面向集合任务的多无人机电力巡检任务分配方法研究

赵慧敏, 罗贺, 阴酉龙, 林世忠, 王国强

系统工程理论与实践 ›› 2025, Vol. 45 ›› Issue (2) : 666-684.

PDF(3024 KB)
中国科学院数学与系统科学研究院期刊网
PDF(3024 KB)
系统工程理论与实践 ›› 2025, Vol. 45 ›› Issue (2) : 666-684. DOI: 10.12011/SETP2023-2060
论文

面向集合任务的多无人机电力巡检任务分配方法研究

    赵慧敏1,4, 罗贺1,4,5, 阴酉龙2, 林世忠3, 王国强1,4,5
作者信息 +

Research on task allocation method for multiple drones power inspection for collective tasks

    ZHAO Huimin1,4, LUO He1,4,5, YIN Youlong2, LIN Shizhong3, WANG Guoqiang1,4,5
Author information +
文章历史 +

摘要

在无人机电力巡检过程中, 一个待巡检部件通常对应多个位置不同且均符合拍摄要求的任务点, 这些拍摄任务点构成一个集合任务. 为了保证巡检质量, 通常要求无人机多次采集待巡检部件的图片, 即访问集合任务中的多个拍摄任务点. 本文针对上述特点, 将面向集合任务的多无人机电力巡检任务分配问题建模为最小最大化多站点家庭旅行商问题(minmax multi-depot family traveling salesman problem, Minmax-MDFTSP), 并设计了一种强化遗传算法框架. 在该框架下, 提出了染色体校验及修正机制、 组合交换变异算子、 基于贪婪策略的局部调优算子, 设计了基于强化学习的遗传算法参数调优方法. 性能实验结果表明, 本文方法在求解质量和求解效率方面均具有明显优势. 此外, 通过消融实验验证了强化遗传算法框架中各个部分的有效性. 最后结合实际场景下的具体案例, 通过实地飞行验证了本文方法相对于现有巡检方式的优势.

Abstract

During the process of using drones for power inspection, a component to be inspected usually corresponds to multiple task points that are different in location but all meet the photography requirements. These task points form a collective task. To ensure the quality of inspection, it is required for the drone to take multiple shots of the components to be inspected, that is, to visit multiple task points in the collective task. In light of the above characteristics, the problem of task allocation for multiple drones power inspection for collective tasks was modelled as minmax multi-depot family traveling salesman problem (Minmax-MDFTSP). A framework that combines reinforcement learning and genetic algorithm was designed to solve the problem. This framework contained a mechanism for checking and correcting chromosomes, a combination exchange mutation operator, a local optimization operator based on a greedy strategy and a parameter tuning method for genetic algorithm based on reinforcement learning. The results of the performance experiment showed that the proposed method in this paper exhibited significant improvements in both solution quality and solving efficiency. Besides, the ablation experiment confirmed the effectiveness of each part in the framework. Finally, in combination with real-world scenarios, the advantages of the proposed method over existing inspection methods were verified through on-site flight validation.

关键词

电力巡检 / 多无人机 / 任务分配 / 强化学习 / 遗传算法

Key words

power inspection / multiple drones / task allocation / reinforcement learning / genetic algorithm

引用本文

EndNote

Ris (Procite)

Bibtex

导出引用
赵慧敏 , 罗贺 , 阴酉龙 , 林世忠 , 王国强. 面向集合任务的多无人机电力巡检任务分配方法研究. 系统工程理论与实践, 2025, 45(2): 666-684 https://doi.org/10.12011/SETP2023-2060
ZHAO Huimin , LUO He , YIN Youlong , LIN Shizhong , WANG Guoqiang. Research on task allocation method for multiple drones power inspection for collective tasks. Systems Engineering - Theory & Practice, 2025, 45(2): 666-684 https://doi.org/10.12011/SETP2023-2060
中图分类号： V279    TM75    TP18   

参考文献

[1] Lv J X, Yan L J, Chu S C, et al. A new hybrid algorithm based on golden eagle optimizer and grey wolf optimizer for 3D path planning of multiple UAVs in power inspection[J]. Neural Computing & Applications, 2022, 34(14): 11911-11936.
[2] 赵蒙, 张博, 胡祥培, 等. 电力巡检车载无人机协同作业优化[J]. 系统管理学报, 2022, 31(6): 1098-1108. Zhao M, Zhang B, Hu X P, et al. Cooperative operation optimization for transmission inspection with vehicle-mounted UAV[J]. Journal of Systems & Management, 2022, 31(6): 1098-1108.
[3] 王宏伟, 刘亚东, 田兵, 等. 基于空间关系的变电站巡检机器人巡检点自主生成方法[J]. 高电压技术, 2022, 48(8): 2982-2990. Wang H W, Liu Y D, Tian B, et al. Autonomous generation method of substation inspection robot inspection point based on spatial relationship[J]. High Voltage Engineering, 2022, 48(8): 2982-2990.
[4] Luo Y H, Yu X, Yang D S, et al. A survey of intelligent transmission line inspection based on unmanned aerial vehicle[J]. Artificial Intelligence Review, 2023, 56(1): 173-201.
[5] Yang L, Fan J F, Liu Y H, et al. A review on state-of-the-art power line inspection techniques[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(12): 9350-9365.
[6] Fan B, Li Y, Zhang R, et al. Review on the technological development and application of UAV systems[J]. Chinese Journal of Electronics, 2020, 29(2): 199-207.
[7] Pan J S, Lv J X, Yan L J, et al. Golden eagle optimizer with double learning strategies for 3D path planning of UAV in power inspection[J]. Mathematics and Computers in simulation, 2022, 193: 509-532.
[8] Guan H C, Sun X L, Su Y J, et al. UAV-lidar aids automatic intelligent powerline inspection[J]. International Journal of Electrical Power & Energy Systems, 2021, 130: 106987.
[9] Liu C, Dong R, Jin L, et al. Viewpoint optimization method for flying robot inspecting transmission towers based on point cloud model[J]. Chinese Journal of Electronics, 2018, 27(5): 976-984.
[10] Xu C, Li Q, Zhou Q, et al. Power line-guided automatic electric transmission line inspection system[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 1-18.
[11] Bernardino R, Paias A. Solving the family traveling salesman problem[J]. European Journal of Operational Research, 2018, 267(2): 453-466.
[12] Moran-Mirabal L F, Gonzalez-Velarde J L, Resende M G C. Randomized heuristics for the family traveling salesperson problem[J]. International Transactions in Operational Research, 2014, 21(1): 41-57.
[13] Bernardino R, Gouveia L, Paias A, et al. The multi-depot family traveling salesman problem and clustered variants: Mathematical formulations and branch-&-cut based methods[J]. Networks, 2022, 80(4): 502-571.
[14] Pop P, Matei O, Pintea C. A two-level diploid genetic based algorithm for solving the family traveling salesman problem[C]// Genetic and Evolutionary Computation Conference, 2018: 340-346.
[15] Bernardino R, Paias A. Heuristic approaches for the family traveling salesman problem[J]. International Transactions in Operational Research, 2021, 28(1): 262-295.
[16] Phan H D, Ellis K, Barca J C, et al. A survey of dynamic parameter setting methods for nature-inspired swarm intelligence algorithms[J]. Neural Computing & Applications, 2020, 32(2): 567-588.
[17] Mou J, Gao K, Duan P, et al. A machine learning approach for energy-efficient intelligent transportation scheduling problem in a real-world dynamic circumstances[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(12): 15527-15539.
[18] Guo K, Yang M, Zhu H. Application research of improved genetic algorithm based on machine learning in production scheduling[J]. Neural Computing & Applications, 2020, 32(7): 1857-1868.
[19] Liang Z, Liu M, Zhong P, et al. Application research of a new neighbourhood structure with adaptive genetic algorithm for job shop scheduling problem[J]. International Journal of Production Research, 2023, 61(2): 362-381.
[20] Cui Y, Hu W, Rahmani A. A reinforcement learning based artificial bee colony algorithm with application in robot path planning[J]. Expert Systems with Applications, 2022, 203: 117389.
[21] Li T, Yin Y, Yang B, et al. A self-learning bee colony and genetic algorithm hybrid for cloud manufacturing services[J]. Computing, 2022, 104(9): 1977-2003.
[22] 张亚宁, 白思俊, 陈志, 等. 基于深度学习的RCPSP调度优先规则实时动态选择算法[J]. 系统工程理论与实践, 2023, 43(7): 2142-2153. Zhang Y N, Bai S J, Chen Z, et al. Real-time dynamic selection algorithm of RCPSP scheduling priority rules based on deep learning[J]. Systems Engineering—Theory & Practice, 2023, 43(7): 2142-2153.
[23] Shahrabi J, Adibi M A, Masoud M. A reinforcement learning approach to parameter estimation in dynamic job shop scheduling[J]. Computers & Industrial Engineering, 2017, 110: 75-82.
[24] Uzunoglu A, Gahm C, Wahl S, et al. Learning-augmented heuristics for scheduling parallel serial-batch processing machines[J]. Computers & Operations Research, 2023, 151: 106122.
[25] Chen Y, Hu J L, Hirasawa K, et al. Optimizing reserve size in genetic algorithms with reserve selection using reinforcement learning[C]// 46th SICE Annual Conference, 2007: 1341-1347.
[26] Meng S S, Zhu Q, Xia F, et al. Research on parameter optimization of dynamic priority scheduling algorithm based on improved reinforcement learning[J]. IET Generation Transmission & Distribution, 2020, 14 (16): 3171-3178.
[27] Li R, Gong W, Lu C. A reinforcement learning based RMOEA/D for bi-objective fuzzy flexible job shop scheduling[J]. Expert Systems with Applications, 2022, 203: 117380.
[28] Chen R H, Yang B, Li S, et al. A Self-learning genetic algorithm based on reinforcement learning for flexible job-shop scheduling problem[J]. Computers & Industrial Engineering, 2020, 149: 106778.
[29] Li R, Zhang X, Jiang L, et al. An adaptive heuristic algorithm based on reinforcement learning for ship scheduling optimization problem[J]. Ocean & Coastal Management, 2022, 230: 106375.
[30] Cao Z C, Lin C R, Zhou M C. A knowledge-based cuckoo search algorithm to schedule a flexible job shop with sequencing flexibility[J]. IEEE Transactions on Automation Science and Engineering, 2021, 18(1): 56-69.
[31] Gu Y, Chen M, Wang L. A self-learning discrete salp swarm algorithm based on deep reinforcement learning for dynamic job shop scheduling problem[J]. Applied Intelligence, 2023, 53: 18925-18958.
[32] Fallahi A, Bani E A, Niaki S T A. A constrained multi-item EOQ inventory model for reusable items: Reinforcement learning-based differential evolution and particle swarm optimization[J]. Expert Systems with Applications, 2022, 207: 118018.
[33] Han S, Fan C, Li X, et al. A modified genetic algorithm for task assignment of heterogeneous unmanned aerial vehicle system[J]. Measurement and Control, 2021, 54(5-6): 994-1014.
[34] Xiao P, Ju H, Li Q, et al. Task planning of space-robot clusters based on modified differential evolution algorithm[J]. Applied Sciences, 2020, 10(14): 5000.
[35] 颜瑞, 陈立双, 朱晓宁, 等. 考虑区域限制的卡车搭载无人机车辆路径问题研究[J]. 中国管理科学, 2022, 30(5): 144-155. Yan R, Chen L S, Zhu X N, et al. Research on vehicle routing problem with truck and drone considering regional restriction[J]. Chinese Journal of Management Science: 2022, 30(5): 144-155.

基金

国家自然科学基金面上项目(71971075,72271076,71871079);国家自然科学基金基础科学中心项目(72188101);国家重点研发计划项目(2019YFE0110300);安徽省博士后研究人员科研活动经费(2022B587)
PDF(3024 KB)

373

Accesses

0

Citation

Detail
段落导航
相关文章

/

〈 〉