presents a path planning method based on an improved dung beetle algorithm and establishes a dynamic collaborative operation model for the path planning of both inspection vehicles and drones. In order to enhance the search efficiency and convergence performance of the dung beetle algorithm, several improvements are introduced: first, the Logistic chaotic mapping is employed to optimize the distribution of initial solutions, thereby increasing search diversity; second, a global search strategy based on the Osprey algorithm is incorporated into the rolling phase of the dung beetle algorithm, enhancing its global search capability and improving both convergence speed and stability; third, an adaptive t-distribution strategy is introduced in the foraging phase to bolster local optimization ability and prevent premature convergence to local optima. Experimental results demonstrate that the improved dung beetle algorithm exhibits superior solution accuracy and convergence speed compared to the basic dung beetle algorithm and several other benchmark algorithms in tests using the CEC2021 standard function set. In practical path planning case studies, the proposed method effectively solves the problem even when the inspection vehicle’s speed is reduced to 30% and 50% and reduces the computation time by 6.38 minutes in normal scenarios, thus demonstrating its practicality and adaptability.