Asynchronous Kalman Filter for Event-Based Star Tracking
Abstract
High precision tracking of stars both from ground and from orbit is a vital capability that enables autonomous alignment of both satellite and ground-based telescopes. Event cameras provide high-dynamic range, high temporal resolution, low latency asynchronous “event” data that captures illumination changes in a scene. Such data is ideally suited for estimating star motion since it has minimal image blur and can capture low-intensity changes in irradiation typical of astronomical observations. In this work, we propose a novel Asynchronous Event-based Star Tracker that processes each event asynchronously to update a Kalman filter that estimates star position and velocity in an image. The proposed tracking method is validated on real and simulated data and shows state-of-that-art tracking performance against existing approach.
Cite
Text
Ng et al. "Asynchronous Kalman Filter for Event-Based Star Tracking." European Conference on Computer Vision Workshops, 2022. doi:10.1007/978-3-031-25056-9_5Markdown
[Ng et al. "Asynchronous Kalman Filter for Event-Based Star Tracking." European Conference on Computer Vision Workshops, 2022.](https://mlanthology.org/eccvw/2022/ng2022eccvw-asynchronous/) doi:10.1007/978-3-031-25056-9_5BibTeX
@inproceedings{ng2022eccvw-asynchronous,
title = {{Asynchronous Kalman Filter for Event-Based Star Tracking}},
author = {Ng, Yonhon and Latif, Yasir and Chin, Tat-Jun and Mahony, Robert E.},
booktitle = {European Conference on Computer Vision Workshops},
year = {2022},
pages = {66-79},
doi = {10.1007/978-3-031-25056-9_5},
url = {https://mlanthology.org/eccvw/2022/ng2022eccvw-asynchronous/}
}