Analysis of Methods for Long Vehicles Speed Estimation Using Anisotropic Magneto-Resistive (AMR) Sensors and Reference Piezoelectric Sensor

Abstract

With rapidly increasing traffic occupancy, intelligent transportation systems (ITSs) are a vital feature for urban areas. This paper analyses methods for estimating long (L > 10 m) vehicle speed and length using a self-developed system, equipped with two anisotropic magneto-resistive (AMR) sensors, and introduces a method for verifying the results. A well-known cross-correlation method of magnetic signatures is not appropriate for calculating the vehicle speed of long vehicles owing to limited resources and a long calculation time. Therefore, the adaptive signature cropping algorithm was developed and used with a difference quotient of a magnetic signature. An additional piezoelectric polyvinylidene fluoride (PVDF) sensor and video camera provide ground truth to evaluate the performances. The prototype system was installed on the urban road and tested under various traffic and weather conditions. The accuracy of results was evaluated by calculating the mean absolute percentage error (MAPE) for different methods and vehicle speed groups. The experimental result with a self-obtained data set of 600 unique entities shows that the average speed MAPE error of our proposed method is lower than 3% for vehicle speed in a range between 40 and 100 km/h.

Research of Distorted Vehicle Magnetic Signatures Recognitions, for Length Estimation in Real Traffic Conditions

“…In this paper, we use data collected from a traffic monitoring station, as explained in paper [ 9 ]. Additionally, this station was upgraded with accelerometer sensors, which were installed on the same PCB (printed circuit board) with magnetometers ( Figure 6 ).…”

“…A variety of sensing techniques are used for increasing road traffic safety. One approach is based on equipping vehicles [ 13 , 14 , 15 ] with new sensors and another by upgrading road infrastructure with traffic monitoring systems, from inductive loops to microwave radars and optical fibers [ 4 , 5 , 6 , 7 , 8 , 9 ]. Although many methods can be used for intelligent traffic systems, there are few key technologies widely accepted as reliable: surveillance camera systems, inductive loops and magnetic field sensors.…”

“…Although many methods can be used for intelligent traffic systems, there are few key technologies widely accepted as reliable: surveillance camera systems, inductive loops and magnetic field sensors. Lately, extensive works have been put on developing intelligent traffic monitoring stations based on magnetic sensors [ 7 , 9 ]. Well-known and mature inductive loop technology were compared with magnetic field sensors in [ 2 ].…”

“…During the previous research [ 9 ], it was noted that at certain conditions speed estimation method from the pair of magnetic signatures gave highly faulty results. After the analysis, it was deducted that most of the time, incorrect results appeared once the wheel of the vehicle passed very close or directly above the sensor.…”

“…The first step for this kind of system is to collect useful road traffic information: vehicle volume, speed, length, type and direction [ 1 , 2 , 3 ]. There are a number of vehicles sensing technologies from simple inductive loops to complex microwave radars and high-speed cameras [ 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 ], and all of them have unique advantages and drawbacks.…”