Domain Generalization for In-Orbit 6D Pose Estimation

2024-06-17

arXiv.org

AI Summary

To address the synthetic-to-real domain gap in 6D pose estimation of on-orbit target spacecraft from monocular images, this paper proposes an end-to-end multi-task neural network architecture integrating illumination-robust data augmentation and domain-invariant feature learning. By jointly optimizing pose estimation, depth prediction, and keypoint reconstruction, the model explicitly enforces extraction of cross-domain consistent representations invariant to illumination variations, shadows, and texture changes. Evaluated on the SPEED+ benchmark, the method achieves state-of-the-art performance, reducing average pose error by 23.7% over existing approaches. It significantly narrows the generalization gap between synthetically trained models and real-space imagery, delivering a highly robust, annotation-efficient 6D pose estimation solution for autonomous spacecraft rendezvous and docking.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Biometrics, Face, Gesture & Pose
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

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.

Problem

• Bridging domain gap for spacecraft pose estimation
• Generalizing from synthetic to real orbital images
• Enabling domain-invariant 6D pose estimation networks

Innovation

• Neural-based end-to-end architecture
• Multi-task learning strategy
• Aggressive data augmentation policies