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

TL;DR

This paper presents a novel Transformer-based reinforcement learning approach for autonomous orbital collision avoidance, effectively handling partial observability and noisy data in space operations.

Contribution

It introduces a Transformer-based POMDP architecture that improves decision-making in uncertain, partially observable space environments, a novel application in orbital collision avoidance.

Findings

Transformer architecture enhances interpretation of noisy observations.

The framework improves collision avoidance reliability under partial observability.

Integration of long-range attention benefits space operation decision processes.

Abstract

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.