Abstract：Monte Carlo localization scheme with a scan matching algorithm is suggested as a robust localization method for mobile robots to accomplish their tasks autonomously.A normal distributions transform(NDT) which is a new approach to laser scan matching is applied to the scan matching algorithm,and this scan matching method transforms the discrete set of 2D points reconstructed from a single scan into a piecewise continuous and differentiable probability distribution defined on the 2D plane,which can be used to match another scan by using Hessian Matrix.Thereby,no point to point correspondences have to be established.Experimental results show that the robot is able to accomplish localization autonomously in an indoor environment using the natural environmental characteristics.
[1] Clark F O. Probabilistic self-localization for mobile robots[J]. IEEE Transactions on Robotics and Automation, 2000, 16(1):55-66. [2] Duckett T, Nehmzow U. Mobile robot self-localisation using occupancy histograms and a mixture of Gaussian location hypotheses[J]. Robotics and Autonomous Systems, 1999, 34(2/3):117-129. [3] Gutmann J S, Schlege C. AMOS:comparison of scan matching approaches for self-localization in indoor environments[A]. Proceedings of the Euromicro Workshop on Advanced Mobile Robots(EUROBOT96)[C]. Kaiserslautern, Germany:1996. 61-67. [4] Bengtsson O, Baerveldt A-J. An improvement of the COX scan matching algorithm for dynamic environments[EB/OL]. http:∥www2.hh. se/staff/bools/kkprojβengtsson_baerveldt ecmr03. pdf, 2003. [5] Bengtsson O, Baerveldt A-J. Localization in changing environmentsestimation of a covariance matrix for the IDC algorithm[J]. Robotics and Autonomous Systems, 2003, 44(1):29-40. [6] Crowley J L, Wallner F, Schiele B. Position estimation using principal components of range data[J]. Robotics and Autonomous Systems,1998, 23(4):267-276. [7] Biber P, Straβer W. The normal distributions transform:a new approach to laser scan matching[A]. Proceedings of the 2003 IEEE/RJS International Conference on Intelligent Robots and Systems[C]. Las Vegas, USA:IEEE, 2003. 2743-2748. [8] Gutmann J-S, Fox D. An experimental comparison of localization methods continued[A] Proceedings of the 2002 IEEE/RSJ International Conference on Intelligent Robots and Systems[C]. Lausanne,Switzerland:IEEE, 2002. 454-459. [9] Thrun S, Fox D, Burgard W, et al. Robust Monte Carlo localization for mobile robots[J]. Artificial Intelligencce, 2001,128(1-2):99-141.
