Mobile robot localization via EKF and UKF: A comparison based on real data

In this work we compare the performance of two well known filters for nonlinear models, the Extended Kalman Filter and the Unscented Kalman Filter, in estimating the position and orientation of a mobile robot. The two filters fuse the measurements taken by ultrasonic sensors located onboard the robot. The experimental results on real data show a substantial equivalence of the two filters, although in principle the approximating properties of the UKF are much better. A switching sensors activation policy is also devised, which allows to obtain an accurate estimate of the robot state using only a fraction of the available sensors, with a relevant saving of battery power. Analyzes using EKF and UKF to fuse measurements from ultrasonic sensors in robotics.Shows that the EKF performs as good as the UKF for mobile robot localization.Proposes a sensor switching rule to use only a fraction of the available sensors.Data comes from a real laboratory setting.

[1]  K. Poolla,et al.  Estimation over Wireless Sensor Networks , 2007, 2007 American Control Conference.

[2]  Jeffrey K. Uhlmann,et al.  Unscented filtering and nonlinear estimation , 2004, Proceedings of the IEEE.

[3]  Yuanxin Wu,et al.  Unscented Kalman filtering for additive noise case: augmented versus nonaugmented , 2005, IEEE Signal Processing Letters.

[4]  B. Anderson,et al.  Optimal Filtering , 1979, IEEE Transactions on Systems, Man, and Cybernetics.

[5]  Konrad Reif,et al.  Stochastic Stability of the Extended Kalman Filter With Intermittent Observations , 2010, IEEE Transactions on Automatic Control.

[6]  Jr. J.J. LaViola,et al.  A comparison of unscented and extended Kalman filtering for estimating quaternion motion , 2003, Proceedings of the 2003 American Control Conference, 2003..

[7]  Paul A. Viola,et al.  Robust Real-Time Face Detection , 2001, International Journal of Computer Vision.

[8]  Andrea Garulli,et al.  Comparison of EKF and UKF for Spacecraft Localization via Angle Measurements , 2011, IEEE Transactions on Aerospace and Electronic Systems.

[9]  R. E. Kalman,et al.  A New Approach to Linear Filtering and Prediction Problems , 2002 .

[10]  G. Raiconi,et al.  On the optimal design of the output transformation for discrete-time linear systems , 1991 .

[11]  Dai Hongde,et al.  Performance Comparison of EKF/UKF/CKF for the Tracking of Ballistic Target , 2012 .

[12]  F. Martinelli.  Robot localization: comparable performance of EKF and UKF in some interesting indoor settings , 2008, 2008 16th Mediterranean Conference on Control and Automation.

[13]  Mohinder S. Grewal,et al.  Kalman Filtering: Theory and Practice Using MATLAB , 2001 .

[14]  I. Petrovic,et al.  Extended Kalman filter based mobile robot pose tracking using occupancy grid maps , 2004, Proceedings of the 12th IEEE Mediterranean Electrotechnical Conference (IEEE Cat. No.04CH37521).

[15]  Konrad Reif,et al.  Stochastic stability of the discrete-time extended Kalman filter , 1999, IEEE Trans. Autom. Control..

[16]  P. Muraca,et al.  An adaptive filter using scanning observation with application to a DPS , 1994 .

[17]  Rudolph van der Merwe,et al.  The Unscented Kalman Filter , 2002 .

[18]  Michael Drumheller,et al.  Mobile Robot Localization Using Sonar , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.