Continuous Map-aware Particle Filter in Off-road Environment

DOWNLOAD DOI: 10.62897/COS2024.2-1.175

Author:

Tamás V. Michaletzky, Tamás Kozma, Balázs Kósa, Tamás Fébert, Viktor Remeli, Viktor Tihanyi

Széchenyi István University, Hungary michaletzky.tamas@techtra.hu


Abstract: Accurate localization is a persistent challenge in mobile object tracking, particularly in environ-ments characterized by uneven sensor coverage and inconsistent sensor reliability. This study explores the integration of map-aware particle filters, leveraging map data to enhance posi-tioning accuracy by assigning higher probabilities to positions near defined paths and lower probabilities to off-road locations. Experiments are made in a simulated environment to test the filters under different sensor reliability. A comparative analysis with traditional particle filtering methods demonstrates that the map-aware approach, when appropriately parameterized for the given environmental condition, significantly improves positioning accuracy, emphasizing its potential for enhancing mobile tracking performance in complex environments. The two proposed methods, DMAPF and CMAPF are also compared under multiple scenarios.


 

REFERENCES

Gustafsson , Gunnarsson F., Bergman N., Forssell U., Jansson J., Karlsson R., Nordlund P.-J., 2002, Particle filters for positioning, navigation, and tracking. IEEE Transactions on Signal Processing, 50, 425-437, DOI: 10.1109/78.978396.

Mansour F., Waters D.W., 2013, Map-assisted Kalman filtering. IEEE International Conference on Acous-tics, Speech and Signal Processing, 3208-3212, DOI: 10.1109/ICASSP.2013.6638250.

Peker A.U., Tosun O., Acarman T., 2011, Particle filter vehicle localization and map-matching using map topology. IEEE Intelligent Vehicles Symposium (IV), 248-253, DOI: 1109/IVS.2011.5940473.

Rechy Rormero A., Borges P. V. K., Pfrunder A., Elfes A., 2018, Map-Aware Particle Filter for Localiza-tion. IEEE International Conference on Robotics and Automation (ICRA), 2940-2947, DOI: 10.1109-2018.8460707.

Sang Min Oh, Tariq, B.N. Walker, Dellaert F., 2004, Map-based priors for localization. IEEE/RSJ Interna-tional Conference on Intelligent Robots and Systems (IROS), 2179-2184, DOI: 10.1109/IROS.2004.1389732.

Widyawan, Klepal M., Pesch D., 2007, A Bayesian Approach for RF-Based Indoor Localisation. 4th Inter-national Symposium on Wireless Communication Systems, 133-137, DOI: 1109/ISWCS.2007.4392316.

Widyawan, Klepal , Beauregard S., 2008, A Backtracking Particle Filter for fusing building plans with PDR displacement estimates. 5th Workshop on Positioning, Navigation and Communication, 207-212, DOI: 10.1109/WPNC.2008.4510376.


 

Connection

E-mail address: cos@sze.hu