Detector networking fusion vehicle tracking method and system in intersection environment

Abstract

The invention discloses a vehicle tracking method and system for detector networking fusion in an intersection environment, and the method comprises the steps: building a coordinate system with unified latitude and longitude based on the arrangement positions of a networking radar detector and a video detector; according to the latitude and longitude unified coordinate system, obtaining one-time fusion data of intersections on the same side and one-time fusion data of intersections on different sides detected by the networking radar detector and the video detector respectively; and performing secondary data fusion on the primary fusion data of the intersections on the same side and the primary fusion data of the intersections on the different sides to complete target tracking. According to the invention, fusion of multiple radars and a video detector in an intersection environment is mainly used for vehicle tracking, and unique vehicle trajectory information is automatically generated through technologies of establishment of a latitude and longitude unified coordinate system of multiple sensors, vehicle data fusion between the radars and the video detector, networking secondary data fusion, trajectory prediction and the like. The problems of multi-path generation and loss of the tracking target are solved, the automation degree is high, and the anti-interference capability is strong; and the target tracking accuracy and precision are high.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent traffic signal control, and in particular discloses a vehicle tracking method and system for detector networking fusion in an intersection environment. Background Art

[0002] With the advancement of intelligent urban construction in my country, intelligent systems for traffic signal control have emerged. Intersections are the core area of ​​traffic signal control. Since each type of sensor has its advantages and disadvantages, it is necessary to combine data from multiple devices in order to maximize detection and tracking capabilities.

[0003] Existing vehicle tracking methods have limitations in terms of reliability and data: 1. Vehicle tracking with a single radar is easily affected by many factors such as weather and solar irradiance, and when the target is outside the radar's field of view or the target's line of sight is obstructed, the tracking target will be lost.

[0004] 2. Currently, traditional millimeter-wave radars will lose tracking targets when the speed is low or stationary.

[0005] 3. The fusion of multiple sensors has the problem of multiple paths and a lot of noise, and ultimately a single trajectory data needs to be formed.

[0006] Therefore, the above-mentioned defects existing in the existing vehicle tracking method are technical problems that need to be solved urgently. Summary of the invention

[0007] The present invention provides a vehicle tracking method and system with detector networking fusion in an intersection environment, aiming to solve at least one defect existing in the above-mentioned existing vehicle tracking method.

[0008] One aspect of the present invention relates to a vehicle tracking method using a detector network fusion in an intersection environment, comprising the following steps: Based on the layout positions of the networked radar detectors and video detectors, a unified longitude and latitude coordinate system is established; According to the unified coordinate system of longitude and latitude, the first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection detected by the networked radar detector and the video detector are obtained respectively; The first fusion data of the intersection on the same side and the first fusion data of the intersection on the opposite side are fused for the second time to complete the target tracking.

[0009] Furthermore, based on the arrangement positions of the networked radar detectors and the video detectors, the steps of establishing a unified longitude and latitude coordinate system include: According to the Beidou positioning system, the layout positions of the networked radar detectors and video detectors are spatially located to obtain the longitude and latitude coordinates; Based on the longitude and latitude coordinates, the origin of the coordinate system is established with the radar layout location, and the speed close to the origin is defined as positive, and the speed away from the origin is defined as negative; the longitude and latitude of the reference object in the coordinate system are aligned to establish a unified longitude and latitude coordinate system.

[0010] Furthermore, the latitude and longitude coordinates include the longitude coordinates and latitude coordinates , according to the unified coordinate system of longitude and latitude, the steps of respectively obtaining the once fused data of the same-side intersection and the once fused data of the opposite-side intersection of the detector network include: Calibrate the longitude and latitude to unify the coordinate system and perform data synchronization processing on the networked radar detector and video detector; Based on longitude coordinates and latitude coordinates , calculate ; in accordance with The vehicle state data is used to predict the vehicle trajectory using a radar detector and a video detector respectively to generate vehicle trajectory prediction data; the vehicle trajectory prediction data includes vehicle trajectory prediction data for the same-side intersection and vehicle trajectory prediction data for the opposite-side intersection; the vehicle trajectory prediction data for the same-side intersection includes first vehicle trajectory prediction data for the same-side intersection generated by the radar detector and second vehicle trajectory prediction data for the same-side intersection generated by the video detector, and the vehicle trajectory prediction data for the opposite-side intersection includes first vehicle trajectory prediction data for the opposite-side intersection generated by the radar detector and second vehicle trajectory prediction data for the opposite-side intersection generated by the video detector; The first vehicle trajectory prediction data of the same-side intersection and the second vehicle trajectory prediction data of the same-side intersection are fused to generate a fusion of the same-side intersection fusion data. Arrays; The first opposite-side intersection vehicle trajectory prediction data and the second opposite-side intersection vehicle trajectory prediction data are fused to generate a fusion of the opposite-side intersection fusion data. Array.

[0011] Furthermore, the steps of performing secondary data fusion on the first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection to complete target tracking include: For a fusion Arrays and One-shot Fusion The array performs secondary data fusion to generate an intersection network fusion array; Based on the intersection networking fusion array, the complete vehicle movement trajectory is drawn in real time to complete target tracking.

[0012] Furthermore, the intersection network fusion array is:

[0013] in, To network and fuse arrays for intersections, For a fusion Arrays, For a fusion Arrays, For a fusion Array.

[0014] Another aspect of the present invention relates to a vehicle tracking system with detector networking fusion in an intersection environment, comprising: Establishing a module for establishing a unified longitude and latitude coordinate system based on the layout positions of the networked radar detectors and the video detectors; An acquisition module is used to obtain the first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection detected by the networked radar detector and the video detector according to the unified coordinate system of longitude and latitude; The tracking module is used to perform secondary data fusion on the first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection to complete target tracking.

[0015] Further, the building module includes: A positioning unit is used to spatially locate the layout of the networked radar detector and the video detector according to the Beidou positioning system to obtain the longitude and latitude coordinates; The establishment unit is used to establish the origin of the coordinate system based on the longitude and latitude coordinates and the radar layout position, define the speed close to the origin as positive, and the speed away from the origin as negative; align the longitude and latitude of the reference object in the coordinate system to establish a unified longitude and latitude coordinate system.

[0016] Furthermore, the acquisition module includes: A synchronization processing unit, used to calibrate the unified coordinate system of longitude and latitude, and perform data synchronization processing on the networked radar detector and the video detector; Calculation unit for longitude coordinates and latitude coordinates , calculate ; The first generation unit is used to generate The vehicle state data is used to predict the vehicle trajectory using a radar detector and a video detector respectively to generate vehicle trajectory prediction data; the vehicle trajectory prediction data includes vehicle trajectory prediction data for the same-side intersection and vehicle trajectory prediction data for the opposite-side intersection; the vehicle trajectory prediction data for the same-side intersection includes first vehicle trajectory prediction data for the same-side intersection generated by the radar detector and second vehicle trajectory prediction data for the same-side intersection generated by the video detector, and the vehicle trajectory prediction data for the opposite-side intersection includes first vehicle trajectory prediction data for the opposite-side intersection generated by the radar detector and second vehicle trajectory prediction data for the opposite-side intersection generated by the video detector; The second generating unit is used to fuse the first vehicle trajectory prediction data of the same-side intersection and the second vehicle trajectory prediction data of the same-side intersection to generate a fusion of the first fusion data of the same-side intersection. Arrays; The third generating unit is used to fuse the first opposite-side intersection vehicle trajectory prediction data and the second opposite-side intersection vehicle trajectory prediction data to generate a first fusion of the opposite-side intersection first fusion data. Array.

[0017] Furthermore, the tracking module includes: The fourth generation unit is used to generate a fusion Arrays and One-shot Fusion The array performs secondary data fusion to generate an intersection network fusion array; The tracking unit is used to draw the complete vehicle movement trajectory in real time based on the intersection network fusion array to complete target tracking.

[0018] Furthermore, the intersection network fusion array is:

[0019] in, To network and fuse arrays for intersections, For a fusion Arrays, For a fusion Arrays, For a fusion Array.

[0020] The beneficial effects achieved by the present invention are: The present invention provides a vehicle tracking method and system with detector networking fusion in an intersection environment, by establishing a unified longitude and latitude coordinate system based on the arrangement positions of the networked radar detector and the video detector; according to the unified longitude and latitude coordinate system, respectively obtain the primary fusion data of the same-side intersection and the primary fusion data of the opposite-side intersection detected by the networked radar detector and the video detector; perform secondary data fusion on the primary fusion data of the same-side intersection and the primary fusion data of the opposite-side intersection to complete target tracking. The vehicle tracking method and system with detector networking fusion in an intersection environment provided by the present invention relates to vehicle trajectory display in the field of intelligent traffic signal control, mainly uses the fusion of multiple radars and video detectors in an intersection environment to track vehicles, automatically generates unique vehicle trajectory information by establishing a unified longitude and latitude coordinate system of multiple sensors, vehicle data fusion between radars and video detectors, networking secondary data fusion, trajectory prediction and other technologies, solves the problem of multi-path generation and loss of tracking targets, has a high degree of automation and strong anti-interference ability; and has high target tracking accuracy and precision. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flow chart of a first embodiment of a vehicle tracking method using a detector network fusion in an intersection environment according to the present invention; Figure 2 The present invention is a schematic diagram of the overall flow of an embodiment of a vehicle tracking method using a detector network fusion in an intersection environment. DETAILED DESCRIPTION

[0022] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0023] like Figure 1 and Figure 2 As shown, the first embodiment of the present invention proposes a vehicle tracking method using a detector network fusion in an intersection environment, comprising the following steps: Step S100: establishing a unified longitude and latitude coordinate system based on the layout positions of the networked radar detectors and the video detectors.

[0024] A detector network of multiple radar detectors and video detectors is established, and a unified longitude and latitude coordinate system is established according to the arrangement positions of the radar detectors and video detectors in the detector network.

[0025] Step S200: According to the unified coordinate system of longitude and latitude, the primary fusion data of the same-side intersection and the primary fusion data of the opposite-side intersection detected by the networked radar detector and the video detector are respectively obtained.

[0026] Get the longitude coordinates according to the established longitude and latitude unified coordinate system and latitude coordinates ; Based on longitude coordinates and latitude coordinates , respectively obtain the once fused data of the same-side intersection and the once fused data of the opposite-side intersection detected by the radar detector and the video detector in the detector network.

[0027] Step S300: perform secondary data fusion on the first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection to complete target tracking.

[0028] The first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection obtained respectively are fused for the second time to complete the target tracking.

[0029] Further, see Figure 1 and Figure 2 The vehicle tracking method using a detector network fusion in an intersection environment provided in this embodiment, step S100 includes: Step S110: According to the Beidou positioning system, the layout positions of the networked radar detectors and the video detectors are spatially positioned to obtain longitude and latitude coordinates.

[0030] According to the Beidou positioning system, the layout positions of the radar detectors and video detectors in the detector network are spatially positioned to obtain the longitude and latitude coordinates.

[0031] Step S120: establish the origin of the coordinate system based on the longitude and latitude coordinates with the radar deployment position, define the speed approaching the origin as positive, and the speed away from the origin as negative; align the longitude and latitude of the reference object in the coordinate system to establish a unified longitude and latitude coordinate system.

[0032] Based on the longitude coordinates in the latitude and longitude coordinates and latitude coordinates , establish the origin of the coordinate system with the radar layout position, define the speed close to the origin as positive, and the speed away from the origin as negative; align the longitude and latitude of the reference object in the coordinate system to establish a unified longitude and latitude coordinate system.

[0033] Preferably, see Figure 1 and Figure 2 The vehicle tracking method using a detector network fusion in an intersection environment provided in this embodiment, step S200 includes: Step S210, calibrate the unified longitude and latitude coordinate system, and perform data synchronization processing on the networked radar detector and the video detector.

[0034] Calibrate the radar time for data synchronization processing with an error within 100ms.

[0035] Step S220: According to the longitude coordinates and latitude coordinates , calculate .

[0036] 1) Filter the noisy data through Kalman filtering algorithm; Vehicle status at the moment As shown in Formula 1: (1) In formula (1), for Vehicle status data at all times, is the longitude coordinate, is the latitude coordinate, is the velocity in longitude direction, is the velocity in the latitude direction.

[0037] 2) Vehicle status at the moment and Error covariance at time The definitions are shown in formulas (2) and (3): (2) (3) In formulas (2) and (3), for Vehicle status data at all times, is the state transfer matrix, for The error covariance at time t, for The error covariance at time t, is the state transposed matrix.

[0038] 3) State transfer matrix The definition is shown in formula (4): (4) In formula (4), is the state transfer matrix, is the change in time.

[0039] 4) Kalman gain at time It is defined as shown in formula (5), where the measurement noise Initially 1: (5) In formula (5), for The Kalman gain at time t, for The error covariance at time t, is the state observation matrix, is the observation transposed matrix, To measure noise.

[0040] 5) Vehicle status at the moment and error covariance The update is as shown in formula (6) and (7), where the observation value use replace: (6) (7) In formulas (6) and (7), for The vehicle status at the moment, for The Kalman gain at time t, is the observed value, is the observation matrix, for The error covariance at time .

[0041] 7) Observation Matrix It is defined as shown in formula (8): (8) In formula (8), is the observation matrix.

[0042] 8) After the target data source disappears, the Kalman filter algorithm is used to make a short-term prediction of the vehicle state. By taking advantage of its data smoothing characteristics, the observation value is set to the state of the last frame before disappearance for 10 consecutive frames, thereby obtaining the prediction value of 10 frames; obtaining the heading angle of the target ,long ,Width The length and width information are easily affected by factors such as occlusion, so the mean filter algorithm is used to process the data, and the target heading angle is processed using the Kalman filter algorithm to generate a complete Vehicle status data at all times.

[0043] (9) (10) (11) In formulas (9)~(11), for Vehicle status data at all times, is the longitude coordinate, is the latitude coordinate, is the velocity in longitude direction, is the latitude speed, The length information of the target; The width information of the target; is the length information of the total target; It is the width information of the total target.

[0044] Step S230: The vehicle state data is obtained by predicting the vehicle trajectory using a radar detector and a video detector respectively to generate vehicle trajectory prediction data; the vehicle trajectory prediction data includes vehicle trajectory prediction data for the same-side intersection and vehicle trajectory prediction data for the opposite-side intersection; the vehicle trajectory prediction data for the same-side intersection includes first vehicle trajectory prediction data for the same-side intersection generated by the radar detector and second vehicle trajectory prediction data for the same-side intersection generated by the video detector, and the vehicle trajectory prediction data for the opposite-side intersection includes first vehicle trajectory prediction data for the opposite-side intersection generated by the radar detector and second vehicle trajectory prediction data for the opposite-side intersection generated by the video detector.

[0045] Step S240: fusing the first same-side intersection vehicle trajectory prediction data and the second same-side intersection vehicle trajectory prediction data to generate a first fusion of the same-side intersection first fusion data. Array.

[0046] Step S250: fusing the first opposite-side intersection vehicle trajectory prediction data and the second opposite-side intersection vehicle trajectory prediction data to generate a first fusion of the opposite-side intersection first fusion data. Array.

[0047] exist Always keep an eye on the same side radar With video detector Perform data fusion: a) Overlapping area, set the array after meeting the fusion conditions To mark the detector source, and add vehicle color With license plate .

[0048] The fusion method uses Euclidean distance Calculate the data in the sample and meet the threshold Require: (12) In formula (12), is the Euclidean distance, for The data of the radar on the same side is always integrated. for The data of the video detector on the same side is fused at all times. is the threshold value.

[0049] Update the labeled detector source after the fusion condition is met, To add a new detector number, Is the original detector serial number: (13) In formula (13), To set the array after meeting the fusion conditions, is the original detector serial number, Add the detector serial number.

[0050] Fusing Arrays If the array is from multiple detectors, the mean data is used: (14) In formula (14), is the fused array, is the longitude coordinate, is the latitude coordinate, is the direction speed, is the heading angle of the target, is the length information of the target, is the width information of the target, To add new vehicle colors, To add a new license plate.

[0051] b) Non-overlapping region, when the labeled array has only a single detector, the fused array The data source is switched to this detector.

[0052] Preferably, see Figure 1 and Figure 2 The vehicle tracking method using a detector network fusion in an intersection environment provided in this embodiment, step S200 includes: Step S310: A fusion Arrays and One-shot Fusion The array is subjected to secondary data fusion to generate an intersection network fusion array.

[0053] Traversal Multiple fusion arrays of the network detector at this moment After a second data fusion, the final intersection network fusion array is generated as follows: (15) In formula (15), To network and fuse arrays for intersections, For a fusion Arrays, For a fusion Arrays, For a fusion Array.

[0054] Step S320: Based on the intersection network fusion array, a complete vehicle motion trajectory is drawn in real time to complete target tracking.

[0055] Based on the calculated big data intersection network fusion array , the complete vehicle motion trajectory can be drawn in real time to achieve the tracking purpose.

[0056] The invention relates to a vehicle tracking system for detector networking fusion in an intersection environment, comprising an establishment module, an acquisition module and a tracking module, wherein the establishment module is used to establish a unified longitude and latitude coordinate system based on the arrangement positions of a networked radar detector and a video detector; the acquisition module is used to respectively acquire the primary fusion data of the same-side intersection and the primary fusion data of the opposite-side intersection detected by the networked radar detector and the video detector according to the unified longitude and latitude coordinate system; and the tracking module is used to perform secondary data fusion on the primary fusion data of the same-side intersection and the primary fusion data of the opposite-side intersection to complete target tracking.

[0057] The building module builds a detector network of multiple radar detectors and video detectors, and builds a unified longitude and latitude coordinate system according to the layout positions of the radar detectors and video detectors in the detector network.

[0058] The acquisition module obtains the longitude coordinates according to the established longitude and latitude unified coordinate system and latitude coordinates ; Based on longitude coordinates and latitude coordinates , respectively obtain the once fused data of the same-side intersection and the once fused data of the opposite-side intersection detected by the radar detector and the video detector in the detector network.

[0059] The tracking module performs a secondary data fusion on the first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection to complete the target tracking.

[0060] Furthermore, the present embodiment provides a vehicle tracking system with a network of detectors in an intersection environment, and the establishment module includes a positioning unit and an establishment unit, wherein the positioning unit is used to spatially locate the layout positions of the networked radar detectors and the video detectors according to the Beidou positioning system to obtain longitude and latitude coordinates; the establishment unit is used to establish the origin of the coordinate system with the radar layout position based on the longitude and latitude coordinates, define the speed close to the origin as positive, and the speed away from the origin as negative; align the longitude and latitude of the reference object in the coordinate system to establish a unified longitude and latitude coordinate system.

[0061] The positioning unit spatially locates the layout positions of the radar detectors and video detectors in the detector network according to the Beidou positioning system to obtain the longitude and latitude coordinates.

[0062] Create a unit based on the longitude coordinate in the latitude and longitude coordinates and latitude coordinates , establish the origin of the coordinate system with the radar layout position, define the speed close to the origin as positive, and the speed away from the origin as negative; align the longitude and latitude of the reference object in the coordinate system to establish a unified longitude and latitude coordinate system.

[0063] Preferably, in a vehicle tracking system with a detector network fusion under an intersection environment provided in this embodiment, the acquisition module includes a synchronization processing unit, a calculation unit, a first generation unit, a second generation unit and a third generation unit, wherein the synchronization processing unit is used to calibrate the longitude and latitude unified coordinate system and perform data synchronization processing on the networked radar detector and the video detector; the calculation unit is used to obtain the data according to the longitude coordinate system. and latitude coordinates , calculate ; The first generation unit is used according to The vehicle state data is respectively predicted by using a radar detector and a video detector to predict the vehicle trajectory to generate vehicle trajectory prediction data; the vehicle trajectory prediction data includes vehicle trajectory prediction data for the same-side intersection and vehicle trajectory prediction data for the opposite-side intersection; the vehicle trajectory prediction data for the same-side intersection includes first vehicle trajectory prediction data for the same-side intersection generated by the radar detector and second vehicle trajectory prediction data for the same-side intersection generated by the video detector, and the vehicle trajectory prediction data for the opposite-side intersection includes first vehicle trajectory prediction data for the opposite-side intersection generated by the radar detector and second vehicle trajectory prediction data for the opposite-side intersection generated by the video detector; the second generating unit is used to fuse the first vehicle trajectory prediction data for the same-side intersection and the second vehicle trajectory prediction data for the same-side intersection to generate a fusion of the same-side intersection fusion data array; a third generating unit, used to fuse the first opposite-side intersection vehicle trajectory prediction data and the second opposite-side intersection vehicle trajectory prediction data to generate a fusion of the opposite-side intersection fusion data Array.

[0064] The synchronization processing unit calibrates the radar time for data synchronization processing with an error within 100ms.

[0065] 1) The computing unit filters the noise data through the Kalman filtering algorithm; Vehicle status at the moment As shown in formula (16): (16) In formula (16), for Vehicle status data at all times, is the longitude coordinate, is the latitude coordinate, is the velocity in longitude direction, is the velocity in the latitude direction.

[0066] 2) Vehicle status at the moment and Error covariance at time The definitions are as shown in formulas (17) and (18): (17) (18) In formulas (17) and (18), for Vehicle status data at all times, is the state transfer matrix, for The error covariance at time t, for The error covariance at time t, is the state transposed matrix.

[0067] 3) State transfer matrix The definition is shown in formula (19): (19) In formula (19), is the state transfer matrix, is the change in time.

[0068] 4) Kalman gain at time It is defined as shown in formula (20), where the measurement noise Initially 1: (20) In formula (20), for The Kalman gain at time t, for The error covariance at time t, is the state observation matrix, is the observation transposed matrix, To measure noise.

[0069] 5) Vehicle status at the moment and error covariance The update of is shown in formulas (21) and (22), where the observation value use replace: (twenty one) (twenty two) In formulas (21) and (22), for The vehicle status at the moment, for The Kalman gain at time t, is the observed value, is the observation matrix, for The error covariance at time .

[0070] 7) Observation Matrix It is defined as shown in formula (23): (twenty three) In formula (23), is the observation matrix.

[0071] 8) After the target data source disappears, the Kalman filter algorithm is used to make a short-term prediction of the vehicle state. By taking advantage of its data smoothing characteristics, the observation value is set to the state of the last frame before disappearance for 10 consecutive frames, thereby obtaining the prediction value of 10 frames; obtaining the heading angle of the target ,long ,Width The length and width information are easily affected by factors such as occlusion, so the mean filter algorithm is used to process the data, and the target heading angle is processed using the Kalman filter algorithm to generate a complete Vehicle status data at all times.

[0072] (twenty four) (25) (26) In formulas (24)~(26), for Vehicle status data at all times, is the longitude coordinate, is the latitude coordinate, is the velocity in longitude direction, is the latitude speed, The length information of the target; The width information of the target; is the length information of the total target; It is the width information of the total target.

[0073] The third generation unit is Always keep an eye on the same side radar With video detector Perform data fusion: a) Overlapping area, set the array after meeting the fusion conditions To mark the detector source, and add vehicle color With license plate .

[0074] The fusion method uses Euclidean distance Calculate the data in the sample and meet the threshold Require: (27) In formula (27), is the Euclidean distance, for The data of the radar on the same side is always integrated. for The data of the video detector on the same side is fused at all times. is the threshold value.

[0075] Update the labeled detector source after the fusion condition is met, To add a new detector number, Is the original detector serial number: (28) In formula (28), To set the array after meeting the fusion conditions, is the original detector serial number, Add the detector serial number.

[0076] Fusing Arrays If the array is from multiple detectors, the mean data is used: (29) In formula (29), is the fused array, is the longitude coordinate, is the latitude coordinate, is the direction speed, is the heading angle of the target, is the length information of the target, is the width information of the target, To add new vehicle colors, To add a new license plate.

[0077] b) Non-overlapping region, when the labeled array has only a single detector, the fused array The data source is switched to this detector.

[0078] Furthermore, in a vehicle tracking system for a detector network fusion under an intersection environment provided in this embodiment, the tracking module includes a fourth generation unit and a tracking unit, wherein the fourth generation unit is used to generate a fusion Arrays and One-shot Fusion The array performs secondary data fusion to generate an intersection network fusion array; the tracking unit is used to draw the complete vehicle movement trajectory in real time based on the intersection network fusion array to complete target tracking.

[0079] The fourth generation unit traversal Multiple fusion arrays of the network detector at this moment After a second data fusion, the final intersection network fusion array is generated as follows: (30) In formula (30), To network and fuse arrays for intersections, For a fusion Arrays, For a fusion Arrays, For a fusion Array.

[0080] The tracking unit calculates the big data intersection network fusion array , the complete vehicle motion trajectory can be drawn in real time to achieve the tracking purpose.

[0081] The vehicle tracking method and system of the detector network fusion in the intersection environment provided in this embodiment, compared with the prior art, establishes a unified longitude and latitude coordinate system based on the layout positions of the networked radar detector and the video detector; obtains the first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection detected by the networked radar detector and the video detector respectively according to the unified longitude and latitude coordinate system; performs secondary data fusion on the first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection to complete target tracking. The vehicle tracking method and system of the detector network fusion in the intersection environment provided in this embodiment, which involves the vehicle trajectory display in the field of intelligent traffic signal control, mainly uses the fusion of multiple radars and video detectors in the intersection environment to track vehicles, automatically generates unique vehicle trajectory information by establishing a unified longitude and latitude coordinate system of multiple sensors, vehicle data fusion between radars and video detectors, networking secondary data fusion, trajectory prediction and other technologies, solves the problem of multi-path generation and loss of tracking targets, has a high degree of automation and strong anti-interference ability; and has high target tracking accuracy and precision.

[0082] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A vehicle tracking method using a detector network fusion in an intersection environment, characterized in that: The following steps are involved: Based on the layout positions of the networked radar detectors and video detectors, a unified longitude and latitude coordinate system is established; According to the unified longitude and latitude coordinate system, the primary fusion data of the same-side intersection and the primary fusion data of the opposite-side intersection detected by the networked radar detector and the video detector are respectively obtained; A secondary data fusion is performed on the first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection to complete target tracking.

2. The vehicle tracking method of detector network fusion in an intersection environment as claimed in claim 1, characterized in that: The step of establishing a unified longitude and latitude coordinate system based on the arrangement positions of the networked radar detectors and the video detectors comprises: According to the Beidou positioning system, the layout positions of the networked radar detectors and video detectors are spatially located to obtain the longitude and latitude coordinates; According to the longitude and latitude coordinates, the origin of the coordinate system is established with the radar deployment position, and the speed close to the origin is defined as positive, and the speed away from the origin is defined as negative; the longitude and latitude of the reference object in the coordinate system are aligned to establish a unified longitude and latitude coordinate system.

3. The vehicle tracking method of detector network fusion in an intersection environment as claimed in claim 2, characterized in that: The latitude and longitude coordinates include longitude coordinates and latitude coordinates According to the unified longitude and latitude coordinate system, the steps of respectively obtaining the once fused data of the same-side intersection and the once fused data of the opposite-side intersection of the detector network include: Calibrate the unified longitude and latitude coordinate system and perform data synchronization processing on the networked radar detector and the video detector; Based on longitude coordinates and latitude coordinates , calculate ; According to the The vehicle state data is respectively predicted by using a radar detector and a video detector to predict the vehicle trajectory to generate vehicle trajectory prediction data; the vehicle trajectory prediction data includes vehicle trajectory prediction data for the same-side intersection and vehicle trajectory prediction data for the opposite-side intersection; the vehicle trajectory prediction data for the same-side intersection includes first vehicle trajectory prediction data for the same-side intersection generated by the radar detector and second vehicle trajectory prediction data for the same-side intersection generated by the video detector, and the vehicle trajectory prediction data for the opposite-side intersection includes first vehicle trajectory prediction data for the opposite-side intersection generated by the radar detector and second vehicle trajectory prediction data for the opposite-side intersection generated by the video detector; The first same-side intersection vehicle trajectory prediction data and the second same-side intersection vehicle trajectory prediction data are fused to generate a fusion of the same-side intersection fusion data. Arrays; The first opposite-side intersection vehicle trajectory prediction data and the second opposite-side intersection vehicle trajectory prediction data are fused to generate a fusion of the opposite-side intersection fusion data. Array.

4. The vehicle tracking method of detector network fusion in an intersection environment as claimed in claim 3, characterized in that: The steps of performing secondary data fusion on the primary fusion data of the same-side intersection and the primary fusion data of the opposite-side intersection to complete target tracking include: For the first fusion The array and the once merged The array performs secondary data fusion to generate an intersection network fusion array; According to the intersection networking fusion array, the complete vehicle movement trajectory is drawn in real time to complete the target tracking.

5. The vehicle tracking method of detector network fusion in an intersection environment as claimed in claim 4, characterized in that: The intersection network fusion array is: in, To network and fuse arrays for intersections, For a fusion Arrays, For a fusion Arrays, For a fusion Array.

6. A vehicle tracking system with detector network fusion in an intersection environment, characterized in that: include: Establishing a module for establishing a unified longitude and latitude coordinate system based on the layout positions of the networked radar detectors and the video detectors; An acquisition module, used to acquire the primary fusion data of the same-side intersection and the primary fusion data of the opposite-side intersection detected by the networked radar detector and the video detector respectively according to the unified longitude and latitude coordinate system; The tracking module is used to perform secondary data fusion on the first fusion data of the same-side intersection and the first fusion data of the opposite-side intersection to complete target tracking.

7. The vehicle tracking system with detector network fusion in the intersection environment as claimed in claim 6, characterized in that: The establishment module includes: A positioning unit is used to spatially locate the layout of the networked radar detector and the video detector according to the Beidou positioning system to obtain the longitude and latitude coordinates; The establishing unit is used to establish the origin of the coordinate system based on the longitude and latitude coordinates and the radar deployment position, define the speed close to the origin as positive and the speed away from the origin as negative; align the longitude and latitude of the reference object in the coordinate system to establish a unified longitude and latitude coordinate system.

8. The vehicle tracking system with detector network fusion in the intersection environment as claimed in claim 7, characterized in that: The acquisition module comprises: A synchronization processing unit, used to calibrate the unified longitude and latitude coordinate system and perform data synchronization processing on the networked radar detector and the video detector; Calculation unit for longitude coordinates and latitude coordinates , calculate ; The first generating unit is used to generate The vehicle state data is respectively predicted by using a radar detector and a video detector to predict the vehicle trajectory to generate vehicle trajectory prediction data; the vehicle trajectory prediction data includes vehicle trajectory prediction data for the same-side intersection and vehicle trajectory prediction data for the opposite-side intersection; the vehicle trajectory prediction data for the same-side intersection includes first vehicle trajectory prediction data for the same-side intersection generated by the radar detector and second vehicle trajectory prediction data for the same-side intersection generated by the video detector, and the vehicle trajectory prediction data for the opposite-side intersection includes first vehicle trajectory prediction data for the opposite-side intersection generated by the radar detector and second vehicle trajectory prediction data for the opposite-side intersection generated by the video detector; The second generating unit is used to fuse the first same-side intersection vehicle trajectory prediction data and the second same-side intersection vehicle trajectory prediction data to generate a fusion of the same-side intersection fusion data. Arrays; The third generating unit is used to fuse the first opposite-side intersection vehicle trajectory prediction data and the second opposite-side intersection vehicle trajectory prediction data to generate a fusion of the opposite-side intersection fusion data. Array.

9. The vehicle tracking system with detector network fusion in the intersection environment as claimed in claim 8, characterized in that: The tracking module includes: The fourth generating unit is used to generate the first fusion The array and the once merged The array performs secondary data fusion to generate an intersection network fusion array; The tracking unit is used to draw the complete vehicle movement trajectory in real time according to the intersection networking fusion array to complete the target tracking.

10. The vehicle tracking system with detector network fusion in the intersection environment as claimed in claim 9, characterized in that: The intersection network fusion array is: in, To network and fuse arrays for intersections, For a fusion Arrays, For a fusion Arrays, For a fusion Array.

Images