A Reduced Camera SLAM Approach for Indoor and Outdoor Navigation Using Laser Information for Landmark Initialization and Relative Motion Information

Manuel Popp, Jamal Atman, Georg Scholz, Jan Ruppelt, Gert F. Trommer

Peer Reviewed

Abstract:	In order to allow Micro Areal Vehicles (MAVs) to autonomously explore urban areas a precise knowledge of the MAV’s kinematic state is essential. Thus, an integrated navigation system is presented in this paper. It provides a precise self-localization in both; indoor and outdoor areas. In order to circumvent the disadvantages of common image or laser based navigation systems, the information from both sensors are fused together on an early stage. That means, laser rangefinder’s distance measurements are referred to image pixels. This results in 2.5-D images. But in order to do so, the pose between both sensors has to be known. Therefore a process that allows a calibration of the sensors’ extrinsic parameters is presented in this paper. Our navigation filter bases on a reduced SLAM approach combined with relative motion measurements. Both elements benefit from the 2.5-D images. The scale factor of the relative translational motion can be simply measured, while the landmarks can be initialized instantaneously with precisely known depth. This results in a navigation system that provides precise and long-term stable navigation solutions, independently from structure and condition of the environment.
Published in:	Proceedings of the 2016 International Technical Meeting of The Institute of Navigation January 25 - 28, 2016 Hyatt Regency Monterey Monterey, California
Pages:	647 - 656
Cite this article:	Popp, Manuel, Atman, Jamal, Scholz, Georg, Ruppelt, Jan, Trommer, Gert F., "A Reduced Camera SLAM Approach for Indoor and Outdoor Navigation Using Laser Information for Landmark Initialization and Relative Motion Information," Proceedings of the 2016 International Technical Meeting of The Institute of Navigation, Monterey, California, January 2016, pp. 647-656. https://doi.org/10.33012/2016.13452
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In