Motion Estimation using Multiple Non-Overlapping Cameras for Small Unmanned Aerial Vehicles - Robotics Institute Carnegie Mellon University

Motion Estimation using Multiple Non-Overlapping Cameras for Small Unmanned Aerial Vehicles

Junsik Kim, Myung Hwangbo, and Takeo Kanade
Conference Paper, Proceedings of (ICRA) International Conference on Robotics and Automation, pp. 3076 - 3081, May, 2008

Abstract

An imaging sensor made of multiple light-weight non-overlapping cameras is an effective sensor for a small unmanned aerial vehicle that has strong payload limitation. This paper presents a method for motion estimation by assuming that such a multi-camera system is a spherical imaging system (that is, the cameras share a single optical center). We derive analytically and empirically a condition for a multi-camera system to be modeled as a spherical camera. Interestingly, not only does the spherical assumption simplify the algorithms and calibration procedure, but also motion estimation based on that assumption becomes more accurate.

BibTeX

@conference{Kim-2008-9953,
author = {Junsik Kim and Myung Hwangbo and Takeo Kanade},
title = {Motion Estimation using Multiple Non-Overlapping Cameras for Small Unmanned Aerial Vehicles},
booktitle = {Proceedings of (ICRA) International Conference on Robotics and Automation},
year = {2008},
month = {May},
pages = {3076 - 3081},
keywords = {Camera Motion Estimation, multiple cameras, Spherical assumption},
}