基于LSTM和启发式方法的遥感卫星地面站天线智能调度

doi:10.7523/j.ucas.2020.0014

摘要/Abstract

摘要： 遥感卫星地面站天线调度是解决遥感卫星数据接收天线资源不足和提高资源使用效率的有效途径。由于天线调度规则复杂,提出一种长短期记忆神经网络和启发式搜索相结合的智能调度方法。首先,使用长短期记忆神经网络模型从历史调度数据中提取天线使用规则,并使用该规则为遥感卫星数据接收任务分配接收天线,得到初始调度方案;其次,使用启发式方法,对初始方案中数据联合接收和资源选择冲突两个问题加以修正,得到实际可行的调度方案。结果表明:本方法与结合启发式规则的遗传算法相比在资源利用率和计算效率上均有提升,证明了本方法的有效性。

关键词: 遥感卫星, 地面站天线调度, 长短期记忆神经网络, 启发式搜索, 智能调度

Abstract: In order to solve the shortage of remote sensing satellite data receiving antennas and improve the utilization of the ground antennas, an intelligent scheduling method which combines LSTM (long short-term memory network) and heuristic search was proposed. First, LSTM is used to extract the antenna using rules from the historical scheduling data of antennas, and then the initial scheduling scheme is obtained by allocating an antenna for each remote sensing data receiving task with the rules; Second, the heuristic search is used to solve the two problems of joint data reception and resource selection conflict in the initial plan, and obtain a practical and feasible scheduling plan. The experiment results show that the method is useful to deal with ground antenna scheduling, improve resource efficiency and reduce computing time to some extent when compared with genetic algorithm.

Key words: remote sensing satellite, ground station antennas scheduling, long short-term memory (LSTM) neural network, heuristic search, intelligent scheduling

中图分类号:

O221.7
V556

孙文军, 马广彬, 田妙苗, 林友明, 黄鹏. 基于LSTM和启发式方法的遥感卫星地面站天线智能调度[J]. 中国科学院大学学报, 2022, 39(4): 532-542.

SUN Wenjun, MA Guangbin, TIAN Miaomiao, LIN Youming, HUANG Peng. Remote sensing satellite ground station antenna intelligent scheduling with LSTM and heuristic search[J]. Journal of University of Chinese Academy of Sciences, 2022, 39(4): 532-542.

参考文献

[1] 金光,武小悦,高卫斌.卫星地面站资源调度优化模型及启发式算法[J].系统工程与电子技术,2004,26(12):1839-1841,1875. DOI:10.3321/j.issn:1001-506X.2014.12.026.
[2] 金光,武小悦,高卫斌.基于冲突的卫星地面站系统资源调度与能力分析[J].小型微型计算机系统,2007,28(2):310-312. DOI:10.3969/j.issn.1000-1220.2007.02.026.
[3] 王远振,高卫斌,聂成.多星地面站系统资源配置优化研究综述[J].系统工程与电子技术,2004,26(4):437-439,453. DOI:10.3321/j.issn:1001-506X.2004.04.005.
[4] Gooley T D. Automating the satellite range scheduling process.Ohio:Air Force Institute of Technology, 1993.
[5] Gooley T D, Borsi J J, Moore J T. Automating air force satellite control network (AFSCN) scheduling[J]. Mathematical and Computer Modelling, 1996, 24(2):91-101. DOI:10.1016/0895-7177(96)00093-3.
[6] Schalck S M. Automating satellite range scheduling.Ohio:Air Force Institute of Technology, 1993.
[7] 贺仁杰.成像侦察卫星调度问题研究.长沙:国防科学技术大学,2004.
[8] 刘洋,陈英武,谭跃进.卫星地面站系统任务调度的动态规划方法[J].中国空间科学技术,2005,25(1):44-47. DOI:10.3321/j.issn:1000-758X.2005.01.008.
[9] Pemberton J C, Galiber F. A constraint-based approach to satellite scheduling[J]. DIMACS Series in Discrete Mathematics and Theoretical Computer Science, 2001, 57:101-114.
[10] Pemberton J C. Towards scheduling over-constrained remote sensing satellites//Proceedings of the 2d International Workshop on Planning and Scheduling for Space. San Francisco, 2000:84-89.
[11] Zhang J W, Xing L N, Peng G S, et al. A large-scale multi-objective satellite data transmission scheduling algorithm based on SVM+NSGA-Ⅱ[J]. Swarm and Evolutionary Computation, 2019, 50:100560. DOI:10.1016/j.swevo.2019.100560.
[12] 王远振,赵坚,聂成.多卫星-地面站系统的Petri网模型研究[J].空军工程大学学报(自然科学版),2003,4(2):7-11. DOI:10.3969/j.issn.1009-3516.2003.02.002.
[13] 王远振,赵坚,聂成.多星地面站设备优化调度方法研究[J].计算机仿真,2003,20(7):17-19,54. DOI:10.3969/j.issn.1006-9348.2003.07.006.
[14] 王远振,赵坚,聂成.任务优先级调度策略性能分析[J].空军工程大学学报(自然科学版),2003,4(3):31-35. DOI:10.3969/j.issn.1009-3516.2003.03.008.
[15] 王远振,赵坚,聂成,等.卫星地面站设备配置效能评价指标体系研究[J].军事运筹与系统工程,2003,17(1):59-64. DOI:10.3969/j.issn.1672-8211.2003.01.015.
[16] 张帆,王钧,李军,等.基于时间序无圈有向图的多准则优化成像调度[J].国防科技大学学报,2005,27(6):61-66. DOI:10.3969/j.issn.1001-2486.2005.06.014.
[17] 陈慧中,王钧,李军,等.卫星成像规划调度系统中的可视化决策支持研究与实现[J].计算机工程与科学,2007,29(7):58-61. DOI:10.3969/j.issn.1007-130X.2007.07.018.
[18] 李云峰,武小悦.基于综合优先度的卫星数传调度算法[J].系统工程学报,2007,22(6):644-648. DOI:10.3969/j.issn.1000-5781.2007.06.014.
[19] 李云峰,武小悦.基于试探性的卫星数传任务调度算法研究[J].系统工程与电子技术,2007,29(5):764-767. DOI:10.3321/j.issn:1001-506X.2007.05.025.
[20] Burrowbridge S E. Optimal allocation of satellite network resources. Blacksburg, VA, USA:Virginia Polytechnic Institute and State University, 1999.
[21] 李云峰,武小悦.遗传算法在卫星数传调度问题中的应用[J].系统工程理论与实践,2008,28(1):124-131. DOI:10.3321/j.issn:1000-6788.2008.01.018.
[22] Sun J, Xhafa F. A genetic algorithm for ground station scheduling//Proceedings of the 2011 International Conference on Complex, Intelligent, and Software Intensive Systems. June 30-July 2, 2011, Seoul, Korea (South). IEEE Computer Society, 2011:138-145. DOI:10.1109/CISIS.2011.29.
[23] 刘民.基于数据的生产过程调度方法研究综述[J].自动化学报,2009,35(6):785-806. DOI:10.3724/SP.J.1004.2009.00785.
[24] 吴启迪,乔非,李莉,等.基于数据的复杂制造过程调度[J].自动化学报,2009,35(6):807-813. DOI:10.3724/SP.J.1004.2009.00807.
[25] Koonce D A, Tsai S C. Using data mining to find patterns in genetic algorithm solutions to a job shop schedule[J]. Computers&Industrial Engineering, 2000, 38(3):361-374. DOI:10.1016/S0360-8352(00)00050-4.
[26] Youssef H, Brigitte M, Noureddine Z. A genetic algorithm and data mining to resolve a job shop schedule//Proceedings of the 8th IEEE International Conference on Emerging Technologies and Factory Automation. October 15-18, 2001, Antibes-Juan les Pins, France. IEEE, 2001:727-728. DOI:10.1109/ETFA.2001.997768.
[27] Harrath Y, Chebel-Morello B, Zerhouni N. A genetic algorithm and data mining based meta-heuristic for job shop scheduling problem//IEEE International Conference on Systems, Man and Cybernetics. October 6-9, 2002, Yasmine Hammamet, Tunisia. Tunisia:IEEE, 2002:280-285. DOI:10.1109/ICSMC.2002.1175709.
[28] Feng S, Li L, Cen L, et al. Using MLP networks to design a production scheduling system[J]. Computers&Operations Research, 2003, 30(6):821-832. DOI:10.1016/s0305-0548(02)00044-8.
[29] Weckman G R, Ganduri C V, Koonce D A. A neural network job-shop scheduler[J]. Journal of Intelligent Manufacturing, 2008, 19(2):191-201. DOI:10.1007/S10845-008-0073.9.
[30] Kumar S, Rao C S P. Application of ant colony, genetic algorithm and data mining-based techniques for scheduling[J]. Robotics and Computer-Integrated Manufacturing, 2009, 25(6):901-908. DOI:10.1016/j.rcim.2009.04.015.
[31] Shahzad A, Mebarki N. Data mining based job dispatching using hybrid simulation-optimization approach for shop scheduling problem[J]. Engineering Applications of Artificial Intelligence, 2012, 25(6):1173-1181. DOI:10.1016/j.engappai.2012.04.001.
[32] 王成龙,李诚,冯毅萍,等.作业车间调度规则的挖掘方法研究[J].浙江大学学报(工学版),2015,49(3):421-429,438.
[33] 王振江.作业车间调度属性选择及调度规则挖掘方法研究.北京:北京化工大学,2016.
[34] 丁建立,王曼.基于关联规则挖掘的航班协同保障数据知识发现研究[J].计算机应用与软件,2016,33(11):21-23,61. DOI:10.3969/j.issn.1000-386x.2016.11.005.
[35] Vinyals O, Fortunato M, Jaitly N. Pointer networks//Proceedings of the 28th International Conference on Neural Information Processing Systems-Vol 2. Cambridge, MA, USA:MIT Press, 2015:2692-2700.
[36] Bello I, Pham H, Le Q V, et al. Neural combinatorial optimization with reinforcement learning. arXiv:1611.09940(2017-01-12).. https://arxiv.org/abs/1611.09940.
[37] Sutskever I, Vinyals O, Le Q V. Sequence to sequence learning with neural networks//Advances in Neural Information Processing Systems, 2014:3104-3112.
[38] Lin K X, Zhao R Y, Xu Z, et al. Efficient large-scale fleet management via multi-agent deep reinforcement learning//The 24th ACM SIGKDD International Conference on Knowledge Discovery&Data Mining. London:ACM, 2018:1774-1783. DOI:10.1145/3219819.3219993.
[39] Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural computation, 1997, 9(8):1735-1780. DOI:10.1162/neco.1997.9.8.1735.
[40] Srivastava N, Hinton G E, Krizhevsky A, et al. Dropout:a simple way to prevent neural networks from overfitting[J]. Journal of Machine Learning Research, 2014, 15(1):1929-1958.