mlsanityData to 14 Oct 2026

Transformer-Based Reinforcement Learning for Autonomous Orbital Collision Avoidance in Partially Observable Environments

arXiv 2602.06088 ↗First posted 5 Feb 2026cs.LG cs.AIcited by 0

On the map: Deep reinforcement learning methods → RL for robot navigation and UAV control

Abstract (arXiv)

We introduce a Transformer-based Reinforcement Learning framework for autonomous orbital collision avoidance that explicitly models the effects of partial observability and imperfect monitoring in space operations. The framework combines a configurable encounter simulator, a distance-dependent observation model, and a sequential state estimator to represent uncertainty in relative motion. A central contribution of this work is the use of transformer-based Partially Observable Markov Decision Process (POMDP) architecture, which leverage long-range temporal attention to interpret noisy and intermittent observations more effectively than traditional architectures. This integration provides a foundation for training collision avoidance agents that can operate more reliably under imperfect monitoring environments.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for robot navigation and UAV control, one of 15 directions in Deep reinforcement learning methods. Builds on 18 of its 38 references that are in the map. First posted on arXiv 5 Feb 2026.

Builds on

18 of 38 references.

References · 18

  1. 2017Attention Is All You Needcited by 60,361
  2. 2017Proximal Policy Optimization Algorithmscited by 15,215
  3. 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,465
  4. 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 2,026
  5. 2019Transformer-XL: Attentive Language Models Beyond a Fixed-Length Contextcited by 1,850
  6. 2015Deep Recurrent Q-Learning for Partially Observable MDPscited by 465
  7. 2019Stabilizing Transformers for Reinforcement Learningcited by 270
  8. 2022TransDreamer: Reinforcement Learning with Transformer World Modelscited by 143
  9. 2018Deep Variational Reinforcement Learning for POMDPscited by 125
  10. 2019Fast Trajectory Optimization via Successive Convexification for Spacecraft Rendezvous with Integer Constraintscited by 5
  11. 2021A Distance-based Anomaly Detection Framework for Deep Reinforcement Learningcited by 5
  12. 2023Social Robot Navigation through Constrained Optimization: a Comparative Study of Uncertainty-based Objectives and Constraintscited by 2
6 more
  1. 2024Recursive Polynomial Method for Fast Collision Avoidance Maneuver Designcited by 2
  2. 2024Risk-Sensitive Orbital Debris Collision Avoidance using Distributionally Robust Chance Constraintscited by 2
  3. 2024Collision Avoidance Maneuvers Optimization in the Presence of Multiple Encounterscited by 1
  4. 2025Chance-constrained Model Predictive Control for Near Rectilinear Halo Orbit spacecraft rendezvouscited by 1
  5. 2025CAMmary: A Review of Spacecraft Collision Avoidance Manoeuvre Design Methodscited by 1
  6. 2025Successive Convexification for Passively-Safe Spacecraft Rendezvous on Near Rectilinear Halo Orbitcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025A Markov Decision Process Framework for Early Maneuver Decisions in Satellite Collision Avoidance
  2. 2026Orbit-Planner: Towards Latent World Models for On-Orbit Obstacle Avoidance of Satellite Agents
  3. 2024On-orbit Servicing for Spacecraft Collision Avoidance With Autonomous Decision Making
  4. 2025OrbitZoo: Real Orbital Systems Challenges for Reinforcement Learning
  5. 2025Multi-Phase Spacecraft Trajectory Optimization via Transformer-Based Reinforcement Learning
  6. 2025RL-AVIST: Reinforcement Learning for Autonomous Visual Inspection of Space Targets
  7. 2020Learning Navigation Costs from Demonstration in Partially Observable Environments
  8. 2025Investigating the Impact of Observation Space Design Choices On Training Reinforcement Learning Solutions for Spacecraft Problems