A Vehicle Speed Estimation Algorithm Based on Wireless AMR Sensors

Abstract

Abstract. This paper proposes an algorithm for vehicle speed estimation based on the use of anisotropic magnetoresistive (AMR) sensors. Speed estimation relies on matching vehicle magnetic signatures from wireless sensors. A scheme based on edit-distance is developed to automatically matching signatures for the vehicles. Experimental results are presented to show that the proposed speed estimation is viable. IntroductionReal time acquisition of traffic information plays a key role in intelligent transportation systems (ITSs). Wireless sensor networks have lots of potential toward providing an ideal solution for traffic information acquisition, such as their low power, small size, low cost, and high accuracy. At present, the sensors used in the traffic information acquisition include the following types: inductive loop detector [1], image (camera) sensor [2,3], acoustic sensor [4,5], infrared sensor [6], ultrasonic sensor [7], etc. The image sensor acquires an abundance of information, but it is vulnerable to bad weather and nighttime operation. The acoustic sensor and infrared sensor are vulnerable to noise in deployed environments. Magnetic sensors based on magneto resistors have recently been proposed for vehicle detection [8,11] because they are quite sensitive, small and more immune to environmental factors such as rain, wind, snow or fog than sensing systems based on video cameras, ultrasound or infrared radiation.Many algorithms have been proposed for dynamic traffic information acquisition system based on wireless AMR sensor. The PATH program of the University of California, Berkeley [12,13] had first extensively explored of AMR sensor network based vehicle detection system. S.Y. Cheung et al. [13] had explored the applications for vehicle detection, speed estimation, and classification. M. Kang et al. [14] proposed a vehicle detector with an AMR sensor and addresses experimental study carried out to show the detector's characteristics and performance. Experiment results show that the vehicle detection accuracy rate is more than 99%, and the accuracy rate to estimate

Orbit-Engineered Anisotropic Magnetoresistive Effect for Constructing a Magnetic Sensor with Ultrahigh Sensitivity

“…The anisotropic magnetoresistance (AMR) effect is an important physical effect first discovered by Thomson, which refers to the phenomenon that the resistivity of a material varies with the angle between the directions of magnetization and current. Based on the AMR effect, highly sensitive linear sensors have been made for the angle or displacement positioning of objects, which plays important roles in the fields of geomagnetic navigation, numerically controlled machine tools, intelligent manufacturing, and vehicle speed measurement. − With the miniaturization and high integration of devices, the size-shrinking AMR sensor needs to maintain excellent magnetoelectric properties to ensure high linear sensitivity and low noise. However, the AMR ratio of the core thin film reduces greatly as the thickness decreases due to the size effect, resulting in significant destruction in the sensor performance.…”

SenseMag: Enabling Low-Cost Traffic Monitoring using Non-invasive Magnetic Sensing

“…To achieve the goal, magnetic dipole models have been used ans studied in [23], [32], [19], [33], [34], [35], where variations of magnetic fields caused by the moving objects have been measured by magnetic sensors to track the vehicles. In addition to speed measurement, passing vehicle counting has been studied in [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], where the wave pattern matching of magnetic signals has been adopted in the algorithms.…”

A nonuniform demagnetizing field model for both static and dynamic responses of anisotropic magnetoresistive thin film sensors

“…Anisotropic magnetoresistive (AMR) sensors [1] are widely used in geomagnetic navigation [2][3][4], vehicle detection [5][6][7][8] and intelligent manufacturing [9] for its high sensitivity and contactless displacement measurement [10,11]. In real applications, the magnetoresistive strips in AMR sensors usually have large length-width ratio of over 20 [12][13][14][15][16], which can significantly affect the magnetoresistive responses.…”