Method and Device for Vehicle Trajectory Tracking and False Trajectory Detection Based on Multiple Geomagnetic Sensors

Abstract

This invention discloses a method and apparatus for vehicle trajectory tracking and false trajectory identification based on multiple geomagnetic sensors. The method includes: geomagnetic sensors deployed on both sides of a road acquiring data of passing vehicles; a base station storing and cleaning the location of the geomagnetic sensors and the vehicle data acquired by those sensors, and packaging and sending the data to a data processing center at fixed time intervals; within a fixed time period, the data processing center processes the received vehicle data acquired by the geomagnetic sensors to generate a set of vehicle trajectories; based on the time difference between adjacent vehicle trajectories passing the same geomagnetic sensor, it determines whether there is a false trajectory among adjacent vehicle trajectories. If so, the adjacent vehicle trajectories are weighted and merged into a single vehicle trajectory, the vehicle trajectory set is updated, and the updated vehicle trajectory set is output; if not, the vehicle trajectory set is output. This invention can output reliable vehicle trajectory information.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of intelligent transportation, and particularly relates to a vehicle trajectory tracking method and device based on multiple geomagnetic sensors and a false trajectory judgment method and device. BACKGROUND

[0002] Traffic is the main driving force for urban development. The rapid increase in the number of cars has led to increasingly severe traffic congestion, traffic accidents, and environmental pollution, which seriously hinders the social and economic development of cities. In order to promote the development of intelligent traffic systems (ITS) and smart cities, diversified Internet of Things sensors are deployed in the traffic system to accurately obtain vehicle information on the road. Among them, geomagnetic sensors are gradually gaining attention in ITS due to their low cost and strong anti-interference capability. Current research on the application of geomagnetic sensors mainly focuses on using a few simple geomagnetic sensors to detect traffic volume, vehicle speed, and other information. As research on geomagnetic sensors deepens, the application of multiple geomagnetic sensors in trajectory tracking systems is attracting attention, and research on tracking the trajectories of passing vehicles is also gaining attention. When using multiple geomagnetic sensors for vehicle trajectory tracking, a key problem is that when a large vehicle that can cause large magnetic field fluctuations drives on a lane, the geomagnetic sensors deployed along the side of the lane may detect two pieces of data for the large vehicle, resulting in the output of two vehicle trajectories during subsequent trajectory tracking, which leads to the generation of false trajectories and affects the credibility of the output trajectories.

[0003] In related technologies, the Ministry of Transport Road Network Monitoring and Emergency Disposal Center proposed a single-lane vehicle trajectory association method based on radar and camera fusion in its co-applied patent document "Vehicle trajectory tracking method and system based on radar and video fusion" (Application No. 202210867558.4, Application Publication No. CN115327527A). This method synchronizes the time stamps of radar devices and camera devices to achieve the goal of target trajectory tracking. By obtaining the time stamp of radar information and the time stamp of camera information, the method realizes the association of sensor data from both sources, performs data fusion, and generates trajectories. The shortcomings of this method are that the deployment environment of cameras and radars is special, and they are usually installed on gantry cranes, making it difficult to deploy on a large scale, with high costs. In addition, the detection accuracy is affected by weather, and it is difficult to detect vehicles in harsh environments.

[0004] Xi'an University of Electronic Science and Technology in its application of the patent document "Single lane vehicle trajectory correlation method based on multiple geomagnetic sensors" (application number: CN202211663151.6, publication number: CN116386345A) proposes a method of deploying multiple geomagnetic sensors at equal intervals on the lane line on one side of a single lane, detecting the timestamp of the vehicle's arrival using multiple geomagnetic sensors and uploading it to a computing device, calculating the predicted timestamp of the vehicle's arrival at the geomagnetic sensor and the correlation method of matching the two diagrams of the geomagnetic sensor measurement time through the computing device, and obtaining the vehicle trajectory information passing through the road; However, this method does not consider the case that large vehicles passing through may cause the geomagnetic sensor to falsely report multiple detection data, resulting in false vehicle trajectories in the final correlation results, affecting the feasibility of its use.

[0005] Therefore, it is urgent to provide a vehicle trajectory tracking and false trajectory judgment method to improve the reliability of vehicle trajectory tracking. SUMMARY

[0006] In order to solve the above-mentioned problems existing in the prior art, the present application provides a vehicle trajectory tracking and false trajectory judgment method and device based on multiple geomagnetic sensors. The technical problem to be solved by the present application is solved by the following technical scheme:

[0007] In a first aspect, the present application provides a vehicle trajectory tracking and false trajectory judgment method based on multiple geomagnetic sensors, comprising:

[0008] The geomagnetic sensors deployed on both sides of the road obtain the data of the passing vehicles, including the magnetic field feature data and the corresponding timestamp, and send the magnetic field feature data and the corresponding timestamp of the vehicles to the base station;

[0009] The base station saves and cleans the position of the geomagnetic sensor and the data of the vehicles obtained by the geomagnetic sensor, and sends them to the data processing center at a fixed time interval;

[0010] In a fixed time period, the data processing center processes the received data of the vehicles obtained by the geomagnetic sensor to generate a vehicle trajectory set;

[0011] According to the time difference of adjacent vehicle trajectories passing through the same geomagnetic sensor, it is judged whether there is a false trajectory in the adjacent vehicle trajectories. If there is, the adjacent vehicle trajectories are weighted and fused to merge into one vehicle trajectory, update the vehicle trajectory set, and output the updated vehicle trajectory set. If not, output the vehicle trajectory set.

[0012] In a second aspect, the present application further provides a vehicle trajectory tracking and false trajectory judgment device based on multiple geomagnetic sensors, comprising:

[0013] data acquisition module, for acquiring data of passing vehicles by geomagnetic sensors deployed on both sides of the road, the data of the vehicles including magnetic field feature data and corresponding time stamps, and sending the magnetic field feature data of the vehicles and the corresponding time stamps to a base station;

[0014] data processing center, for the base station to save and clean the positions of the geomagnetic sensors and the data of the vehicles acquired by the geomagnetic sensors, and send the data to the data processing center at a fixed time interval;

[0015] vehicle trajectory generation module, for the data processing center to process the data of the vehicles acquired by the geomagnetic sensors received within a fixed time period, and generate a vehicle trajectory set;

[0016] false trajectory judgment module, for judging whether there is a false trajectory in adjacent vehicle trajectories according to the time difference of the adjacent vehicle trajectories passing through the same geomagnetic sensor, if there is, then the adjacent vehicle trajectories are fused by weighting and merged into one vehicle trajectory, the vehicle trajectory set is updated, and the updated vehicle trajectory set is output, if there is not, then the vehicle trajectory set is output.

[0017] The present application has the following advantages:

[0018] The present application provides a vehicle trajectory tracking and false trajectory judgment method and device based on multiple geomagnetic sensors, which first deploys geomagnetic sensors on the roadside to collect data of vehicles passing through the sensors; secondly, generates original vehicle trajectories by using detection data of multiple geomagnetic sensors in a data center; then, detects false trajectories caused by false detection data of the sensors in the generated trajectories on the basis of the generated trajectories by using a Bayesian filtering method; finally, fuses data of false trajectories caused by false detection data with real trajectories, eliminates false false trajectories, and finally outputs credible vehicle trajectory information. The method can be used to realize intelligent highway.

[0019] The present application will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is a flowchart of the vehicle trajectory tracking and false trajectory judgment method based on multiple geomagnetic sensors provided by the embodiments of the present application;

[0021] Figure 2 is a schematic diagram of the geomagnetic sensor deployment diagram provided by the embodiments of the present application;

[0022] Figure 3 is a schematic diagram of the data association between the existing vehicle trajectory set and the data of the vehicles acquired by the geomagnetic sensors provided by the embodiments of the present application;

[0023] Figure 4 is a flowchart of judging a false trajectory provided by an embodiment of the present application. DETAILED DESCRIPTION

[0024] The present application will be further described in conjunction with specific embodiments, but the embodiments of the present application are not limited thereto.

[0025] The present application aims at the deficiencies of the prior art, and provides a vehicle trajectory tracking and false trajectory judgment method based on multiple geomagnetic sensors, to solve the problems of current vehicle trajectory tracking algorithm complexity, high cost of cameras and radars and other sensors, weather influence, difficulty in large-scale deployment, and false trajectories in some current vehicle trajectory tracking systems based on multiple geomagnetic sensors.

[0026] The technical idea for achieving the purpose of the present application is that the vehicle data used is derived from geomagnetic sensors deployed on road lane lines, which does not affect normal traffic and is easy to maintain, and can solve the problem of destroying the road surface and affecting normal traffic in the existing deployment scheme, and also brings trouble to subsequent maintenance. The geomagnetic sensor has low cost and is not affected by weather, which is easy to deploy on a large scale and can solve the problem of high cost of cameras and radars and other sensors and serious weather influence. In addition, the present application can output reliable vehicle trajectories in time by real-time processing of the vehicle passing time collected by the geomagnetic sensor in the data processing center, which does not bring pressure to the geomagnetic sensor computing capability.

[0027] Please refer to Figure 1 , Figure 1 is a flowchart of the vehicle trajectory tracking and false trajectory judgment method based on multiple geomagnetic sensors provided by an embodiment of the present application, and the vehicle trajectory tracking and false trajectory judgment method based on multiple geomagnetic sensors provided by the present application comprises:

[0028] S101, the geomagnetic sensor deployed on both sides of the road acquires data of the passing vehicle, the data of the vehicle includes magnetic field feature data and corresponding time stamp, and sends the magnetic field feature data of the vehicle and the corresponding time stamp to the base station.

[0029] Specifically, in the embodiment, a plurality of geomagnetic sensors are arranged along the road lane line, and the magnetic field disturbance caused by the vehicle passing through the geomagnetic sensor triggers the detection of the geomagnetic sensor. Then, the geomagnetic sensor transmits the vehicle magnetic field feature data and the corresponding timestamp of the vehicle passing through to the base station through wireless connection. The base station, as an intermediate medium for data transmission, receives the timestamp information and vehicle magnetic field feature information detected by each geomagnetic sensor, saves the timestamp data, vehicle passing position data and vehicle magnetic field feature of the vehicle passing through sent by each geomagnetic sensor, and sends the sensor detection data to the data processing center.

[0030] In the embodiment, please refer to Figure 2 , Figure 2 is a schematic diagram of the geomagnetic sensor deployment diagram provided by the embodiment of the present application, in which Figure 2 n geomagnetic sensors A1, A2,..., An are arranged along the road lane line at equal intervals, and n is a natural number. n

[0031] The geomagnetic sensor used in the embodiment is an RM3100 series three-axis geomagnetic sensor, but it is not limited to other geomagnetic sensors on the market that can detect the magnetic field disturbance of the vehicle.

[0032] In the embodiment, the geomagnetic sensor is arranged on the lane line of the lane to detect the vehicle, which overcomes the defect that the geomagnetic sensor arranged in the middle of the road will damage the road surface, affect normal traffic and be difficult to maintain in the prior art. The present application can realize the judgment of large vehicle mis-detection without affecting the traffic and facilitating the maintenance.

[0033] In addition, the geomagnetic sensor used in the embodiment has lower power consumption than cameras and radars and other sensors, is easy to install, has low cost and is not easily affected by weather factors, which overcomes the defect that the vehicle detection method in the prior art has high cost and is easily affected by weather factors, resulting in high cost and instability of the vehicle detection scheme. The present application can realize the detection of the vehicle trajectory and provide the corresponding trajectory information at low cost and without being affected by the weather.

[0034] S102, the base station saves and cleans the position of the geomagnetic sensor and the data of the vehicle obtained by the geomagnetic sensor, and packages and sends them to the data processing center at a fixed time interval.

[0035] Specifically, in the embodiment, the geomagnetic sensors A1, A2,..., An are arranged along the road lane line, and the magnetic field disturbance caused by the vehicle passing through the geomagnetic sensor triggers the detection of the geomagnetic sensor. Then, the geomagnetic sensor transmits the vehicle magnetic field feature data and the corresponding timestamp of the vehicle passing through to the base station through wireless connection. The base station, as an intermediate medium for data transmission, receives the timestamp information and vehicle magnetic field feature information detected by each geomagnetic sensor, saves the timestamp data, vehicle passing position data and vehicle magnetic field feature of the vehicle passing through sent by each geomagnetic sensor, and sends the sensor detection data to the data processing center. n ​Magnetic field data are collected at the respective locations of the sensors. This data is processed to obtain discrete magnetic field disturbance data caused by vehicles passing by the sensor deployment points. A vehicle detection algorithm is then used to detect whether vehicles enter or leave the detection range of the geomagnetic sensors and to record the timestamps of vehicle detection. Geomagnetic sensors A1, A2, ..., A... n The system wirelessly transmits vehicle magnetic field characteristic data and corresponding timestamps to the base station. The base station, acting as an intermediary for data transmission, receives the timestamp information and vehicle magnetic field characteristic information detected by each geomagnetic sensor, and stores the vehicle information, including the timestamp and magnetic field characteristics, sent by each geomagnetic sensor. The base station manages the data from geomagnetic sensors A1, A2, ..., A... n The time data uploaded during vehicle inspection is cleaned, and the data from geomagnetic sensors A1, A2, ..., A... n The uploaded timestamp and vehicle magnetic field characteristics are tagged with geomagnetic sensor information (including the geomagnetic sensor deployment location and number, etc.).

[0036] S103. Within a fixed time period, the data processing center processes the vehicle data received from the geomagnetic sensor to generate a set of vehicle trajectories.

[0037] Specifically, in this embodiment, within a fixed time period, the data processing center processes the vehicle data acquired by the geomagnetic sensor to generate a set of vehicle trajectories, including:

[0038] The data processing center sorts the locations of the geomagnetic sensors, takes the data of the vehicle acquired by the first geomagnetic sensor as the starting point of the vehicle trajectory, and saves it in the data processing center.

[0039] Starting with the second geomagnetic sensor, the data processing center sequentially associates the vehicle data acquired by each geomagnetic sensor with the existing vehicle trajectories. If the existing vehicle trajectory can be associated with the vehicle data of the current geomagnetic sensor, the existing vehicle trajectory is updated; otherwise, a new vehicle trajectory is generated. The updated existing vehicle trajectory and the new vehicle trajectory constitute the vehicle trajectory set.

[0040] In this embodiment, the data center processes the received geomagnetic sensor data at fixed time intervals. The data to be processed by the data center at the current time interval is denoted as (T). s-1 ,T s All data within the time period; firstly, sort according to the sensor deployment location information in the geomagnetic sensor information tag it carries, and regard the data of each vehicle at the first geomagnetic sensor location as the starting point of each vehicle trajectory.

[0041] This embodiment defines (T)s-1 ,T s The vehicle data received from the j-th geomagnetic sensor within the time period is: in, This represents the number of vehicle data reports received from the j-th geomagnetic sensor within that time period. This represents the vehicle's transit time and position detected by the geomagnetic sensor. In this embodiment, the vehicle trajectory is defined as the following measurement data sequence: in, This represents the estimated state of the vehicle when the i-th vehicle passes the n-th geomagnetic sensor, including the vehicle's position, speed, and the time it takes for the vehicle to pass the n-th geomagnetic sensor.

[0042] The nth vehicle data reported by the first geomagnetic sensor For example, let's explain how to set the state of the vehicle trajectory start point. Set the state of the vehicle trajectory start point to:

[0043]

[0044] Indicates the time of reporting the test. This indicates the magnitude of the reported vehicle's magnetic field strength;

[0045] Starting from the location of the second geomagnetic sensor, the vehicle data acquired by each geomagnetic sensor is sequentially associated with the existing vehicle trajectory. If the vehicle trajectory can be associated with the data at the current geomagnetic sensor location, step S1031 is executed. If the vehicle trajectory cannot be associated with the data at the current location, step S1032 is executed. If, after the association is completed, there is data that cannot be associated with the existing vehicle trajectory, step S1033 is executed.

[0046] Optionally, the data processing center sequentially correlates the vehicle data acquired by each geomagnetic sensor with existing vehicle trajectories, including:

[0047] Data association is performed by calculating the proximity between existing vehicle trajectories and vehicle data acquired by geomagnetic sensors, wherein the data association includes:

[0048] The predicted time for the vehicle trajectory to travel from the current geomagnetic sensor to the next geomagnetic sensor is calculated using the following expression:

[0049]

[0050] in, Let represent the predicted time for the i-th existing vehicle trajectory to travel from the k-th geomagnetic sensor to the (k+1)-th geomagnetic sensor, where i = 1, 2, ..., I, and I represents the total number of existing vehicle trajectories reaching the k-th geomagnetic sensor, where k = 1, 2, ..., n. x represents the reporting time of the i-th existing vehicle trajectory reaching the k-th geomagnetic sensor. k The coordinates of the position of the k-th geomagnetic sensor, x k+1 The coordinates represent the position of the (k+1)th geomagnetic sensor. This represents the estimated speed of the i-th existing vehicle trajectory as it passes the k-th geomagnetic sensor;

[0051] The estimated magnetic field strength for each vehicle is then calculated, expressed as follows:

[0052]

[0053] Wherein, the current estimated magnetic field strength of the i-th vehicle trajectory is the average of the magnetic field strengths of the historical associated data of that vehicle, m j This represents the magnitude of the magnetic field strength of the vehicle data associated with the i-th vehicle trajectory at the j-th geomagnetic sensor;

[0054] According to the predicted time Calculate prediction time The approximation of the timestamps of vehicle data acquired by geomagnetic sensors is expressed as follows:

[0055]

[0056] Among them, a ij This represents the timestamp corresponding to the predicted data of the i-th vehicle passing through the (k+1)-th geomagnetic sensor and the data of the j-th vehicle obtained by the (k+1)-th geomagnetic sensor. The probability of allocation; j = 1, 2, ..., J This represents the magnitude of the magnetic field strength of the vehicle data for the j-th vehicle, where j = 1, 2, ..., J, and J represents the total number of timestamps corresponding to the vehicle data acquired by the (k+1)-th geomagnetic sensor. This indicates the estimated magnetic field strength of the vehicle.

[0057] This can also be understood as, please see Figure 3 , Figure 3This is a schematic diagram illustrating the association between an existing set of vehicle trajectories and vehicle data acquired by a geomagnetic sensor, as provided in this embodiment of the invention. The data association problem is viewed as a data allocation or assignment problem. Arrows represent allocation results with a high probability. First, the time it takes for each vehicle trajectory to pass through the current geomagnetic sensor is predicted. Then, the closeness between the predicted time and the time corresponding to the vehicle data reported by the sensor is calculated. The closeness is used as the allocation cost to generate an allocation matrix. The rows of the allocation matrix represent the corresponding vehicle trajectories, and the columns represent the measurement points to be allocated. The matrix originally represents the corresponding allocation probability. After the allocation matrix is ​​established, it is viewed as a bipartite graph model, and the final allocation result is confirmed by the Hungarian algorithm.

[0058] In this embodiment, event detection is performed in the data center, which does not put computational pressure on the edge geomagnetic sensors. This overcomes the problem in the prior art where the complex vehicle trajectory tracking algorithm puts computational pressure on the computing devices.

[0059] S1031. If the existing vehicle trajectory can be associated with the vehicle's data on the current geomagnetic sensor, then update the existing vehicle trajectory, including:

[0060] Based on the vehicle data of the current geomagnetic sensor associated with the existing vehicle trajectory, the motion state of the existing vehicle trajectory in the current geomagnetic sensor is estimated by Kalman filtering, and the existing vehicle trajectory is updated.

[0061] Specifically, assume that the vehicle state estimate at the k-th geomagnetic sensor is: The vehicle data obtained from the (k+1)th geomagnetic sensor is the j-th vehicle data. The reporting time is The corresponding geomagnetic sensor deployment location is z k+1 The corresponding Kalman filter includes two steps: state prediction and state update.

[0062] In this embodiment, assuming the vehicle moves at a constant speed from the k-th geomagnetic sensor to the (k+1)-th geomagnetic sensor, the corresponding state prediction process is as follows:

[0063]

[0064] in, Represents the state transition matrix;

[0065] Represents the noise matrix;

[0066] ε v =3m / s 2 Additive white Gaussian noise representing velocity, H k =

[100] represents the observation matrix, Sk+1 R represents the covariance matrix of the predicted measurements. k =4 indicates the measurement error, which is related to the detection accuracy of the geomagnetic sensor. In actual use, it can be adjusted appropriately to achieve a better estimation effect.

[0067] The corresponding state update process is represented as follows:

[0068]

[0069] in,

[0070] Finally, the state estimate of the vehicle is obtained.

[0071] S1032. If the existing vehicle trajectory cannot be associated with vehicle data at the current geomagnetic sensor, predict the time it takes for the existing vehicle trajectory to pass through the current geomagnetic sensor. If the predicted time does not exceed the first threshold T... s Then, the predicted time of the existing vehicle trajectory passing through the current geomagnetic sensor is used as the timestamp corresponding to the vehicle data acquired when the vehicle passes through the geomagnetic sensor, and the timestamp is filled in. If this time exceeds the first threshold T, s If so, the existing vehicle trajectory will no longer be associated with other vehicle data, and will wait to be associated with vehicle data in the next fixed time period.

[0072] Specifically, for a vehicle trajectory i that cannot be correlated with data, predict the time it takes for it to pass through the (k+1)th geomagnetic sensor location. If the predicted time No more than T s If the sensor fails to detect the vehicle's trajectory, it is considered that a sensor miss has occurred, meaning that the (k+1)th sensor did not detect the vehicle's trajectory data. Data supplementation is then performed on the trajectory, using the predicted elapsed time as the time the vehicle passed the sensor.

[0073]

[0074] S1033. If the vehicle data acquired by all the geomagnetic sensors cannot be associated with the existing vehicle trajectory, then the vehicle data shall be used as the starting point of the vehicle trajectory and participate in the association of the vehicle data acquired by the next geomagnetic sensor.

[0075] S104. Based on the time difference between adjacent vehicle trajectories passing through the same geomagnetic sensor, determine whether there are false trajectories among the adjacent vehicle trajectories. If there are, weighted fusion of the adjacent vehicle trajectories is performed to merge them into one vehicle trajectory, the vehicle trajectory set is updated, and the updated vehicle trajectory set is output. If there are no false trajectories, the vehicle trajectory set is output. Optionally, the output object can be any traffic management platform or other data center that has established a communication relationship with the data center.

[0076] Specifically, in this embodiment, please refer to Figure 4 , Figure 4 This is a flowchart of a method for determining false trajectories provided in an embodiment of the present invention. The method for determining whether a false trajectory exists among adjacent vehicle trajectories includes:

[0077] If the absolute value of the time difference between adjacent vehicle trajectories passing through the same geomagnetic sensor is less than the second threshold T, the probability of the first false detection event is used to determine whether it was generated by the same vehicle. If it was generated by the same vehicle, then there is a false trajectory; if it was not generated by the same vehicle, then there is no false trajectory.

[0078] If the absolute value of the time difference between adjacent vehicle trajectories passing through the same geomagnetic sensor is greater than or equal to the second threshold T, the probability of the second false detection event is used to determine whether it was generated by the same vehicle. If it was generated by the same vehicle, then there is a false trajectory; if it was not generated by the same vehicle, then there is no false trajectory.

[0079] In this embodiment, for the two original vehicle trajectories and Determine if there are cases where the absolute value of the time difference between passing through the same sensor is less than the threshold T.

[0080] For the geomagnetic sensor used in this embodiment, there is a local time synchronization error of ±50ms. At the same time, continuous sensor detection triggered by the same vehicle can be considered as the vehicles passing by one after the other, which generally does not exceed 200ms. Therefore, in this embodiment, the time difference threshold T is set to 200ms.

[0081] Let φ represent a false detection event. φ = 1 indicates that the data for both trajectories were generated by the same vehicle, meaning one of them is a false trajectory; φ = 0 indicates that the two trajectories belong to two different vehicles, and both are normal trajectories. Then, the probability of a false detection event occurring up to the k-th geomagnetic sensor is:

[0082]

[0083] Where, Δt k This represents the absolute value of the time difference between the two vehicles when they pass the k-th geomagnetic sensor. If the time it takes for vehicle trajectory v1 to pass the k-th geomagnetic sensor is... The time it takes for vehicle trajectory v2 to pass the k-th geomagnetic sensor is but Based on Δt k To determine if the value is less than the threshold T, calculate the value of the above formula.

[0084] The expression for the probability of the first false detection event is:

[0085]

[0086] Where, φ k Pr(Δt) represents the false detection event of the k-th geomagnetic sensor. k <T|φ k =1) and Pr(Δt) k <T|φ k =0) represents the likelihood function;

[0087] If Pr(φ) k =1|Δt 0,1,.....,k-1 ,Δt k The value of ) exceeds the set threshold P T If Pr(φ) is a false trajectory, then it is considered that there is a false trajectory among the adjacent vehicle trajectories. k =0|Δt 0,1,.....,k-1 ,Δt k The value of ) exceeds the set threshold P S If so, it is assumed that there are no false trajectories among the adjacent vehicle trajectories.

[0088] The expression for the probability of the second false detection event is:

[0089]

[0090] Where, φ k Pr(Δt) represents the false detection event of the k-th geomagnetic sensor. k ≥T|φ k =1) and Pr(Δt) k ≥T|φ k =0) represents the likelihood function;

[0091] If Pr(φ) k =1|Δt 0,1,.....,k-1 ,Δt k The value of ) exceeds the set threshold P T If Pr(φ) is a false trajectory, then it is considered that there is a false trajectory among the adjacent vehicle trajectories. k =0|Δt 0,1,.....,k-1 ,Δt k The value of ) exceeds the set threshold P S If so, it is assumed that there are no false trajectories among the adjacent vehicle trajectories. Optionally, P T=0.85, which can be adjusted according to the detection effect in specific applications.

[0092] It should be noted that Δt can be observed when false detection events actually occur and when no false detection events occur. k The initial value Pr(φ) is obtained from the distribution. k =1|Δt0) and Pr(φ k =0|Δt0) all take values ​​of 0.5, C is a normalization constant, ensuring Pr(φ k =1|Δt 0,1,.....,k-1 ,Δt k ) and Pr(φ k =0|Δt 0,1,.....,k ,Δt k The sum of ) is 1.

[0093] Optionally, the two trajectories can be weighted and merged to form a unified vehicle trajectory. The weighting method is average weighting, that is, the weight of each trajectory is 0.5.

[0094] In this embodiment, the presence of false trajectories in the generated trajectory is identified, making the vehicle trajectory information provided by the present invention more reliable.

[0095] In summary, this invention provides a method for vehicle trajectory tracking and false trajectory identification based on multiple geomagnetic sensors. First, geomagnetic sensors are deployed on the roadside to collect data as vehicles pass by. Second, the original vehicle trajectory is generated in a data center using the detection data from the multiple geomagnetic sensors. Then, based on the generated trajectory, a Bayesian filtering method is used to detect false trajectories caused by sensor misdetection. Finally, the data from the false trajectories caused by misdetection are fused with the real trajectory to eliminate erroneous false trajectories, ultimately outputting reliable vehicle trajectory information. The method proposed in this invention can be used to achieve intelligent highway management.

[0096] Based on the same inventive concept, this invention also provides a vehicle trajectory tracking and false trajectory judgment device based on multiple geomagnetic sensors, applied to the vehicle trajectory tracking and false trajectory judgment method based on multiple geomagnetic sensors provided in the above embodiments of this invention. The method embodiments are described above and will not be repeated here. The device includes:

[0097] The data acquisition module is used to acquire data of passing vehicles by geomagnetic sensors deployed on both sides of the road. The vehicle data includes magnetic field characteristic data and corresponding timestamps, and the magnetic field characteristic data and corresponding timestamps of the vehicles are sent to the base station.

[0098] The data processing center is used by the base station to save and clean the location of the geomagnetic sensor and the vehicle data acquired by the geomagnetic sensor, and then package and send it to the data processing center at fixed time intervals.

[0099] The vehicle trajectory generation module is used to process the vehicle data obtained by the geomagnetic sensor received by the data processing center within a fixed time period to generate a set of vehicle trajectories.

[0100] The false trajectory detection module is used to determine whether there is a false trajectory among adjacent vehicle trajectories based on the time difference between adjacent vehicle trajectories passing through the same geomagnetic sensor. If there is a false trajectory, the adjacent vehicle trajectories are weighted and merged into a single vehicle trajectory, the vehicle trajectory set is updated, and the updated vehicle trajectory set is output. If there is no false trajectory, the vehicle trajectory set is output.

[0101] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations are intended to cover non-exclusive inclusion, such that an article or device comprising a list of elements includes not only those elements but also other elements not expressly listed. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device comprising said element. Terms such as "connected" or "linked" are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect. The orientations or positional relationships indicated by terms such as "upper," "lower," "left," and "right" are based on the orientations or positional relationships shown in the accompanying drawings and are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as limiting the invention.

[0102] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features or characteristics described may be combined in any suitable manner in one or more embodiments or examples. In addition, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0103] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A multi-magnetic sensor-based vehicle trajectory tracking and false trajectory determination method, characterized in that, The method comprises the following steps: A geomagnetic sensor arranged on both sides of a road acquires data of a passing vehicle, the data of the vehicle comprising magnetic field feature data and a corresponding timestamp, and sends the magnetic field feature data of the vehicle and the corresponding timestamp to a base station; The base station saves and cleans the position of the geomagnetic sensor and the data of the vehicle acquired by the geomagnetic sensor, and sends the data to a data processing center at a fixed time interval; In a fixed time period, the data processing center processes the data of the vehicle acquired by the geomagnetic sensor received, and generates a vehicle trajectory set; According to the time difference of adjacent vehicle trajectories passing through the same geomagnetic sensor, it is determined whether there is a false trajectory in the adjacent vehicle trajectories, if there is, the adjacent vehicle trajectories are fused by weighting and merged into one vehicle trajectory, the vehicle trajectory set is updated, and the updated vehicle trajectory set is output, if there is not, the vehicle trajectory set is output; The determination of whether there is a false trajectory in the adjacent vehicle trajectories comprises: If the absolute value of the time difference of the adjacent vehicle trajectories passing the same geomagnetic sensor is less than a second threshold and whether it is the same vehicle generating the first false detection event, if it is the same vehicle, there is a false trajectory, and if it is not the same vehicle, there is no false trajectory. If the absolute value of the time difference of the adjacent vehicle trajectories passing the same geomagnetic sensor is greater than or equal to a second threshold value If the same vehicle is determined to be generated by the second false detection event, the false trajectory exists, and if the same vehicle is not determined to be generated, the false trajectory does not exist. The expression of the probability of the first false detection event is: ； wherein, represents a false detection event of the i-th geomagnetic sensor, and represents a likelihood function;​ If the value exceeds a set threshold , it is considered that there is a false trajectory in the adjacent vehicle trajectory, if the value exceeds a set threshold , it is considered that there is no false trajectory in the adjacent vehicle trajectory, denotes the absolute value of the time difference of the two vehicles when passing through the first magnetic sensor, is a normalization constant; The expression of the probability of the second false detection event is: ； wherein, represents a false detection event of the i-th geomagnetic sensor, and represents a likelihood function;​ If the value exceeds a set threshold , it is considered that there is a false trajectory in the adjacent vehicle trajectory, and if the value exceeds a set threshold , it is considered that there is no false trajectory in the adjacent vehicle trajectory.

2. The multi-magnetic sensor based vehicle trajectory tracking and false trajectory determination method according to claim 1, characterized in that, In the fixed time period, the data processing center processes the data of the vehicle acquired by the geomagnetic sensor received, and generates a vehicle trajectory set, which comprises: The data processing center sorts the positions of the geomagnetic sensors, takes the data of the vehicle acquired by the first geomagnetic sensor as the starting point of the vehicle trajectory, and saves it in the data processing center; Starting from the second geomagnetic sensor, the data processing center sequentially associates the data of the vehicle acquired by each geomagnetic sensor with the existing vehicle trajectory, if the existing vehicle trajectory can be associated with the data of the vehicle at the current geomagnetic sensor, the existing vehicle trajectory is updated, otherwise, a new vehicle trajectory is generated; wherein the updated existing vehicle trajectory and the new vehicle trajectory constitute the vehicle trajectory set.

3. The multi-magnetic sensor based vehicle trajectory tracking and false trajectory determination method according to claim 2, characterized in that, The data processing center sequentially associates the data of the vehicle acquired by each geomagnetic sensor with the existing vehicle trajectory, which comprises: The data association comprises: The expression of the predicted time of the vehicle trajectory from the current geomagnetic sensor to the next geomagnetic sensor is: ； in, Indicates the first The existing vehicle trajectory from the first The geomagnetic sensor reached the first Prediction time of a geomagnetic sensor , Indicates reaching the th The total number of existing vehicle trajectories from each geomagnetic sensor. , Indicates the first The existing vehicle trajectory reaches the first The reporting time of each geomagnetic sensor Indicates the first The coordinates of the location of each geomagnetic sensor Indicates the first The coordinates of the location of each geomagnetic sensor Indicates the first The existing vehicle trajectory passes through the first Estimated velocity from a geomagnetic sensor; According to the predicted time , the predicted time is calculated in proximity to the respective time stamp of the vehicle data acquired by the geomagnetic sensor, whose expression is: ； wherein, denotes the th vehicle's predicted passage through the th geomagnetic sensor and the th geomagnetic sensor, denotes the th vehicle's data's corresponding timestamp assigned to the th vehicle's data, denotes the th vehicle's data's vehicle magnetic field strength magnitude, denotes the th geomagnetic sensor's data's corresponding timestamp, denotes the vehicle's estimated magnetic field strength magnitude.

4. The multi-magnetic sensor based vehicle trajectory tracking and false trajectory determination method according to claim 2, characterized in that, If the existing vehicle trajectory can be associated with the data of the vehicle at the current geomagnetic sensor, the existing vehicle trajectory is updated, which comprises: According to the vehicle data of the current geomagnetic sensor associated with the existing vehicle trajectory, the motion state of the existing vehicle trajectory at the current geomagnetic sensor is estimated by Kalman filtering, and the existing vehicle trajectory is updated.

5. The multi-magnetic sensor based vehicle trajectory tracking and false trajectory determination method according to claim 2, characterized in that, If the existing vehicle trajectory cannot be associated with the data of the current geomagnetic sensor related to the vehicle, predict the time when the existing vehicle trajectory passes the current geomagnetic sensor, if the predicted time does not exceed the first threshold , then the predicted time of the existing vehicle trajectory passing the current geomagnetic sensor is supplemented as the time stamp corresponding to the data of the vehicle obtained when the vehicle passes the geomagnetic sensor, and if the predicted time exceeds the first threshold , the existing vehicle trajectory is no longer associated with other vehicle data, and waits to be associated with the vehicle data of the next fixed time period.

6. The multi-magnetic sensor based vehicle trajectory tracking and false trajectory determination method according to claim 2, characterized in that, If the vehicle data acquired by all the geomagnetic sensors cannot be associated with the existing vehicle trajectory, the vehicle data is taken as the starting point of the vehicle trajectory, and participates in the association of the vehicle data acquired by the next geomagnetic sensor.

7. A multi-magnetic sensor based vehicle trajectory tracking and false trajectory determination device, characterized by, The method comprises the following steps: A data acquisition module is arranged on both sides of the road to acquire data of passing vehicles, the data of the vehicles including magnetic field feature data and corresponding time stamps, and the magnetic field feature data and corresponding time stamps of the vehicles are sent to a base station; A data processing center is arranged to save and clean the data of the vehicles acquired by the geomagnetic sensor and the position of the geomagnetic sensor, and to send the data to the data processing center at a fixed time interval; A vehicle trajectory generation module is arranged to process the data of the vehicles acquired by the geomagnetic sensor received by the data processing center to generate a vehicle trajectory set within a fixed time period; A false trajectory judgment module is arranged to judge whether there is a false trajectory in adjacent vehicle trajectories according to the time difference of the adjacent vehicle trajectories passing through the same geomagnetic sensor, if there is, to perform weighted fusion on the adjacent vehicle trajectories to merge them into one vehicle trajectory, update the vehicle trajectory set, and output the updated vehicle trajectory set, and if there is not, to output the vehicle trajectory set; The judgment of whether there is a false trajectory in the adjacent vehicle trajectories includes: If the absolute value of the time difference of the adjacent vehicle trajectories passing the same geomagnetic sensor is less than a second threshold and whether it is the same vehicle generating the first false detection event, if it is the same vehicle, there is a false trajectory, and if it is not the same vehicle, there is no false trajectory. If the absolute value of the time difference of the adjacent vehicle trajectories passing the same geomagnetic sensor is greater than or equal to a second threshold value If the same vehicle is determined to be generated by the probability of the second false detection event, a false trajectory exists, and if the same vehicle is determined not to be generated, a false trajectory does not exist. The expression of the probability of the first false detection event is: ； wherein, represents a false detection event of the i-th geomagnetic sensor, and represents a likelihood function;​ If the value exceeds a set threshold , it is considered that there is a false trajectory in the adjacent vehicle trajectory, if the value exceeds a set threshold , it is considered that there is no false trajectory in the adjacent vehicle trajectory, represents the absolute value of the time difference of the two vehicles when passing through the first magnetic sensor, is a normalization constant; The expression of the probability of the second false detection event is: ； in, Indicates the first False detection events of a geomagnetic sensor and Represents the likelihood function; If the value exceeds a set threshold a false trajectory is considered to exist in the adjacent vehicle trajectories, and if the value exceeds a set threshold a false trajectory is considered not to exist in the adjacent vehicle trajectories.

Images