An overview on sensor map based localization for automated driving

In the urban environment or under complex traffic situation, conventional localization methods like GPS, dead reckoning or SLAM are not precise enough for autonomous driving. So high-precision localization for intelligent vehicles becomes a hot problem. Among all the high-precision localization methods, the category based on sensor maps outperforms others because of its accuracy and real-time property. In this paper, a detailed review and comparison of sensor-map based localization algorithms and the key techniques in it are presented. First we categorize the localization methods according to the sensors used and types of map. We summarize there are three key problems in localization: 1) model of maps: the form of map according to different environments; 2) high-precision map generation: methods of generating such a high-precision map; 3) localization: methods of estimating vehicle's position and pose from the detected environment information aligned with the maps generated in advance. In light of the above, in this paper, the state-of-the-art techniques of high-precision localization for intelligent vehicles are reviewed in detail, and the corresponding pro and con are discussed.

[1]  Peter Kulchyski and , 2015 .

[2]  Andriy Myronenko,et al.  Point Set Registration: Coherent Point Drift , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Ming Yang,et al.  Road DNA based localization for autonomous vehicles , 2016, 2016 IEEE Intelligent Vehicles Symposium (IV).

[4]  Jihong Lee,et al.  Multi-robot cooperative localization with optimally fused information of odometer and GPS , 2007, 2007 International Conference on Control, Automation and Systems.

[5]  Paul J. Besl,et al.  A Method for Registration of 3-D Shapes , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Ignacio Parra,et al.  Accurate Global Localization Using Visual Odometry and Digital Maps on Urban Environments , 2012, IEEE Transactions on Intelligent Transportation Systems.

[7]  Otman A. Basir,et al.  GPS Localization Accuracy Classification: A Context-Based Approach , 2013, IEEE Transactions on Intelligent Transportation Systems.

[8]  Klaus C. J. Dietmayer,et al.  Localization based on region descriptors in grid maps , 2014, 2014 IEEE Intelligent Vehicles Symposium Proceedings.

[9]  Patrick Weber,et al.  OpenStreetMap: User-Generated Street Maps , 2008, IEEE Pervasive Computing.

[10]  Paul Newman,et al.  Continuous vehicle localisation using sparse 3D sensing, kernelised rényi distance and fast Gauss transforms , 2013, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[11]  Ming Yang,et al.  Ground-Texture-Based Localization for Intelligent Vehicles , 2009, IEEE Transactions on Intelligent Transportation Systems.

[12]  Peter Biber,et al.  The normal distributions transform: a new approach to laser scan matching , 2003, Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453).

[13]  Paul Newman,et al.  Work smart, not hard: Recalling relevant experiences for vast-scale but time-constrained localisation , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[14]  Hiroshi Murase,et al.  Single camera vehicle localization using SURF scale and dynamic time warping , 2014, 2014 IEEE Intelligent Vehicles Symposium Proceedings.

[15]  Andreas Nüchter,et al.  Automatic Appearance-Based Loop Detection from 3 D Laser Data Using the Normal Distributions Transform , 2009 .