Domain Generalization for In-Orbit 6D Pose Estimation

TL;DR

This paper presents a novel neural network architecture and training strategy to improve domain generalization for 6D spacecraft pose estimation from monocular images, effectively bridging the gap between synthetic training data and real orbital images.

Contribution

It introduces an end-to-end neural architecture combined with multi-task learning and aggressive data augmentation to enhance domain-invariant feature learning for spacecraft pose estimation.

Findings

Achieves state-of-the-art accuracy on SPEED+ dataset.

Effectively closes the domain gap between synthetic and real images.

Ablation studies confirm the effectiveness of key components.

Abstract

We address the problem of estimating the relative 6D pose, i.e., position and orientation, of a target spacecraft, from a monocular image, a key capability for future autonomous Rendezvous and Proximity Operations. Due to the difficulty of acquiring large sets of real images, spacecraft pose estimation networks are exclusively trained on synthetic ones. However, because those images do not capture the illumination conditions encountered in orbit, pose estimation networks face a domain gap problem, i.e., they do not generalize to real images. Our work introduces a method that bridges this domain gap. It relies on a novel, end-to-end, neural-based architecture as well as a novel learning strategy. This strategy improves the domain generalization abilities of the network through multi-task learning and aggressive data augmentation policies, thereby enforcing the network to learn domain-invariant features. We demonstrate that our method effectively closes the domain gap, achieving state-of-the-art accuracy on the widespread SPEED+ dataset. Finally, ablation studies assess the impact of key components of our method on its generalization abilities.