Look Who it Was Random video

Efficient Multi-camera Visual-Inertial SLAM for Micro Aerial Vehicles

4,632 views · Published 22 August 2016 · 2:20 · Indexed 5 October 2026

Channel: nimbro · 2016 · Science & Technology

Watch on YouTube

We present an extension of the recently introduced monocular
ORB-SLAM for multiple cameras alongside an inertial
measurement unit (IMU). Our main contributions are: Embedding
the multi-camera setup into the underlying graph
SLAM approach that defines the upcoming sparse optimization
problems on several adjusted subgraphs, integration of an
IMU filter that supports visual tracking, and enhancements
of the original algorithm in local map estimation and keyframe
creation. The SLAM system is evaluated on a public stereo
SLAM dataset for flying robots and on a new dataset with
three mounted cameras.
The main advantages of the proposed method are its restricted
computational load, high positional accuracy, and low
number of parameters.

Sebastian Houben, Jan Quenzel, Nicola Krombach, and Sven Behnke:
Efficient Multi-camera Visual-Inertial SLAM for Micro Aerial Vehicles.
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

http://www.ais.uni-bonn.de/papers/IROS_2016_Houben.pdf

More from this channel