Satellite Pose Estimation Challenge: Dataset, Competition Design and Results

Abstract

Reliable pose estimation of uncooperative satellites is a key technology for enabling future on-orbit servicing and debris removal missions. The Kelvins Satellite Pose Estimation Challenge aims at evaluating and comparing monocular vision-based approaches and pushing the state-of-the-art on this problem. This work is based on the Satellite Pose Estimation Dataset, the first publicly available machine learning set of synthetic and real spacecraft imageries. The choice of dataset reflects one of the unique challenges associated with spaceborne computer vision tasks, namely the lack of spaceborne images to train and validate the developed algorithms. This work briefly reviews the basic properties and the collection process of the dataset which was made publicly available. The competition design, including the definition of performance metrics and the adopted testbed, is also discussed. The main contribution of this paper is the analysis of the submissions of the 48 competitors, which compares the performance of different approaches and uncovers what factors make the satellite pose estimation problem especially challenging.

Accepted to IEEE Transactions on Aerospace and Electronic Systems

References in corpus (7)

• YOLOv3: An Incremental Improvement
• High-Resolution Representations for Labeling Pixels and Regions
• Rethinking the Inception Architecture for Computer Vision
• Deep Learning for Spacecraft Pose Estimation from Photorealistic Rendering
• Estimating 6D Pose From Localizing Designated Surface Keypoints
• Towards Robust Learning-Based Pose Estimation of Noncooperative Spacecraft
• Pose Estimation for Non-Cooperative Rendezvous Using Neural Networks

Cited by in corpus (24)

• Artificial Intelligence for Satellite Communication: A Review
• Robust Multi-Task Learning and Online Refinement for Spacecraft Pose Estimation across Domain Gap
• A Survey on Deep Learning-Based Monocular Spacecraft Pose Estimation: Current State, Limitations and Prospects
• Adaptive Neural Network-based Unscented Kalman Filter for Robust Pose Tracking of Noncooperative Spacecraft
• Performance Study of YOLOv5 and Faster R-CNN for Autonomous Navigation around Non-Cooperative Targets
• Challenges for Monocular 6D Object Pose Estimation in Robotics
• Autonomous Angles-Only Multi-Target Tracking for Spacecraft Swarms
• Real-Time, Flight-Ready, Non-Cooperative Spacecraft Pose Estimation Using Monocular Imagery
• Lessons from a Space Lab -- An Image Acquisition Perspective
• Mobile-URSONet: an Embeddable Neural Network for Onboard Spacecraft Pose Estimation
• Advancing 6-DoF Instrument Pose Estimation in Variable X-Ray Imaging Geometries
• Robust On-Manifold Optimization for Uncooperative Space Relative Navigation with a Single Camera
• SU-Net: Pose estimation network for non-cooperative spacecraft on-orbit
• Online Supervised Training of Spaceborne Vision during Proximity Operations using Adaptive Kalman Filtering
• Domain Generalization for In-Orbit 6D Pose Estimation
• Bridging the Domain Gap for Flight-Ready Spaceborne Vision
• A Spacecraft Dataset for Detection, Segmentation and Parts Recognition
• Event-RGB Fusion for Spacecraft Pose Estimation Under Harsh Lighting
• Robotic Testbed for Rendezvous and Optical Navigation: Multi-Source Calibration and Machine Learning Use Cases
• SPEED+: Next-Generation Dataset for Spacecraft Pose Estimation across Domain Gap
• Improved 3D Gaussian Splatting of Unknown Spacecraft Structure Using Space Environment Illumination Knowledge
• SPARK: SPAcecraft Recognition leveraging Knowledge of Space Environment
• Wide-Depth-Range 6D Object Pose Estimation in Space
• Equivariant Filter for Relative Attitude and Target's Angular Velocity Estimation