Tight Coupling of Laser Scanner and Inertial Measurements for a Fully Autonomous Relative Navigation Solution

Andrey Soloviev, Dustin Bates, Frank van Graas

Abstract:	The paper describes a fully autonomous relative navigation solution for urban environments (indoor and outdoor). The navigation solution is derived by combining measurements from a two-dimensional (2D) laser scanner with measurements from inertial sensors. Navigation relies on the availability of structures (lines and surfaces) within the scan range (80 m, typically). Features (e.g. lines and corners) are first extracted from 2D laser scans and then used for position and heading determination. Inertial navigation states (position, velocity and attitude) are applied to predict feature displacements between laser scans, computationally adjust a 2D scan plane for tilting of the scanner platform, and to coast through scans where features extracted from scan images do not provide sufficient information for navigation. Parameters of features extracted from scan images are applied to periodically re-calibrate inertial states in order to reduce the error drift of inertial navigation parameters. The calibration herein uses a Kalman filter that performs a range-domain fusion of laser scanner and inertial data. Indoor and outdoor live data test results are used to demonstrate performance characteristics of the laser/inertial integrated navigation. Positioning at the cmlevel is demonstrated for indoor scenarios where welldefined features and good feature geometry are available. Test data from challenging urban environments show position errors at the meter-level after approx. 200 m of travel (between 0.6% and 0.8% of distance traveled).
Published in:	Proceedings of the 2007 National Technical Meeting of The Institute of Navigation January 22 - 24, 2007 The Catamaran Resort Hotel San Diego, CA
Pages:	1089 - 1103
Cite this article:	Updated citation: Published in NAVIGATION: Journal of the Institute of Navigation
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In