Online Supervised Training of Spaceborne Vision during Proximity Operations using Adaptive Kalman Filtering
arXiv:2309.11645 · doi:10.1109/ICRA57147.2024.10610138
Abstract
This work presents an Online Supervised Training (OST) method to enable robust vision-based navigation about a non-cooperative spacecraft. Spaceborne Neural Networks (NN) are susceptible to domain gap as they are primarily trained with synthetic images due to the inaccessibility of space. OST aims to close this gap by training a pose estimation NN online using incoming flight images during Rendezvous and Proximity Operations (RPO). The pseudo-labels are provided by adaptive unscented Kalman filter where the NN is used in the loop as a measurement module. Specifically, the filter tracks the target's relative orbital and attitude motion, and its accuracy is ensured by robust on-ground training of the NN using only synthetic data. The experiments on real hardware-in-the-loop trajectory images show that OST can improve the NN performance on the target image domain given that OST is performed on images of the target viewed from a diverse set of directions during RPO.
arXiv commentAccepted to ICRA'2024. Final revised version
References in corpus (10)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Albumentations: fast and flexible image augmentations
- Towards Deep Learning Models Resistant to Adversarial Attacks
- Model Adaptation: Unsupervised Domain Adaptation without Source Data
- Revisiting Batch Normalization For Practical Domain Adaptation
- Satellite Pose Estimation Challenge: Dataset, Competition Design and Results
- Robust Multi-Task Learning and Online Refinement for Spacecraft Pose Estimation across Domain Gap
- Adaptive Neural Network-based Unscented Kalman Filter for Robust Pose Tracking of Noncooperative Spacecraft
- Adaptive and Dynamically Constrained Process Noise Estimation for Orbit Determination
- Analytical Process Noise Covariance Modeling for Absolute and Relative Orbits