DEVICE WITH AMR SENSORS IMPLEMENTING IT

FIELD OF THE INVENTION

The invention relates to the field of land vehicle speed measurement, in particular, speed determination by using several magnetic sensors placed near moving vehicles and by computing the time difference (delay) of the occurrence of the signal in different sensors. The new method for computing the time difference of the occurrence of signals described reduces the computing time, allows more efficient use of energy.

DESCRIPTION OF THE RELATED ART

This invention provides a new method for computing land vehicle speed using AMR sensors and device implementing that method. At least two AMR sensors that measure changes in the magnetic field of the Earth in time (magnetic signatures) are placed in the pavement or in another place close to passing vehicles (at a particular known distance from each other (e. g. 20-30 cm) when the vehicle is moving above/near AMR sensors. After measuring the time difference between dependencies of the magnetic field on time by different sensors, the vehicle speed is determined. The more points that form the signature, the more and more accurate information is obtained from the measurement, on the other hand, the large amount of data lengthens the data processing time and uses more energy. When controlling traffic flows, it is needed to fast measure the vehicle speed, to use fast data processing and transmission without loss of information quality.

This invention provides a new method for fast computing of vehicle speed. When computing speed, the use of the difference of “mass centres” to compute the cross correlation function maximum allows reducing the volume of computations from 50 to 1000 times. The mass centres Ri and R2 are computed from signatures of both AMR sensors. The computed difference of mass centres R1 and R2 is used as the initial difference when computing the cross-correlation coefficient of signatures. Then, the cross-correlation coefficient maximum is searched using the set of shifts of R2-R1 ± d. 2d - width of the maximum search area. In this way, the cross-correlation computing requiring multiple computational resources is performed only at points of 2d range. The new computation method is applied by measuring the average speed on a particular road stretch using two meters at a known distance from each other with technical means for vehicle identification.

Document US6208268 (B1 ), (published on 27 March 2001 ) provides a similar system for determining the vehicle speed: two magnetic sensors arranged along the direction of movement of the vehicles detects a change in the magnetic field, compares the difference in time measured by different sensors, the vehicle speed is computed from the measured difference. The cited document provides a simple computation method for determining the speed without taking into account the time taken by the computational means to compute the vehicle speed. Nothing is mentioned about using energy efficiently. It is unclear whether the method is accurate enough or whether the computation is suitable for reliably determining the speed with acceptable error.

Document US2013057264 (A1 ) (published on 7 March 2013) discloses a similar method for determining the vehicle speed using magnetic sensors, the magnetic signature they generate, and the time computation between signals generated by the different sensors. The document provides a computation method using Anderson's function to compute the cross-correlation between signals. The document does not provide data on the computation speed, but it is also clear from the computation method that the computation method is not designed for fast processing of large amounts of measurement data while maintaining the efficiency of speed measurement.

Summarizing documents of the prior related art documents, the following

Deficiencies can be distinguished:

- the computation method provided does not consider the large amount of data and the time it takes to process such data; not suitable for fast processing of measurement data;

- the measurement and computation method does not ensure acceptable speed measurement errors;

- the cited documents do not mention the possibility of computing the average vehicle speed on a particular road stretch of known length;

- nothing is mentioned about energy efficiency.

This invention provides a technical solution that does not have the above deficiencies.

THE SUMMARY OF THE INVENTION When measuring the vehicle speed with two magnetic sensors, the time difference between dependencies measured by two different sensors (magnetic signatures) is determined. The higher the number of points in the signature, the more accurate information can be obtained from the measurements. On the other hand, a high number of points lengthens the processing time and increases the energy consumption for computations. When it is necessary to accurately identify a vehicle that exceeds the set speed, the data must be processed extremely fast to allow the capture camera (or other vehicle identification means) to capture the offender.

This invention provides a new fast method for determining the speed and device implementing it. The method for determining the speed uses the difference of “mass centres” as the initial information for computing the cross-correlation coefficient. Then, the cross-correlation coefficient maximum is searched using a set of shifts, the centre of which is the difference of “mass centres”. The computation method provided not only shortens the computation time, but also allows more efficient use of the energy used by the speed meter.

In addition, using two speed meters (having vehicle identification means) arranged at a known distance, the average speed of the vehicle on the stretch between these two meters can be determined.

THE PREFERRED EMBODIMENT

Many technical means are used to measure the speed of land vehicles; one of methods to determine the vehicle speed is by using magnetism phenomena. One method to set speed is by using AMR (Anisotropic Magnetoresistive) sensors. AMR sensors are located near moving vehicles, e. g. they can be placed on the road, under the road pavement, or near the lane. When a vehicle having steel parts interacting with magnetic fields appears above/near sensors, the natural magnetic field of the Earth is distorted in the proximity of the sensor. AMR sensors measure the change in the magnetic field of the Earth. If more than one AMR sensor is placed along the direction of vehicle movement, the moving vehicle initially changes the magnetic field measured by one AMR sensor and then the magnetic field measured by another AMR sensor (in the direction of vehicle movement behind the first AMR sensor). The dependency on time of the change in the magnetic field of the same vehicle in the different sensors is similar, so knowing the distance between these AMR sensors and setting the time at which the measured dependencies differ, the vehicle speed, vehicle length (where appropriate), other settings can be determined.

The magnetic sensor transforms the time dependency of the magnetic field in the three-dimensional space into an electrical signal. The higher resolution (i.e. the number of points per time unit) of said dependencies (or a signal, signature) allows to more accurately determine vehicle speed, to measure other parameters. However, a higher resolution, i. a higher amount of data overloads the computing hardware that processes the data. More time is needed to process the data, and if the vehicle speed measurement is for the purpose of determining and capturing vehicles which are exceeding the speed (e. g. when taking pictures), the long processing time may cause the offender to be far from where he can be captured by the camera.

For the purposes of this description, the terms “signal”, “signature” mean the change in the magnetic field and/or change in the corresponding electrical parameter in time, the dependency on time of the parameter.

The present invention provides a method for reducing the amount of computations efficiently and without losing the necessary information, so that data processing (e. g., computation of the cross-correlation function maximum between signals) occurs faster.

One of known methods to compute the difference of the occurrence in time of two signals is the cross-correlation computation. In the present invention, the scope of the cross-correlation function computations can be reduced by using the difference of “mass centres” to compute the cross-correlation function maximum. The computed difference of mass centres is used as the initial difference in the search for the location of the cross-correlation coefficient maximum of signatures.

The method for computing vehicle speed using AMR sensors provided by the present invention has the following steps:

- Using at least two AMR sensors placed on the road/into the pavement, near moving vehicles and at a distance L in the direction of moving vehicles, along the roadway, continuously measuring the magnetic field of the Earth in real time;

- When the vehicle is moving above/near the sensor, the dependency in time of the change in the magnetic field of the Erath (magnetic signature) is detected and the data of at least two AMR sensors is recorded into the memory of a microcontroller (or other computing device);

- The Earth’s magnetic field distortion module is computed from Earth's magnetic field three-dimensional space components x, y and z measured by sensors; - Mass centres R1 and R2 of the magnetic signatures obtained by at least two AMR sensors are computed:

where Si and S2 - discrete signal modules, N - number of points depending on threshold level. Threshold level - optionally set signal value from which it is assumed that the signal value exceeds the noise level from which all values of the signal, signature is used to derive the computation results.

- The difference between mass centres R2-R1 of magnetic signatures is computed.

- With this difference, both signals are cross-correlated between R2-Ri-d and R2- R1 + d. d - half of the search range.

- The difference between both signals D corresponding the cross-correlation coefficient maximum of magnetic signatures is found.

- The vehicle speed is computed

where L - distance between sensors; D - time difference between “magnetic signatures” (expressed in discrete references); t _s - duration between two adjacent references.

The computed difference between the mass centres Ri and R2 is used as the initial difference for computing the cross-correlation coefficient of signatures. Then, the cross-correlation coefficient maximum is searched using the set of shifts of R2-R1 ± d. 2d - width of the maximum search area. In this way, the cross-correlation computation requiring multiple computational resources is performed only at points of the range 2d.

If vehicle speed measurement is performed to identify vehicles that exceed the speed at that location, then the vehicle identification method is required. The use of the present invention can be achieved by conventional vehicle identification means when the vehicle registration number is captured by image cameras. The image capture camera is positioned at a distance from the magnetic speed meters to enable the speed measuring device to process the data to determine the vehicle speed, to transmit the signal so that the image capture camera would be ready to capture and would capture the vehicle.

This invention provides a method for determining vehicle speed by known electronic computing means as usual having a processor (s), microcontroller (s) or other means for computational, logical operations; instant and/or permanent memory modules for data storage; communication means for communication between internal elements and interfaces with external elements and other technical means specific to devices performing such or similar computations. A part of the device performing the measurement of the change in the magnetic field of the Earth is connected to the part of the device performing the measured data processing. Magnetoresistive sensors are the main component of this part of the device; in this case, AMR sensors are used. AMR sensors measure a three-dimensional magnetic field. The data measured by them are transmitted by wired or wireless (e. g. radio) means for data transmission to the part of the device performing the data processing.

A possible application of the described invention is for determining the average vehicle speed. In this embodiment of the invention, the instantaneous speed meters (with vehicle identification means) described must be spaced apart from each other in order to compute the average speed (e. g., 1 km). The first instantaneous speed meter placed in the drive direction capture the instantaneous speed and then uses technical means to accurately identify the vehicle (e. g. an image camera). In one embodiment, said magnetic signature of the vehicle may be used to identify vehicles. The magnetic signature (in a form of an electrical signal) will be different for different vehicles; the form depends on the amount of ferromagnetic materials, the shape, the magnetization of the ferromagnetic materials in the vehicle and other elements of the vehicle that cause the magnetic field distortion. A known method of accurately identifying vehicles is to capture state vehicle registration numbers by image capture cameras located behind magnetic speed meters in the direction of vehicle movement in the present invention. The average speed of the vehicle between these two speed meters shall be computed by capturing the vehicle at the first vehicle measuring point and then at the second measuring point and after the time taken for the vehicle to drive between the first and second speed meters (between which is a known distance ). The magnetic signature of the vehicle which exceeds the speed and the captured state number are transmitted to the next speed meter, linking the two gauges to a common system for measuring the average speed. The second gauge uses the magnetic signature of the vehicle to capture the correct vehicle using the image capture camera.

In order to illustrate and describe the invention, the description of the preferred embodiments is presented above. This is not a detailed or restrictive description to determine the exact form or embodiment. The above description should be viewed more than the illustration, not as a restriction. It is obvious that specialists in this field can have many modifications and variations. The embodiment is chosen and described in order to best understand the principles of the present invention and their best practical application for the various embodiments with different modifications suitable for a specific use or implementation adaptation. It is intended that the scope of the invention is defined by the definition added to it and its equivalents, in which all of these definitions have meaning within the broadest limits, unless otherwise stated.

In the embodiments described by those skilled in the art, modifications may be made without deviating from the scope of this invention as defined in the following definition.