Impact of Partially Observable Markov Decision Process in Next Generation Satellite for Remote Sensing

Year : 2025 | Volume : 03 | Issue : 02 | Page : 20 28
    By

    K.V.V. Subba Rao,

  • Anantham Srujana Jyothi,

  • Manas Kumar Yogi,

  1. Assistant Professor, CSE Department, Pragati Engineering College (A), Surampalem , East Godavari district, Andhra Pradesh, India
  2. Assistant Professor, CSE-AI&ML Department, Pragati Engineering College (A), Surampalem , East Godavari district, Andhra Pradesh, India
  3. Student, CSE Department, Pragati Engineering College (A), Surampalem , East Godavari district, Andhra Pradesh, India

Abstract

The integration of Partially Observable Markov Decision Processes (POMDPs) in next- generation satellite systems represents a transformative advancement in remote sensing technology. This article explores how POMDP frameworks address the inherent uncertainties and incomplete observability challenges in satellite operations, including dynamic task scheduling, resource allocation, and adaptive sensing strategies. By modeling satellite decision-making under uncertainty, POMDPs enable autonomous systems to optimize mission objectives while managing constraints such as limited power, bandwidth, and computational resources. This article looks at how POMDP-based techniques handle important operational issues in satellite systems, such as adaptive sensing techniques, real- time resource allocation, risk-aware planning, and dynamic job scheduling. POMDP models enable autonomous onboard systems to respect mission-critical constraints like limited power, restricted bandwidth, onboard computational capacity, and orbital mechanics while optimizing mission objectives like imaging quality, area coverage, response time, and energy efficiency by explicitly representing uncertainty in state transitions and observational inputs. The study examines key application areas including Earth observation, disaster monitoring, and multi-satellite coordination, demonstrating how POMDP-based approaches enhance data acquisition quality, reduce latency, and improve overall mission efficiency. The paper examines a variety of application domains, such as target tracking, cooperative multi-satellite coordination, environmental and disaster monitoring, and Earth observation, where POMDPs show special usefulness. These applications demonstrate how POMDP-driven decision- making can improve overall mission robustness, decrease task delay, and improve data collecting accuracy under extremely dynamic and unpredictable circumstances. This paper highlights the increasing significance of POMDP frameworks in enabling intelligent, adaptable, and robust remote sensing systems through an analysis of recent theoretical advances, algorithmic breakthroughs, and practical demonstrations in operational situations. According to the article’s conclusion, POMDP-based strategies will be essential to the development of completely autonomous satellite missions that can react instantly to changing operational requirements and environmental conditions. Through analysis of recent implementations and theoretical frameworks, this article highlights the critical role of POMDPs in enabling intelligent, adaptive remote sensing systems capable of responding to evolving environmental conditions and mission requirements in real-time.

Keywords: Markov, Remote, Decision, Satellite, Network, Partially Observable Markov Decision Processes (POMDPs)

[This article belongs to International Journal of Satellite Remote Sensing ]

How to cite this article:
K.V.V. Subba Rao, Anantham Srujana Jyothi, Manas Kumar Yogi. Impact of Partially Observable Markov Decision Process in Next Generation Satellite for Remote Sensing. International Journal of Satellite Remote Sensing. 2025; 03(02):20-28.
How to cite this URL:
K.V.V. Subba Rao, Anantham Srujana Jyothi, Manas Kumar Yogi. Impact of Partially Observable Markov Decision Process in Next Generation Satellite for Remote Sensing. International Journal of Satellite Remote Sensing. 2025; 03(02):20-28. Available from: https://journals.stmjournals.com/ijsrs/article=2025/view=235434


References

  1. Zhang G, Li X, Hu G, Li Y, Wang X, Zhang Z. Marl-based multi-satellite intelligent task planning method. Ieee Access. 2023 Nov 28;11:135517-28.
  2. Lauri M, Hsu D, Pajarinen J. Partially observable markov decision processes in robotics: A survey. IEEE Transactions on Robotics. 2022 Sep 22;39(1):21-40.
  3. Chen H, Peng S, Du C, Li J. Earth observation satellites: task planning and scheduling. Springer Nature; 2023 Sep 4.
  4. Hady MA, Hu S, Pratama M, Cao J, Kowalczyk R. Multi-Agent Reinforcement Learning for Autonomous Multi-Satellite Earth Observation: A Realistic Case Study. arXiv preprint arXiv:2506.15207. 2025 Jun 18.
  5. Chai F, Zhang Q, Yao H, Xin X, Wang F, Xu M, Xiong Z, Niyato D. Multi-agent DDPG based resource allocation in NOMA-enabled satellite IoT. IEEE Transactions on Communications. 2024 May 7;72(10):6287-300.
  6. Zhou X, Wang W, Wang T, Lei Y, Zhong F. Bayesian reinforcement learning for multi-robot decentralized patrolling in uncertain environments. IEEE Transactions on Vehicular Technology. 2019 Oct 22;68(12):11691-703.
  7. Wang Y, Wang L, Chen X, Liang D. Offshore petroleum leaking source detection method from remote sensing data via deep reinforcement learning with knowledge transfer. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2022 Jul 18;15:5826-40.
  8. Oestreich CE. Robust Control and Learning for Autonomous Spacecraft Proximity Operations with Uncertainty (Doctoral dissertation, Massachusetts Institute of Technology).
  9. Stephens R, Lacerda B, Hawes N. Hierarchical planning for resource-constrained long-term monitoring missions in time-varying environments.
  10. Lu K, Liu C, Li C, Chen R. Flight Dynamics Modeling and Dynamic Stability Analysis of Tilt‐Rotor Aircraft. International Journal of Aerospace Engineering. 2019;2019(1):5737212.
  11. Xing BY, Pan F, Feng XX, Li WX, Gao Q. Autonomous landing of a micro aerial vehicle on a moving platform using a composite landmark. International Journal of Aerospace Engineering. 2019;2019(1):4723869.
  12. Zhu Y, Liang S, Gong M, Yan J. Decomposed POMDP optimization-based sensor management for multi-target tracking in passive multi-sensor systems. IEEE Sensors Journal. 2022 Jan 6;22(4):3565-78.
  13. Ross S, Pineau J, Paquet S, Chaib-Draa B. Online planning algorithms for POMDPs. Journal of Artificial Intelligence Research. 2008 Jul 29;32:663-704.
  14. Al Amin MA, Shetty S, Njilla LL, Tosh DK, Kamhoua CA. Dynamic cyber deception using partially observable Monte‐Carlo planning framework. Modeling and Design of Secure Internet of Things. 2020 Jul 15:331-55.
  15. Sun L, Yang Y, Duan Q, Shi Y, Lyu C, Chang YC, Lin CT, Shen Y. Multi-agent coordination across diverse applications: A survey. arXiv preprint arXiv:2502.14743. 2025 Feb 20.
  16. Shorinwa O, Schwager M. Distributed model predictive control via separable optimization in multiagent networks. IEEE Transactions on Automatic Control. 2023 Apr 21;69(1):230-45.
  17. Harris A, Naik K. Autonomous Command and Control for Earth-Observing Satellites using Deep Reinforcement Learning. In2023 IEEE Aerospace Conference 2023 Mar 4 (pp. 1-12). IEEE.
  18. Li J, Li C, Wang F. Automatic scheduling for earth observation satellite with temporal specifications. IEEE Transactions on Aerospace and Electronic Systems. 2020 Jan 15;56(4):3162-9.
  19. Liang H, Zhang W. A survey and taxonomy of resource allocation methods in wireless networks. Journal of Communications and Information Networks. 2021 Dec 24;6(4):372-84.
  20. Aziz HA, Das S. Deep reinforcement learning applications in connected-automated transportation systems. InDeep Learning and Its Applications for Vehicle Networks 2023 May 12 (pp. 133-164). CRC Press.
  21. Du Y, Wang T, Xin B, Wang L, Chen Y, Xing L. A data-driven parallel scheduling approach for multiple agile earth observation satellites. IEEE Transactions on Evolutionary Computation. 2019 Aug 9;24(4):679-93.
Regular Issue Subscription Review Article
Volume 03
Issue 02
Received 15/11/2025
Accepted 19/11/2025
Published 31/12/2025
Publication Time 46 Days
168

Login


My IP

PlumX Metrics