Login / Signup

Star-Tracker Algorithm for Smartphones and Commercial Micro-Drones.

Revital MarbelBoaz Ben-MosheRoi Yozevitch
Published in: Sensors (Basel, Switzerland) (2020)
This paper presents a star-tracking algorithm to determine the accurate global orientation of autonomous platforms such as nano satellites, U A V s, and micro-drones using commercial-off-the-shelf ( C O T S ) mobile devices such as smartphones. Such star-tracking is especially challenging because it is based on existing cameras which capture a partial view of the sky and should work continuously and autonomously. The novelty of the proposed framework lies both in the computational efficiency and the ability of the star-tracker algorithm to cope with noisy measurements and outliers using affordable C O T S mobile platforms. The presented algorithm was implemented and tested on several popular platforms including: Android mobile devices, commercial-micro drones, and Raspberry Pi. The expected accuracy of the reported orientation is [0.1°,0.5°].
Keyphrases
  • machine learning
  • deep learning
  • neural network
  • high resolution