Reproduce “Star Tracking using an Event Camera” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Tat-Jun Chin, Samya Bagchi, Anders Eriksson, Andre van Schaik
Researchers from The University of Adelaide, University of Queensland, and Western Sydney University developed an algorithmic pipeline for star tracking using event cameras. This work demonstrates the feasibility of achieving high-precision spacecraft attitude estimation with event cameras, achieving an RMSE angular error typically at or below 1 degree, comparable to commercial star trackers.
Key insights
- Event cameras, despite their noisy and asynchronous data, can be effectively leveraged for high-precision star tracking if coupled with a robust, specialized algorithmic pipeline.
- The proposed 'augmented rotation averaging' formulation, which incorporates sparse but accurate absolute rotation measurements, effectively mitigates drift that is common in sequential pose estimation from relative measurements.
- Combining robust relative pose estimation (via Trimmed ICP), augmented rotation averaging, and a rotation-only bundle adjustment provides a comprehensive solution for fusing event data into accurate attitude estimates.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.