Evaluation of Vehicle Tracking for Traffic Monitoring Based on Road Surface Mounted Magnetic Sensors

Abstract The aim of this work is to evaluate a vehicle tracking scheme as a means of monitoring traffic on roads. The scheme can be used as a component in a traffic monitoring system which can provide traffic management systems and road maintainers with traffic information. Vehicle tracking is achieved by determining vehicle position, velocity and magnetic moment using a nonlinear weighted least squares method ( NWLS ) on readings from two 3-axes magnetic sensors. The tracking was performed both in simulation and in real life. The traffic monitoring system is composed of two adjacently glue attached wireless sensor nodes, which are placed at a distance of 1 m along the road. A potential misalignment of the sensors due to placement errors is analyzed in simulation and addressed.

[1]  Jens Eliasson,et al.  iRoad - cooperative road infrastructure systems for driver support , 2009 .

[2]  Niklas Wahlström.  Target Tracking using Maxwell’s Equations , 2010 .

[3]  Ondrej Karpis,et al.  Sensor for vehicles classification , 2012, 2012 Federated Conference on Computer Science and Information Systems (FedCSIS).

[4]  Bing W Kwan,et al.  Magnetoresistors for vehicle detection and identification , 1997, 1997 IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation.

[5]  M. Birsan.  Non-linear Kalman filters for tracking a magnetic dipole , 2004 .

[6]  M. Birsan.  Unscented particle filter for tracking a magnetic dipole target , 2005, Proceedings of OCEANS 2005 MTS/IEEE.

[7]  Fredrik Gustafsson,et al.  Toward Autonomous Collision Avoidance by Steering , 2007, IEEE Transactions on Intelligent Transportation Systems.

[8]  Fredrik Gustafsson,et al.  Magnetometers for tracking metallic targets , 2010, 2010 13th International Conference on Information Fusion.

[9]  Sing Yiu Cheung,et al.  Traffic Surveillance by Wireless Sensor Networks: Final Report , 2007 .

[10]  Roger Alimi,et al.  Ferromagnetic Mass Localization in Check Point Configuration Using a Levenberg Marquardt Algorithm , 2009, Sensors.

[11]  David Törnqvist,et al.  Statistical Fault Detection with Applications to IMU Disturbances , 2006 .