基于指针网络的空间目标遍历交会序列规划

doi:10.7527/S1000-6893.2023.28698

Abstract

Abstract:

Traversal rendezvous mission planning of multiple space targets for a single spacecraft is a mixed-integer programming problem with high complexity， which involves the combinatorial optimization of the top-level rendezvous sequence and the continuous optimization of the base-level flight trajectories. Existing methods that integrally optimize all discrete and continuous variables are inefficient and difficult to achieve the optimum. We propose a learning-based method that can efficiently obtain the near-optimal sequence mainly using the pointer networks. First， the neural network model for multiple-space-target traversal rendezvous planning is constructed as the decision agent of sequencing. Second， an unsupervised learning method based on the asynchronous advantage actor-critic algorithm is proposed to avoid the expensive computational cost in obtaining training labels. Finally， an estimation method to rapidly approximate the actual transfer cost is embedded in the training process to improve the efficiency of calculating rewards. Case studies show that the proposed training method performs efficiently， and the well-trained agent can rapidly predict the optimal sequence with a probability more than 88.7%.

Key words: aerospace mission planning, rendezvous sequence planning, moving target traveling salesman problem, combinatorial optimization, pointer network, reinforcement learning

CLC Number:

V448

Jiacheng ZHANG, Yuehe ZHU, Yazhong LUO. Space target rendezvous sequence planning via pointer networks[J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(15): 528698-528698.

Figures/Tables 13

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Table 1

Value ranges of dynamics attributes of targets

算例	参数类型	最小值	最大值
算例1	初始位置横向分量x₀/m	0	100
	初始位置纵向分量y₀/m	0	100
	移动速度横向分量v_x_，0/（m·s^-1）	1.5	5
	移动速度纵向分量v_y_，0/（m·s^-1）	1.5	5
	2目标间转移时长 $Δ t$ /s	0	10
算例2	半长轴a/AU	2.4	3
	偏心率e	0.1	0.2
	轨道倾角I₀/（°）	5	15
	升交点赤经Ω₀/（°）	0	360
	近地点幅角ω₀/（°）	0	360
	真近点角θ₀/（°）	0	360
	2目标间转移时长 $Δ t$ /d	300	1 200
算例3	半长轴a/km	6 990	7 280
	偏心率e	0	0.02
	轨道倾角I₀/（°）	80	85
	升交点赤经Ω₀/（°）	0	360
	近地点幅角ω₀/（°）	0	360
	真近点角θ₀/（°）	0	360
	两目标间转移时长 $Δ t$ /d	30	180

Table 1

Table 2

Table 3

Parameters of ant colony optimization

参数	数值	参数	数值
蚂蚁数量N	100	启发式权重 $β$	2
最大迭代次数G	500	全局信息素挥发因子 $ρ$	0.1
信息素常量Q	20	局部信息素挥发因子 $γ$	0.1
信息素因子 $α$	2	解构造选择概率 $q 0$	0.1

Table 3

Table 4

Fig.7

Fig.8

Table 5

Multi-spacecraft rendezvous sequences obtained by sequencer-assisted optimization

航天器序号	交会目标顺序	$Δ v$ /（km·s^-1）
1	2， 78， 3， 19， 14， 37， 18， 41， 62， 80， 106， 13， 121， 0， 107， 10， 9， 34， 63， 25， 27， 61， 101， 94， 90， 83， 28， 77， 73	3.871
2	65， 112， 40， 87， 66， 51， 97， 17， 8， 22， 29， 46， 115， 92， 33， 88， 109， 55， 21， 93， 75， 89	3.299
3	31， 47， 43， 98， 52， 111， 57， 16， 15， 58， 104， 42， 74， 7， 6， 24， 95， 35， 39	3.195
4	69， 122， 103， 117， 91， 118， 84， 100， 48， 82， 60， 85， 99， 5， 120， 119， 54	2.029
5	44， 86， 59， 4， 105， 102， 36， 23， 68， 67， 114， 76	1.737
6	30， 49， 20， 116， 32， 81， 72， 64， 108， 79， 50， 12， 53， 56， 113	1.437
7	71， 96， 45， 1， 38， 11， 110， 26， 70	0.796

Table 5

References 40

1	SHAN M H， GUO J， GILL E. Review and comparison of active space debris capturing and removal methods［J］. Progress in Aerospace Sciences， 2016， 80： 18-32.
2	SIZOV D A， ASLANOV V S. Space debris removal with harpoon assistance： Choice of parameters and optimization［J］. Journal of Guidance， Control， and Dynamics， 2021， 44（4）： 767-778.
3	LI Y X， HUO J， MA P， et al. Target localization method of non-cooperative spacecraft on on-orbit service［J］. Chinese Journal of Aeronautics， 2022， 35（11）： 336-348.
4	ZHANG J， PARKS G T， LUO Y Z， et al. Multispacecraft refueling optimization considering the J2 perturbation and window constraints［J］. Journal of Guidance， Control， and Dynamics， 2014， 37（1）： 111-122.
5	GAO Y T， LU X， PENG Y M， et al. Trajectory optimization of multiple asteroids exploration with asteroid 2010TK₇ as main target［J］. Advances in Space Research， 2019， 63（1）： 432-442.
6	PELONI A， CERIOTTI M， DACHWALD B. Solar-sail trajectory design for a multiple near-earth-asteroid rendezvous mission［J］. Journal of Guidance， Control， and Dynamics， 2016， 39（12）： 2712-2724.
7	HELVIG C S， ROBINS G， ZELIKOVSKY A. The moving-target traveling salesman problem［J］. Journal of Algorithms， 2003， 49（1）： 153-174.
8	SAAD S， WAN JAAFAR W N， JAMIL S J. Solving standard traveling salesman problem and multiple traveling salesman problem by using branch-and-bound［C］∥ AIP Conference Proceedings. 2013.
9	TOMANOVÁ P， HOLÝ V. Ant colony optimization for time-dependent travelling salesman problem［C］∥Proceedings of the 2020 4th International Conference on Intelligent Systems， Metaheuristics & Swarm Intelligence. New York： ACM， 2020： 47-51.
10	ZHAO J F， FENG W M， YUAN J P. A novel two-level optimization strategy for multi-debris active removal mission in LEO［J］. Computer Modeling in Engineering & Sciences， 2020， 122（1）： 149-174.
11	朱阅訸. 面向大规模目标访问任务的飞行序列规划方法［D］. 长沙：国防科技大学， 2020.
	ZHU Y H. Flight sequence planning method for large-scale-object visiting mission［D］. Changsha： National University of Defense Technology， 2020 （in Chinese）.
12	SHANG H B， LIU Y X. Assessing accessibility of main-belt asteroids based on Gaussian process regression［J］. Journal of Guidance， Control， and Dynamics， 2017， 40（5）： 1144-1154.
13	HUANG A Y， LUO Y Z， LI H N. Fast estimation of perturbed impulsive rendezvous via semi-analytical equality-constrained optimization［J］. Journal of Guidance， Control， and Dynamics， 2020， 43（12）： 2383-2390.
14	ZHU Y H， LUO Y Z. Fast approximation of optimal perturbed long-duration impulsive transfers via artificial neural networks［J］. IEEE Transactions on Aerospace and Electronic Systems， 2021， 57（2）： 1123-1138.
15	ZHU Y H， LUO Y Z. Fast evaluation of low-thrust transfers via multilayer perceptions［J］. Journal of Guidance， Control， and Dynamics， 2019， 42（12）： 2627-2637.
16	VIAVATTENE G， CERIOTTI M. Artificial neural networks for multiple NEA rendezvous missions with continuous thrust［J］. Journal of Spacecraft and Rockets， 2022， 59（2）： 574-586.
17	CUI P Y， QIAO D， CUI H T， et al. Target selection and transfer trajectories design for exploring asteroid mission［J］. Science China Technological Sciences， 2010， 53（4）： 1150-1158.
18	CERF M. Multiple space debris collecting mission—debris selection and trajectory optimization［J］. Journal of Optimization Theory and Applications， 2013， 156（3）： 761-796.
19	HUANG A Y， LUO Y Z， LI H N. Global optimization of multiple-spacecraft rendezvous mission via decomposition and dynamics-guide evolution approach［J］. Journal of Guidance， Control， and Dynamics， 2022， 45（1）： 171-178.
20	WANG H J， YANG Z， ZHOU W G， et al. Online scheduling of image satellites based on neural networks and deep reinforcement learning［J］. Chinese Journal of Aeronautics， 2019， 32（4）： 1011-1019.
21	LITTLE B D， FRUEH C E. Space situational awareness sensor tasking： Comparison of machine learning with classical optimization methods［J］. Journal of Guidance， Control， and Dynamics， 2020， 43（2）： 262-273.
22	刘冰雁，叶雄兵，周赤非，等. 基于改进DQN的复合模式在轨服务资源分配［J］. 航空学报， 2020， 41（5）： 323630.
	LIU B Y， YE X B， ZHOU C F， et al. Allocation of composite mode on-orbit service resource based on improved DQN［J］. Acta Aeronautica et Astronautica Sinica， 2020， 41（5）： 323630 （in Chinese）.
23	IZZO D， MÄRTENS M， PAN B F. A survey on artificial intelligence trends in spacecraft guidance dynamics and control［J］. Astrodynamics， 2019， 3（4）： 287-299.
24	SONG Y， GONG S P. Solar-sail trajectory design for multiple near-Earth asteroid exploration based on deep neural networks［J］. Aerospace Science and Technology， 2019， 91： 28-40.
25	IZZO D， ÖZTÜRK E. Real-time guidance for low-thrust transfers using deep neural networks［J］. Journal of Guidance， Control， and Dynamics， 2021， 44（2）： 315-327.
26	ZAVOLI A， FEDERICI L. Reinforcement learning for robust trajectory design of interplanetary missions［J］. Journal of Guidance， Control， and Dynamics， 2021， 44（8）： 1440-1453.
27	SÁNCHEZ-SÁNCHEZ C， IZZO D. Real-time optimal control via deep neural networks： Study on landing problems［J］. Journal of Guidance， Control， and Dynamics， 2018， 41（5）： 1122-1135.
28	SCORSOGLIO A， D’AMBROSIO A， GHILARDI L， et al. Image-based deep reinforcement meta-learning for autonomous lunar landing［J］. Journal of Spacecraft and Rockets， 2022， 59（1）： 153-165.
29	YANG B， LI S A， FENG J L， et al. Fast solver for J2-perturbed lambert problem using deep neural network［J］. Journal of Guidance， Control， and Dynamics， 2022， 45（5）： 875-884.
30	PENG H， BAI X L. Artificial neural network–based machine learning approach to improve orbit prediction accuracy［J］. Journal of Spacecraft and Rockets， 2018， 55（5）： 1248-1260.
31	VINYALS O， FORTUNATO M， JAITLY N. Pointer networks［DB/OL］. arXiv preprint： 1506.03134， 2015.
32	GU S S， HAO T， YAO H M. A pointer network based deep learning algorithm for unconstrained binary quadratic programming problem［J］. Neurocomputing， 2020， 390： 1-11.
33	GU S S， YAO H M. Pointer network based deep learning algorithm for the maximum clique problem［J］. International Journal on Artificial Intelligence Tools， 2021， 30（1）： 2140004.
34	GU S S， YANG Y E. A deep learning algorithm for the max-cut problem based on pointer network structure with supervised learning and reinforcement learning strategies［J］. Mathematics， 2020， 8（2）： 298.
35	马一凡，赵凡宇，王鑫，等. 基于改进指针网络的卫星对地观测任务规划方法［J］. 浙江大学学报（工学版）， 2021， 55（2）： 395-401.
	MA Y F， ZHAO F Y， WANG X， et al. Satellite earth observation task planning method based on improved pointer networks［J］. Journal of Zhejiang University （Engineering Science）， 2021， 55（2）： 395-401 （in Chinese）.
36	HOCHREITER S， SCHMIDHUBER J. Long short-term memory［J］. Neural Computation， 1997， 9（8）： 1735-1780.
37	KIM Y. Convolutional neural networks for sentence classification［DB/OL］. arXiv preprint： 1408.5882， 2014.
38	NUDT. Problem data of the GTOC11： Candidate asteroids［EB/OL］. .
39	ESA. Problem data of the GTOC9： Debris orbits［EB/OL］. .
40	BANG J， AHN J. Multitarget rendezvous for active debris removal using multiple spacecraft［J］. Journal of Spacecraft and Rockets， 2019， 56（4）： 1237-1247.

超参数	数值
超参数	算例1	算例2	算例3
特征属性数量	4	6	9
一维卷积核大小	1	3	3
特征属性嵌入维度	128	512	512
LSTM隐层维度	128	512	512
Attention匹配维度	128	512	512
训练中的目标数量	15	15	15
环境交互总回合数	2 000	5 000	8 000
优化器学习率	0.001	0.001	0.001
Softmax蒸馏温度初值	5.0	5.0	5.0
测试中的目标数量	15~20	15~20	15~20

性能指标	数值
性能指标	算例1	算例2	算例3
模型训练时间/s	23.62	4.77×10³	3.09×10⁴
Actor规划时间/ms	0.255	3.74	5.15
蚁群算法收敛时间/s	8.31	96.13	312.93
Actor得最优序列概率/%	96.15	92.38	88.71
Actor得近最优序列概率/%	100.00	96.31	92.11
Actor故障概率/%	0	0.82	1.92
Critic计算时间/ms	0.183	2.17	2.89
Critic误差率/%	7.21	7.33	9.37
Critic故障概率/%	13.82	16.35	12.55

[1]	Zhilin FAN, Hongyong YANG, Yilin HAN. Target round-up control for multi-agent systems based on reinforcement learning [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(S1): 727487-727487.
[2]	Xiaowei FU, Zhe XU, Jindong ZHU, Nan WANG. Maneuvering decision-making of multi-UAV attack-defence confrontation based on PER-MATD3 [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(7): 327083-327083.
[3]	Xizhen GAO, Liang TANG, Huang HUANG. Deep reinforcement learning in autonomous manipulation for celestial bodies exploration: Applications and challenges [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(6): 26762-026762.
[4]	Pan ZHOU, Jiangtao HUANG, Sheng ZHANG, Gang LIU, Bowen SHU, Jigang TANG. Intelligent air combat decision making and simulation based on deep reinforcement learning [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(4): 126731-126731.
[5]	Lei DONG, Hongbing CHEN, Xi CHEN, Changxiao ZHAO. Distributed multi-agent coalition task allocation strategy for single pilot operation mode based on DQN [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(13): 327895-327895.
[6]	Wenxue CHEN, Changsheng GAO, Wuxing JING. Trust region policy optimization guidance algorithm for intercepting maneuvering target [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(11): 327596-327596.
[7]	Junpeng HUI, Ren WANG, Jifeng GUO. Intelligent guidance for no⁃fly zone avoidance based on reinforcement learning [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(11): 327416-327416.
[8]	Sheng ZHANG, Pan ZHOU, Yang HE, Jiangtao HUANG, Gang LIU, Jigang TANG, Huaizhi JIA, Xin DU. Air combat maneuver decision-making test based on deep reinforcement learning [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(10): 128094-128094.
[9]	HUI Junpeng, WANG Ren, YU Qidong. Generating new quality flight corridor for reentry aircraft based on reinforcement learning [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(9): 325960-325960.
[10]	FU Xiaowei, WANG Hui, XU Zhe. Cooperative pursuit strategy for multi-UAVs based on DE-MADDPG algorithm [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(5): 325311-325311.
[11]	LUO Qing, ZHANG Tao, SHAN Peng, ZHANG Wentao, LIU Zihao. Generating reconfiguration blueprints for IMA systems based on improved Q-learning [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(8): 525792-525792.
[12]	SUN Zhixiao, YANG Shengqi, PIAO Haiyin, BAI Chengchao, GE Jun. A survey of air combat artificial intelligence [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(8): 525799-525799.
[13]	REN Feng, GAO Chuanqiang, TANG Hui. Machine learning for flow control: Applications and development trends [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(4): 524686-524686.
[14]	YANG Jianan, HOU Xiaolei, HU Yu Hen, LIU Yong, PAN Quan, FENG Qian. Heuristic enhanced reinforcement learning method for large-scale multi-debris active removal mission planning [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(4): 524354-524354.
[15]	XIANG Xiaojia, YAN Chao, WANG Chang, YIN Dong. Coordination control method for fixed-wing UAV formation through deep reinforcement learning [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(4): 524009-524009.

Space target rendezvous sequence planning via pointer networks

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 13

References 40

Related Articles 15

Recommended Articles

Metrics

Comments