Deep reinforcement learning-based spacecraft attitude control with pointing keep-out constraint

TL;DR

This paper presents a DRL-based approach using SAC for spacecraft attitude control with pointing constraints, employing a new state representation and curriculum learning to ensure effective reorientation within the keep-out zone.

Contribution

The paper introduces a novel DRL framework with a specialized state representation and curriculum learning for constrained spacecraft attitude control.

Findings

Effective control within pointing constraints demonstrated in simulations

New state representation improves constraint handling

Curriculum learning accelerates training convergence

Abstract

This paper implements deep reinforcement learning (DRL) for spacecraft reorientation control with a single pointing keep-out zone. The Soft Actor-Critic (SAC) algorithm is adopted to handle continuous state and action space. A new state representation is designed to explicitly include a compact representation of the attitude constraint zone. The reward function is formulated to achieve the control objective while enforcing the attitude constraint. A curriculum learning approach is used for the agent training. Simulation results demonstrate the effectiveness of the proposed DRL-based method for spacecraft pointing-constrained attitude control.