Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 37 of the 37 references it cites were found in a registry or an index.

DRL-Based Dynamic Destroy Approaches for Agile-Satellite Mission Planning

Open access
Huang, Wei, Li, Zongwang, He, Xiaohe, Xiang, Junyan, Du, Xu, Liang, Xuwen · 2023 · Remote Sensing

Abstract

Agile-satellite mission planning is a crucial issue in the construction of satellite constellations. The large scale of remote sensing missions and the high complexity of constraints in agile-satellite mission planning pose challenges in the search for an optimal solution. To tackle the issue, a dynamic destroy deep-reinforcement learning (D3RL) model is designed to facilitate subsequent optimization operations via adaptive destruction to the existing solutions. Specifically, we first perform a clustering and embedding operation to reconstruct tasks into a clustering graph, thereby improving data utilization. Secondly, the D3RL model is established based on graph attention networks (GATs) to enhance the search efficiency for optimal solutions. Moreover, we present two applications of the D3RL model for intensive scenes: the deep-reinforcement learning (DRL) method and the D3RL-based large-neighborhood search method (DRL-LNS). Experimental simulation results illustrate that the D3RL-based approaches outperform the competition in terms of solutions’ quality and computational efficiency, particularly in more challenging large-scale scenarios. DRL-LNS outperforms ALNS with an average scheduling rate improvement of approximately 11% in Area instances. In contrast, the DRL approach performs better in World scenarios, with an average scheduling rate that is around 8% higher than that of ALNS.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/rs15184503

Cites

37 of 37 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Agile satellite observation. Agile satellites with three-axis maneuvering capabilities must schedule a considerable number of remote sensing missions within a limited time frame. Targets beyond the observation range remain undetectable and are therefore considered invalid.
Figure 1 Agile satellite observation. Agile satellites with three-axis maneuvering capabilities must schedule a considerable number of remote sensing missions within a limited time frame. Targets beyond the observation range remain undetectable and are therefore considered invalid.CC-BY · From the PDF
Figure 2 The process of attention and aggregation operation in graph attention layer.
Figure 2 The process of attention and aggregation operation in graph attention layer.CC-BY · From the PDF
Figure 3 Actor-network. The attention mechanism is incorporated into the network architecture, where the initial input tasks are embedded as graph nodes. These graph nodes undergo multiple layers of GAT networks to extract node features, ultimately obtaining the set of nodes to be destroyed in the next iteration.
Figure 3 Actor-network. The attention mechanism is incorporated into the network architecture, where the initial input tasks are embedded as graph nodes. These graph nodes undergo multiple layers of GAT networks to extract node features, ultimately obtaining the set of nodes to be destroyed in the next iteration.CC-BY · From the PDF
Figure 4 Agent-environment interaction in AEOS.
Figure 4 Agent-environment interaction in AEOS.CC-BY · From the PDF
Figure 5 Training curves for the Area and World scenarios. The two plots show the average scheduled rate of Area and World instances, respectively. The results were obtained by conducting experiments with 5 different random seeds. ( a ) Training process for Area targets. ( b ) Training process for World targets.
Figure 5 Training curves for the Area and World scenarios. The two plots show the average scheduled rate of Area and World instances, respectively. The results were obtained by conducting experiments with 5 different random seeds. ( a ) Training process for Area targets. ( b ) Training process for World targets.CC-BY · From the PDF
Figure 6 Optimization scheduled rate performance contrast. Boxplot illustrating comparison based on 10 different instances. Each box represents the interquartile range (IQR) of the data distribution, with the median marked by a horizontal line inside the box. Triangle markers indicate the mean values of the test results. ( a ) Scene: Area. ( b ) Scene: World.
Figure 6 Optimization scheduled rate performance contrast. Boxplot illustrating comparison based on 10 different instances. Each box represents the interquartile range (IQR) of the data distribution, with the median marked by a horizontal line inside the box. Triangle markers indicate the mean values of the test results. ( a ) Scene: Area. ( b ) Scene: World.CC-BY · From the PDF
Figure 7 Optimization Scheduled rate performance contrast. ( a ) Scene: Area. ( b ) Scene: World.
Figure 7 Optimization Scheduled rate performance contrast. ( a ) Scene: Area. ( b ) Scene: World.CC-BY · From the PDF
Figure 8 Cont .
Figure 8 Cont .CC-BY · From the PDF
Figure 8 Performance comparison of algorithms with 1500 iterations. Each instance size evaluated across 10 test scenarios. Note: The GA algorithm fails to find a feasible solution when the number of planning tasks exceeds 200. ( a ) Scene: Area instance size: 50. ( b ) Scene: Area instance size: 100. ( c ) Scene: Area instance size: 150. ( d ) Scene: Area instance size: 200. ( e ) Scene: World instance size: 100. ( f ) Scene: World instance size: 200. ( g ) Scene: World instance size: 300. ( h ) Scene: World instance size: 400.
Figure 8 Performance comparison of algorithms with 1500 iterations. Each instance size evaluated across 10 test scenarios. Note: The GA algorithm fails to find a feasible solution when the number of planning tasks exceeds 200. ( a ) Scene: Area instance size: 50. ( b ) Scene: Area instance size: 100. ( c ) Scene: Area instance size: 150. ( d ) Scene: Area instance size: 200. ( e ) Scene: World instance size: 100. ( f ) Scene: World instance size: 200. ( g ) Scene: World instance size: 300. ( h ) Scene: World instance size: 400.CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 30 references

Research works printed but not linked: 4

  1. Singh, P., Pandey, P.C., Petropoulos, G.P., Pavlides, A., Srivastava, P.K., Koutsias, N., Deng, K.A.K., and Bao, Y. (2020). Hyperspectral Remote Sensing, Elsevier.
  2. Hottung, A., and Tierney, K. (2019). Neural large neighborhood search for the capacitated vehicle routing problem. arXiv.
  3. Iacopino, C., Harrison, S., and Brewer, A. (2015, January 25–27). Mission planning systems for commercial small-sat earth observation constellations. Proceedings of the 9th International Workshop on Planning and Scheduling for Space (IWPSS), VenueBuenos Aires, Argentina.
  4. Veličković, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., and Bengio, Y. (2017). Graph attention networks. arXiv.

Cited by

See all 11

Metadata

Authors
Huang, W., Li, Z., He, X., Xiang, J., Du, X., & Liang, X.
Affiliations
  • ShanghaiTech University (CN)
  • University of Chinese Academy of Sciences (CN)
  • Henan Academy of Sciences (CN)
Corresponding author
Liang, Xuwen
Venue
Remote Sensing
Publisher
MDPI AG
Published
2023-09-13
Volume
15
Issue
18
Pages
4503
ISSN
2072-4292 (ISSN-L)
OpenAlex
W4386748751
Cited by
11
References
30
Citations by year
2026: 4 · 2025: 3 · 2024: 4
FWCI
1.34
Type
journal article
Language
en
Open access
Locations
  • Remote Sensing (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
National Key R&D Program of China (2022YFB3902800)
Topics
  • Physical Sciences › Engineering › Aerospace Engineering › Satellite Communication Systems
  • Physical Sciences › Engineering › Industrial and Manufacturing Engineering › Scheduling and Optimization Algorithms
  • Physical Sciences › Engineering › Industrial and Manufacturing Engineering › Vehicle Routing Optimization Methods
Keywords
deep reinforcement learning, graph attention networks, large neighborhood search

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: DRL-Based Dynamic Destroy Approaches for Agile-Satellite Mission Planning. https://app.cochat.ai/papers/10.3390/rs15184503
The paper
Huang, W., Li, Z., He, X., Xiang, J., Du, X., & Liang, X. (2023). DRL-Based Dynamic Destroy Approaches for Agile-Satellite Mission Planning. Remote Sensing. https://doi.org/10.3390/rs15184503

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.