A method, device, and edge computing node for processing vehicle recognition data

Abstract

The present invention provides a method, an apparatus and an edge computing node for processing vehicle recognition data. The method for processing vehicle recognition data includes: obtaining first vehicle-side vehicle data of a first vehicle, second vehicle-side vehicle data of a second vehicle, and roadside perception data of each roadside vehicle; obtaining second vehicle estimated recognition data on the vehicle side, first vehicle estimated recognition data on the roadside, and second vehicle estimated recognition data on the roadside according to the first vehicle-side vehicle data, the second vehicle-side vehicle data, and the roadside perception data; and obtaining recognition data of the first vehicle and recognition data of the second vehicle according to the first vehicle-side vehicle data, the second vehicle estimated recognition data on the vehicle side, the first vehicle estimated recognition data on the roadside, the second vehicle estimated recognition data on the roadside, the vehicle-side weight value, and the roadside weight value. The solution of the present invention can improve the accuracy of recognizing connected vehicles (the first vehicle) and the second vehicle (other target objects).

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent transportation, and particularly relates to a method, an apparatus, and an edge computing node for processing vehicle recognition data. Background Art

[0002] Currently, in vehicle-road collaborative applications, the perception, recognition, tracking, and visual presentation of traffic participating targets play a key role in upper-layer application scenarios and user driving behavior guidance. In the presentation of the human-machine interface (HMI) of a connected host vehicle carrying an on-board unit (OBU), no duplicate removal processing of the host vehicle is performed, which will result in the user seeing the effect that the host vehicle follows, runs parallel to, or trails a vehicle, and this vehicle is caused by the road-side perception device identifying, tracking, and fusing and then giving it to the road-side unit (RSU) to broadcast through the direct communication interface PC5. Such a presentation effect will, at best, affect the user's driving experience, and at worst, cause misunderstandings, induce changes in driving actions, and lead to safety accidents.

[0003] The existing vehicle duplicate removal methods mainly compare the host vehicle information with the trajectory information of the vehicles sensed by the road-side unit through the OBU or the road-side unit of the connected vehicle to identify the connected vehicle and other targets (other vehicles). However, this requires a high degree of accuracy in the road-side unit's identification of the vehicle's trajectory information and positioning accuracy. Therefore, the accuracy of identifying the connected vehicle and other targets (other vehicles) is low. Summary of the Invention

[0004] Embodiments of the present invention provide a method, an apparatus, and an edge computing node for processing vehicle recognition data to solve the problem of low accuracy in identifying connected vehicles and other targets (other vehicles) in the existing technology.

[0005] To solve the above technical problems, the embodiments of the present invention provide the following technical solutions:

[0006] Embodiments of the present invention provide a method for processing vehicle recognition data, which is characterized by including:

[0007] Obtaining the first vehicle-side vehicle data of the first vehicle and the second vehicle-side vehicle data of the second vehicle sent by the first vehicle, and obtaining the road-side perception data of each road-side vehicle sent by the road-side unit RSU;

[0008] According to the first vehicle-side vehicle data, the second vehicle-side vehicle data, and the road-side perception data, obtaining the estimated recognition data of the second vehicle on the vehicle side, the estimated recognition data of the first vehicle on the road side, and the estimated recognition data of the second vehicle on the road side;

[0009] Based on the first vehicle-side vehicle data, the second vehicle estimation and recognition data of the vehicle side, the first vehicle estimation and recognition data of the roadside, the second vehicle estimation and recognition data of the roadside, the vehicle-side weight value, and the roadside weight value, obtain the recognition data of the first vehicle and the recognition data of the second vehicle;

[0010] Wherein, the first vehicle is a vehicle carrying an on-vehicle unit OBU, and the second vehicle is a vehicle within a first preset range of the first vehicle.

[0011] An embodiment of the present invention further provides a vehicle recognition data processing device, including:

[0012] A data acquisition module, configured to acquire the first vehicle-side vehicle data of the first vehicle and the second vehicle-side vehicle data of the second vehicle sent by the first vehicle, and acquire the roadside perception data of each roadside vehicle sent by a roadside unit RSU;

[0013] A first processing module, configured to obtain the second vehicle estimation and recognition data of the vehicle side, the first vehicle estimation and recognition data of the roadside, and the second vehicle estimation and recognition data of the roadside according to the first vehicle-side vehicle data, the second vehicle-side vehicle data, and the roadside perception data;

[0014] A second processing module, configured to obtain the recognition data of the first vehicle and the recognition data of the second vehicle according to the first vehicle-side vehicle data, the second vehicle estimation and recognition data of the vehicle side, the first vehicle estimation and recognition data of the roadside, the second vehicle estimation and recognition data of the roadside, the vehicle-side weight value, and the roadside weight value;

[0015] Wherein, the first vehicle is a vehicle carrying an on-vehicle unit OBU, and the second vehicle is a vehicle within a first preset range of the first vehicle.

[0016] An embodiment of the present invention further provides an edge computing node, including: a processor, a memory, and a program stored on the memory and executable on the processor, and when the program is executed by the processor, the steps of the vehicle recognition data processing method described in any one of the above are implemented.

[0017] An embodiment of the present invention further provides a readable storage medium, on which a program is stored, and when the program is executed by a processor, the steps in the vehicle recognition data processing method described in any one of the above are implemented.

[0018] The beneficial effects of the present invention are:

[0019] The vehicle identification data processing method provided by the embodiment of the present invention obtains the first vehicle-side vehicle data of the first vehicle sent by the first vehicle and the second vehicle-side vehicle data of the second vehicle located near the first vehicle, and obtains the roadside perception data of each roadside vehicle sent by the roadside unit RSU. According to the first vehicle-side vehicle data, the second vehicle-side vehicle data, and the roadside perception data, the estimated identification data of the second vehicle on the vehicle side, the estimated identification data of the first vehicle on the roadside, and the estimated identification data of the second vehicle on the roadside are obtained. According to the first vehicle-side vehicle data, the estimated identification data of the second vehicle on the vehicle side, the estimated identification data of the first vehicle on the roadside, the estimated identification data of the second vehicle on the roadside, the vehicle-side weight value, and the roadside weight value, the identification data of the first vehicle and the identification data of the second vehicle are obtained, which can improve the accuracy of identifying the connected vehicle (the first vehicle) and the second vehicle (other target objects). BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It represents the flowchart of the vehicle identification data processing method provided by the embodiment of the present invention;

[0021] Figure 2 It represents the schematic diagram of the positions of the first vehicle and the second vehicle provided by the embodiment of the present invention;

[0022] Figure 3 It represents the schematic diagram of the relationship between the vehicle-side weight value and the roadside weight value provided by the embodiment of the present invention;

[0023] Figure 4 It represents the specific flowchart of the vehicle identification data processing method provided by the embodiment of the present invention;

[0024] Figure 5 It represents the schematic diagram of the structure of the vehicle identification data processing device provided by the embodiment of the present invention;

[0025] Figure 6 It represents the schematic diagram of the structure of the edge computing node provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0027] The terms "first", "second", etc. in the specification of the present invention are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such terms can be interchanged under appropriate circumstances, so that the embodiments of the present invention can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same type, and do not limit the number of objects. For example, the first object can be one or more. In addition, "and / or" in the specification means at least one of the connected objects, and the character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0028] Before describing the specific embodiments of the present invention, the following explanations are made first:

[0029] There are mainly two known methods for removing the duplicate of the host vehicle itself. One is that the network-connected host vehicle carrying the OBU compares the roadside vehicle perception information received by the RSU broadcast with its own vehicle information for duplicate removal. It mainly relies on the longitude and latitude information and the continuity of the trajectory for judgment. However, it has high requirements for the recognition accuracy and positioning accuracy of the roadside perception system, and in case of occlusion, the data will be lost, affecting the judgment result. The implementation difficulty is relatively large, and it will increase the workload of the network-connected host vehicle carrying the OBU. The other is to perform the duplicate removal process of the host vehicle itself through roadside equipment, mainly relying on the roadside equipment to judge by comparing the longitude and latitude information and the continuity of the trajectory of the network-connected host vehicle carrying the OBU and the vehicles recognized by the roadside perception system. Compared with placing the decision-making end at the vehicle end, it can reduce the workload of the vehicle-end equipment. However, it not only strongly depends on the recognition accuracy and positioning accuracy of the roadside perception system, but also affects the presentation requirements of the vehicle-road collaborative cloud control platform for all detected targets.

[0030] To solve the problem of low recognition accuracy of identifying network-connected vehicles and other targets (other vehicles) in the existing technology, the embodiments of the present invention provide a vehicle recognition data processing method, device, and edge computing node.

[0031] As Figure 1 shown, the embodiments of the present invention provide a vehicle recognition data processing method, including:

[0032] Step 101: Obtain the first vehicle-side vehicle data of the first vehicle and the second vehicle-side vehicle data of the second vehicle sent by the first vehicle, and obtain the roadside perception data of each roadside vehicle sent by the roadside unit RSU; wherein, the first vehicle is a vehicle carrying an on-vehicle unit OBU, that is, the first vehicle is a network-connected vehicle, and the second vehicle is a vehicle within a first preset range of the first vehicle, that is, the second vehicle is a vehicle around the first vehicle.

[0033] It should be noted that the vehicle recognition data processing method provided by the embodiments of the present invention is executed by a vehicle-road collaborative cloud control platform, which is a mobile edge computing (MEC) node.

[0034] In this step, the MEC obtains the first vehicle-side vehicle data of the first vehicle and the second vehicle-side vehicle data of the second vehicle sent by the OBU, and obtains the roadside perception data of each roadside vehicle sent by the RSU corresponding to the first vehicle. The vehicle-side vehicle data and the roadside perception data are used as the data input part of the MEC. Among them, the vehicle-side vehicle data is also called "Object_veh", which mainly comes from connected vehicles equipped with OBU and having sensing capabilities. The roadside perception data is also called "Object_road", which mainly comes from the structured data after the fusion of various roadside sensors.

[0035] Exemplarily, please refer to Figure 2 , according to the vehicle-side vehicle data and the roadside perception data, it can be known that there will be several vehicles, such as Figure 2 shown, HV represents the host vehicle (the first vehicle), the B-type bus, including B1, B2, and B3, the T-type vehicle is a truck, including T1 and T2, and the C-type vehicle is a car, including C1, C2, C3, C4, C5, C6, C7, C8, and C9. For the vehicle-side vehicle data sent by the first vehicle, there may be blind spots for C4, C5, and C6. For the roadside perception data, there may be large vehicle occlusion for C1.

[0036] The data in this step comes from the vehicle side and the roadside, which improves the structural diversity of the data.

[0037] Step 102: Obtain the second vehicle estimated recognition data on the vehicle side, the first vehicle estimated recognition data on the roadside, and the second vehicle estimated recognition data on the roadside according to the first vehicle-side vehicle data, the second vehicle-side vehicle data, and the roadside perception data.

[0038] After receiving the vehicle - side vehicle data and roadside perception data sent by the first vehicle, in this step, the first vehicle - side vehicle data and the second vehicle - side vehicle data are matched to obtain the second vehicle estimation and recognition result on the vehicle side. The first vehicle - side vehicle data, the second vehicle - side vehicle data, and the roadside perception data are matched to obtain the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, and the second vehicle estimation and recognition data on the roadside. That is, after matching the first vehicle - side vehicle data, the second vehicle - side vehicle data, and the roadside perception data, the host vehicle (the first vehicle) and other target vehicle (the second vehicle) in the vehicle - side vehicle data can be obtained more accurately through the first vehicle - side vehicle data and the second vehicle estimation and recognition data on the vehicle side, and the host vehicle (the first vehicle) and other target vehicle (the second vehicle) in the roadside perception data can be confirmed more accurately through the first vehicle estimation and recognition data on the roadside and the second vehicle estimation and recognition data on the roadside.

[0039] Step 103: Obtain the recognition data of the first vehicle and the recognition data of the second vehicle according to the first vehicle - side vehicle data, the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, the second vehicle estimation and recognition data on the roadside, the vehicle - side weight value, and the roadside weight value.

[0040] Among them, the vehicle - side weight value and the roadside weight value are determined according to the first vehicle - side vehicle data and the second vehicle - side vehicle data.

[0041] After obtaining the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, and the second vehicle estimation and recognition data on the roadside, in this step, the vehicle - side weight value and the roadside weight value are determined by using the first vehicle - side vehicle data and the second vehicle - side vehicle data. According to the vehicle - side weight value and the roadside weight value, the first vehicle - side vehicle data, the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, and the second vehicle estimation and recognition data on the roadside are processed by using the weighted average method to obtain the recognition data of the first vehicle and the recognition data of the second vehicle.

[0042] After processing the first vehicle - side vehicle data, the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, and the second vehicle estimation and recognition data on the roadside by using the weighted average method, the recognition accuracy of the connected vehicle (the first vehicle) and the recognition accuracy of other target vehicles (the second vehicle) can be further improved.

[0043] It should be noted that for the vehicle - side vehicle data, the basic requirements for the data structure are as follows. That is, the vehicle - side vehicle data includes the following sub - data: data sending timestamp, data number (this data number is the number of the data - collecting vehicle, that is, the target object number, which is also the number of the first vehicle), vehicle longitude data, vehicle latitude data, vehicle type (including motor vehicles and non - motor vehicles), vehicle model (including large vehicles and small vehicles), body color (including yellow, white, black, gray, blue, red, and other colors), vehicle size data (vehicle size data includes vehicle length data, vehicle width data, and vehicle height data), license plate number, vehicle speed, vehicle heading angle, OBU carrying flag bit (master vehicle flag bit), and may also include other information (temporary identity document (ID), acceleration, vehicle altitude, etc.).

[0044] The master vehicle flag bit is used to indicate whether the vehicle is the first vehicle or the second vehicle.

[0045] Among them, assume that in the vehicle - side vehicle data, the data sending timestamp is veh_timestap, the data number is veh_idα, the vehicle longitude data is veh_lonα, the vehicle latitude data is veh_latα, the vehicle type is veh_xα, the vehicle model is veh_yα, the body color is veh_colorα, the vehicle length data is veh_lα, the vehicle width data is veh_wα, the vehicle height data is veh_hα, the license plate number is veh_numbα, the vehicle speed is veh_speedα, the vehicle heading angle is veh_headingα, and the master vehicle flag bit is veh_hvflag.

[0046] For the roadside perception data, the basic requirements for the data structure are as follows. That is, the roadside perception data includes the following sub - data: data sending timestamp, data number (this data number is the number of the data - collecting object, that is, the target object number, which is also the number of the RSU), vehicle longitude data, vehicle latitude data, vehicle type (including motor vehicles and non - motor vehicles), vehicle model (including large vehicles and small vehicles), body color (including yellow, white, black, gray, blue, red, and other colors), vehicle size data (vehicle size data includes vehicle length data, vehicle width data, and vehicle height data), license plate number, vehicle speed, vehicle heading angle, and may also include other information (acceleration, vehicle altitude, etc.).

[0047] Among them, it is assumed that in the roadside perception data, the data sending timestamp is road_timestap, the data number is road_idα, the vehicle longitude data is road_lonα, the vehicle latitude data is road_latα, the vehicle type is road_xα, the vehicle model is road_yα, the body color is road_colorα, the vehicle length data is road_lα, the vehicle width data is road_wα, the vehicle height data is road_hα, the license plate number is road_numbα, the vehicle speed is road_speedα, and the vehicle heading angle is road_headingα.

[0048] The estimated recognition data and the finally processed recognition data also include the above-mentioned sub-data.

[0049] In an alternative embodiment of the present invention, after obtaining the vehicle-side vehicle data and the roadside perception data, a determination is made on the output of the vehicle-side vehicle data and the roadside perception data, mainly judging from the integrity of the data. If there is a lack of basic data (sub-data) in the data, the data is discarded. If the integrity requirement is met, subsequent recognition data processing is performed. Specifically, in the case where the first target vehicle-side vehicle data and the first target roadside perception data do not include the first target sub-data, the first target vehicle-side vehicle data and the first target roadside perception data are deleted; wherein, the first target vehicle-side vehicle data is the second vehicle-side vehicle data of any second vehicle, the first target roadside perception data is the roadside perception data of any roadside vehicle; the first target sub-data is one or more of the sub-data.

[0050] In an alternative embodiment of the present invention, step 102 includes:

[0051] Using the first vehicle-side vehicle data and each piece of roadside perception data, determine the third vehicle among the roadside vehicles that is most similar to the first vehicle. Specifically, since the first vehicle-side vehicle data corresponding to the first vehicle includes the host vehicle flag bit veh_hvflag, the sub-data in the first vehicle-side vehicle data can be determined according to the host vehicle flag bit: the data transmission timestamp is veh_timestap, the data number is veh_idα, the vehicle longitude data is veh_lonα, the vehicle latitude data is veh_latα, the vehicle type is veh_xα, the vehicle model is veh_yα, the body color is veh_colorα, the vehicle length data is veh_lα, the vehicle width data is veh_wα, the vehicle height data is veh_hα, the license plate number is veh_numbα, the vehicle speed is veh_speedα, the vehicle heading angle is veh_headingα. Traverse and compare the above sub-data with each sub-data in the roadside perception data to find the suspected host vehicle in the roadside perception data, that is, the third vehicle most similar to the first vehicle, and obtain the roadside perception data of the third vehicle. Among them, the sub-data used in the roadside perception data includes: the data transmission timestamp is road_timestap, the data number is road_idα, the vehicle longitude data is road_lonα, the vehicle latitude data is road_latα, the vehicle type is road_xα, the vehicle model is road_yα, the body color is road_colorα, the vehicle length data is road_lα, the vehicle width data is road_wα, the vehicle height data is road_hα, the license plate number is road_numbα, the vehicle speed is road_speedα, the vehicle heading angle is road_headingα.

[0052] According to the vehicle longitude data and vehicle latitude data in the first vehicle-side vehicle data and the vehicle longitude data and vehicle latitude data in the second vehicle-side vehicle data, determine the first distance between the first vehicle and each second vehicle, that is, according to the longitude and latitude data of the vehicle in the vehicle-side vehicle data, the relative position and the first distance S between the first vehicle and each second vehicle in the vehicle-side vehicle data can be obtained;

[0053] According to the vehicle longitude data and vehicle latitude data in the roadside perception data of the third vehicle and the vehicle longitude data and vehicle latitude data in the roadside perception data of the fourth vehicle, determine the second distance between the third vehicle and each fourth vehicle, where the fourth vehicle is a vehicle among the roadside vehicles that is within a second preset range of the third vehicle, that is, the relative position and the second distance L between the suspected host vehicle and each fourth vehicle near the suspected host vehicle in the roadside perception data can be obtained;

[0054] It should be noted that both the first preset range and the second preset range can be set according to requirements, and the first preset range and the second preset range can be the same or different.

[0055] Based on the first distance, the second distance, the second vehicle data on the vehicle side, the roadside perception data of the third vehicle, and the roadside perception data of the fourth vehicle, the initial estimated identification data of the second vehicle on the vehicle side, the initial estimated identification data of the first vehicle on the roadside, and the initial estimated identification data of the second vehicle on the roadside are obtained. Specifically, the first distance and the second distance are considered, and feature matching is performed on the second vehicle data on the vehicle side and the roadside perception data of the fourth vehicle. If the matching is successful, it is retained; if the matching is unsuccessful, it is discarded, thereby obtaining the initial estimated identification data of the second vehicle on the vehicle side and the initial estimated identification data of the second vehicle on the roadside, and using the roadside perception data of the third vehicle as the initial estimated identification data of the first vehicle on the roadside.

[0056] Among them, the initial estimated identification data also includes the above-mentioned sub-data.

[0057] Based on the initial estimated identification data of the second vehicle on the vehicle side, the initial estimated identification data of the first vehicle on the roadside, the initial estimated identification data of the second vehicle on the roadside, and the first vehicle data on the vehicle side, the estimated identification data of the second vehicle on the vehicle side, the estimated identification data of the first vehicle on the roadside, and the estimated identification data of the second vehicle on the roadside are obtained. Specifically, after feature matching, secondary special attribute matching is performed on the first vehicle data on the vehicle side and the initial estimated identification data of the first vehicle on the roadside. The data with unsuccessful matching is discarded, and the data with successful matching is retained to obtain the initial estimated identification data of the first vehicle on the roadside. Secondary special attribute matching is performed on the initial estimated identification data of the second vehicle on the vehicle side and the initial estimated identification data of the second vehicle on the roadside. The data with unsuccessful matching is discarded, and the data with successful matching is retained to obtain the estimated identification data of the second vehicle on the vehicle side and the estimated identification data of the second vehicle on the roadside.

[0058] Furthermore, the process of performing feature matching on the second vehicle data on the vehicle side and the roadside perception data of the fourth vehicle, that is, the process of obtaining the estimated identification data of the second vehicle on the vehicle side, the estimated identification data of the first vehicle on the roadside, and the estimated identification data of the second vehicle on the roadside based on the first distance, the second distance, the second vehicle data on the vehicle side, the roadside perception data of the third vehicle, and the roadside perception data of the fourth vehicle, includes:

[0059] When the difference between the first distance S corresponding to the first target vehicle in the second vehicle and the second distance L corresponding to the second target vehicle in the fourth vehicle is less than or equal to a first preset distance (optionally, the first preset distance is 0.5 meters, preferably, the first preset distance is 0.2 meters), perform the Hungarian matching on the second vehicle-side vehicle data of the first target vehicle and the roadside perception data of the second target vehicle. In the case of successful matching, use the second vehicle-side vehicle data of the first target vehicle as the estimated recognition data of the second vehicle on the vehicle side, and use the roadside perception data of the second target vehicle as the estimated recognition data of the second vehicle on the roadside.

[0060] Specifically, when the first condition is satisfied, it is determined that the matching between the second vehicle-side vehicle data of the first target vehicle and the roadside perception data of the second target vehicle is successful;

[0061] Among them, the first condition includes one or more of the following:

[0062] The absolute value of the difference between the vehicle longitude data in the second vehicle-side vehicle data of the first target vehicle and the vehicle longitude data in the roadside perception data of the second target vehicle is less than a second preset distance; where the second preset distance is 0.5 meters;

[0063] The absolute value of the difference between the vehicle latitude data in the second vehicle-side vehicle data of the first target vehicle and the vehicle latitude data in the roadside perception data of the second target vehicle is less than a second preset distance; where the second preset distance is 0.5 meters;

[0064] The absolute value of the difference between the vehicle heading angle in the second vehicle-side vehicle data of the first target vehicle and the vehicle heading angle in the roadside perception data of the second target vehicle is less than a first preset angle; where the first preset angle is 5°.

[0065] The absolute value of the difference between the vehicle speed in the second vehicle-side vehicle data of the first target vehicle and the vehicle speed in the roadside perception data of the second target vehicle is less than a first preset speed; where the first preset speed is 1 m / s.

[0066] The absolute value of the difference between the vehicle size data in the second vehicle-side vehicle data of the first target vehicle and the vehicle size data in the roadside perception data of the second target vehicle is less than a first preset size. Where the first preset size is 0.5 m.

[0067] It should be noted that when matching the longitude and latitude data, the longitude and latitude data are converted into position data before comparing the errors to determine whether the absolute value of the difference is less than the second preset distance.

[0068] Perform the Hungarian matching on the vehicle longitude data, vehicle latitude data, vehicle heading angle, vehicle speed, and vehicle size data in the sub-data, retain the successfully matched data, and discard the failed-matched data to obtain the second vehicle estimated recognition data on the vehicle side and the second vehicle estimated recognition data on the roadside.

[0069] Further, the process of performing secondary attribute special matching on the data, that is, obtaining the second vehicle estimated recognition data on the vehicle side, the first vehicle estimated recognition data on the roadside, and the second vehicle estimated recognition data on the roadside according to the second vehicle initial estimated recognition data on the vehicle side, the first vehicle initial estimated recognition data on the roadside, the second vehicle initial estimated recognition data on the roadside, and the first vehicle side vehicle data, includes:

[0070] When the second target sub-data in the second vehicle initial estimated recognition data on the vehicle side matches successfully with the second target sub-data in the second vehicle initial estimated recognition data on the roadside, use the second vehicle initial estimated recognition data on the vehicle side as the second vehicle estimated recognition data on the vehicle side, and use the second vehicle initial estimated recognition data on the roadside as the second vehicle estimated recognition data on the roadside;

[0071] When the second target sub-data in the first vehicle initial estimated recognition data on the roadside matches successfully with the second target sub-data in the first vehicle side vehicle data, use the first vehicle initial estimated recognition data on the roadside as the first vehicle estimated recognition data on the roadside;

[0072] Wherein, the second target sub-data includes at least one of the following:

[0073] Vehicle type; vehicle model; body color; license plate number.

[0074] It should also be noted that the second target sub-data may further include vehicle altitude and / or vehicle acceleration.

[0075] On the basis of feature matching, the purpose of performing secondary special attribute matching is to ensure the accuracy and precision of vehicle matching. Performing secondary special attribute matching includes, but is not limited to, matching of vehicle type, vehicle model, body color, license plate number, vehicle altitude, and vehicle acceleration.

[0076] Exemplarily, the specific method of secondary special attribute matching is:

[0077] When the vehicle type in the vehicle - side data (including the first vehicle - side vehicle data and the second vehicle initial estimation and recognition data on the vehicle side) matches the vehicle type in the roadside data (including the second vehicle initial estimation and recognition data on the roadside and the first vehicle initial estimation and recognition data on the roadside), veh_xn&road_xn is set to 1; when they do not match, veh_xn&road_xn is set to 0. Here, the vehicle types are divided into motor vehicles and non - motor vehicles. In actual operations, veh_xn and road_xn can also be taken in hexadecimal;

[0078] When the vehicle model in the vehicle - side data (including the first vehicle - side vehicle data and the second vehicle initial estimation and recognition data on the vehicle side) matches the vehicle model in the roadside data (including the second vehicle initial estimation and recognition data on the roadside and the first vehicle initial estimation and recognition data on the roadside), veh_yn&road_yn is set to 1; when they do not match, veh_yn&road_yn is set to 0. Here, the vehicle models are divided into large vehicle models and small vehicle models. In actual operations, veh_yn and road_yn can also be taken in hexadecimal;

[0079] When the body color in the vehicle - side data (including the first vehicle - side vehicle data and the second vehicle initial estimation and recognition data on the vehicle side) matches the body color in the roadside data (including the second vehicle initial estimation and recognition data on the roadside and the first vehicle initial estimation and recognition data on the roadside), veh_colorn&road_colorn is set to 1; when they do not match, veh_color n&road_color n is set to 0. Here, the body colors are divided into yellow, white, black, gray, blue, red, and other colors, which are represented by 1, 2, 3, 4, 5, 6, 7 respectively. In actual operations, the above colors can also be taken in hexadecimal;

[0080] When the license plate number in the vehicle - side data (including the first vehicle - side vehicle data and the second vehicle initial estimation and recognition data on the vehicle side) matches the license plate number in the roadside data (including the second vehicle initial estimation and recognition data on the roadside and the first vehicle initial estimation and recognition data on the roadside), veh_numbn&road_numbn is set to 1; when they do not match, veh_numbn&road_numbn is set to 0. Here, in actual operations, the license plate number can also be taken in hexadecimal;

[0081] After the above judgment, determine the value of the judgment value n = (veh_xn & road_xn) & (veh_yn & road_yn) & (veh_colorn & road_colorn) & (veh_numbn & road_numbn), where & is the "AND" operation, that is, (veh_xn & road_xn) & (veh_yn & road_yn) & (veh_colorn & road_color n) & (veh_numbn & road_numbn) is equal to the sum of veh_xn & road_xn, veh_yn & road_yn, veh_colorn & road_colorn, and veh_numbn & road_numbn. If the value of the judgment value n is greater than the first threshold, the vehicle-side data and the roadside data match in secondary special attributes, and the vehicle-side data and the roadside data are retained. If the value of the judgment value n is less than or equal to the first threshold, it is determined that the vehicle-side data and the roadside data do not match in secondary special attributes, and the vehicle-side data and the roadside data are discarded.

[0082] After the secondary special attribute matching, the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, and the second vehicle estimation and recognition data on the roadside can be obtained more accurately. Furthermore, based on the first vehicle-side vehicle data, the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, and the second vehicle estimation and recognition data on the roadside, the host vehicle (the first vehicle) and other target vehicle (the second vehicle) in the vehicle-side data and the roadside data can be identified.

[0083] In an alternative embodiment of the present invention, before weighting the vehicle-side data (including the first vehicle-side vehicle data and the second vehicle estimation and recognition data on the vehicle side) and the roadside data (including the first vehicle estimation and recognition data on the roadside and the second vehicle estimation and recognition data on the roadside) using the vehicle-side weight value and the roadside weight value, the first distance S between the first vehicle and each second vehicle is used to determine the vehicle-side weight value and the roadside weight value. Specifically, the method for determining the vehicle-side weight value and the roadside weight value is as follows:

[0084] When the position of the host vehicle (the first vehicle) is known, according to the first distance S between the first vehicle and each second vehicle, determine the weight distribution ratio, where K 车 +K 路 = 1; K 车 represents the vehicle-side weight value, and K 路 represents the roadside weight value.

[0085] When the first distance S between the first vehicle and the second vehicle is from 0 to S1, K 车 = (-3)*S / 4 + 0.75;

[0086] When the first distance S between the first vehicle and the second vehicle is from S1 to S2, K 车 = 0.3;

[0087] When the first distance S between the first vehicle and the second vehicle is from S2 to S3, K 车 = (-2)*S / 3 + 0.6.

[0088] Then use the above formula to find K 车 , and then according to K 车 + K 路 = 1 to find K 路 .

[0089] Among them, the values of S1, S2, and S3 can be set according to requirements.

[0090] The schematic diagram of the relationship between the vehicle side weight value and the roadside weight value is as Figure 3 shown.

[0091] In an optional embodiment of the present invention, step 103 includes:

[0092] According to the third target data in the first vehicle side vehicle data, the corresponding third target data in the first vehicle estimated recognition data on the roadside, the vehicle side weight value, and the roadside weight value, obtain a first data difference; wherein, the third target data is one of the following: vehicle longitude data; vehicle latitude data; vehicle size data (including vehicle length data, vehicle width data, and vehicle height data).

[0093] Specifically, according to the vehicle longitude data in the first vehicle side vehicle data, the vehicle longitude data in the first vehicle estimated recognition data on the roadside, the vehicle side weight value, and the roadside weight value, obtain a longitude intermediate value, and according to the vehicle longitude data in the first vehicle side vehicle data and the vehicle longitude data in the first vehicle estimated recognition data on the roadside, obtain a longitude average value. The first data difference corresponding to the vehicle longitude data is the difference between the longitude intermediate value and the longitude average value. The formula is as follows:

[0094] δ_lonα = Med_lonα - aver_lonα

[0095] Med_lonα = (car_lonα * K 车 + road_lonα * K 路 ) / 1

[0096] aver_lonα = (car_lonα + road_lonα) / 2

[0097] Among them, δ_lonα represents the first data difference corresponding to the vehicle longitude data, Med_lonα represents the longitude median, aver_lonα represents the longitude average, car_lonα represents the vehicle longitude data in the first vehicle data on the vehicle side, road_lonα represents the vehicle longitude data in the first vehicle estimated recognition data on the roadside, and K 车 represents the vehicle side weight value, and K 路 represents the roadside weight value.

[0098] The latitude median is obtained based on the vehicle latitude data in the first vehicle data on the vehicle side, the vehicle latitude data in the first vehicle estimated recognition data on the roadside, the vehicle side weight value, and the roadside weight value. The latitude average is obtained based on the vehicle latitude data in the first vehicle data on the vehicle side and the vehicle latitude data in the first vehicle estimated recognition data on the roadside. The first data difference corresponding to the vehicle latitude data is the difference between the latitude median and the latitude average. The formula is as follows:

[0099] δ_latα = Med_latα - aver_latα

[0100] Med_latα = (car_latα * K 车 + road_latα * K 路 ) / 1

[0101] aver_latα = (car_latα + road_latα) / 2

[0102] Among them, δ_latα represents the first data difference corresponding to the vehicle latitude data, Med_latα represents the latitude median, aver_latα represents the latitude average, car_latα represents the vehicle latitude data in the first vehicle data on the vehicle side, road_latα represents the vehicle latitude data in the first vehicle estimated recognition data on the roadside, and K 车 represents the vehicle side weight value, and K 路 represents the roadside weight value.

[0103] The vehicle length median is obtained based on the vehicle length data in the first vehicle data on the vehicle side, the vehicle length data in the first vehicle estimated recognition data on the roadside, the vehicle side weight value, and the roadside weight value. The vehicle length average is obtained based on the vehicle length data in the first vehicle data on the vehicle side and the vehicle length data in the first vehicle estimated recognition data on the roadside. The first data difference corresponding to the vehicle length is the difference between the vehicle length median and the vehicle length average. The formula is as follows:

[0104] δ_lα = Med_lα - aver_lα

[0105] Med_lα = (car_lα * K 车 + road_lα * K 路 ) / 1

[0106] aver_lα = (car_lα + road_lα) / 2

[0107] Where, δ_lα represents the first data difference corresponding to the vehicle length, Med_lα represents the median value of the vehicle length, aver_lα represents the average value of the vehicle length, car_lα represents the vehicle length data in the first vehicle side data, road_lα represents the vehicle length data in the first vehicle estimated recognition data on the roadside, K 车 represents the vehicle side weight value, K 路 represents the roadside weight value.

[0108] The vehicle width median value is obtained based on the vehicle width data in the first vehicle side data, the vehicle width data in the first vehicle estimated recognition data on the roadside, the vehicle side weight value and the roadside weight value. The vehicle width average value is obtained based on the vehicle width data in the first vehicle side data and the vehicle width data in the first vehicle estimated recognition data on the roadside. The first data difference corresponding to the vehicle width is the difference between the vehicle width median value and the vehicle width average value. The formula is as follows:

[0109] δ_wα = Med_wα - aver_wα

[0110] Med_wα = (car_wα * K 车 + road_wα * K 路 ) / 1

[0111] aver_wα = (car_wα + road_wα) / 2

[0112] Where, δ_wα represents the first data difference corresponding to the vehicle width, Med_wα represents the vehicle width median value, aver_wα represents the vehicle width average value, car_wα represents the vehicle width data in the first vehicle side data, road_wα represents the vehicle width data in the first vehicle estimated recognition data on the roadside, K 车 represents the vehicle side weight value, K 路 represents the roadside weight value.

[0113] The vehicle height median value is obtained based on the vehicle height data in the first vehicle side data, the vehicle height data in the first vehicle estimated recognition data on the roadside, the vehicle side weight value and the roadside weight value. The vehicle height average value is obtained based on the vehicle height data in the first vehicle side data and the vehicle height data in the first vehicle estimated recognition data on the roadside. The first data difference corresponding to the vehicle height is the difference between the vehicle height median value and the vehicle height average value. The formula is as follows:

[0114] δ_hα = Med_hα - aver_hα

[0115] Med_hα = (car_hα * K 车 + road_hα * K路 ) / 1

[0116] aver_hα = (car_hα + road_hα) / 2

[0117] Wherein, δ_hα represents the first data difference corresponding to the vehicle height, Med_hα represents the median value of the vehicle height, aver_hα represents the average value of the vehicle height, car_hα represents the vehicle height data in the first vehicle data on the vehicle side, road_hα represents the vehicle height data in the first vehicle estimated recognition data on the roadside, K 车 represents the vehicle side weight value, K 路 represents the roadside weight value.

[0118] Obtain the recognition data of the first vehicle according to the first vehicle data on the vehicle side and the first data difference, or, obtain the recognition data of the first vehicle according to the first vehicle estimated recognition data on the roadside and the first data difference;

[0119] Specifically, if the first data difference (including the first data difference corresponding to the vehicle longitude data as described above, the first data difference corresponding to the vehicle latitude data as described above, the first data difference corresponding to the vehicle length as described above, the first data difference corresponding to the vehicle width as described above, and the first data difference corresponding to the vehicle height as described above) is greater than or equal to the corresponding threshold value, then delete the corresponding first vehicle data on the vehicle side. If it is less than, then retain the corresponding first vehicle data on the vehicle side and delete it to obtain the recognition data of the first vehicle. Or, if the first data difference (including the first data difference corresponding to the vehicle longitude data as described above, the first data difference corresponding to the vehicle latitude data as described above, the first data difference corresponding to the vehicle length as described above, the first data difference corresponding to the vehicle width as described above, and the first data difference corresponding to the vehicle height as described above) is greater than or equal to the corresponding threshold value, then delete the corresponding first vehicle estimated recognition data on the roadside. If it is less than, then retain the corresponding first vehicle estimated recognition data on the roadside to obtain the recognition data of the first vehicle.

[0120] Obtain a second data difference according to the third target data in the second vehicle estimated recognition data on the vehicle side, the corresponding third target data in the second vehicle estimated recognition data on the roadside, the vehicle side weight value, and the roadside weight value;

[0121] Specifically, obtain a longitude median value according to the vehicle longitude data in the second vehicle estimated recognition data on the vehicle side, the vehicle longitude data in the second vehicle estimated recognition data on the roadside, the vehicle side weight value, and the roadside weight value. Obtain a longitude average value according to the vehicle longitude data in the second vehicle estimated recognition data on the vehicle side and the vehicle longitude data in the second vehicle estimated recognition data on the roadside. The second data difference corresponding to the vehicle longitude data is the difference between the longitude median value and the longitude average value;

[0122] Obtain a latitude intermediate value based on the vehicle latitude data in the second vehicle estimation and recognition data on the vehicle side, the vehicle latitude data in the second vehicle estimation and recognition data on the roadside, the vehicle side weight value, and the roadside weight value. Obtain a latitude average value based on the vehicle latitude data in the second vehicle estimation and recognition data on the vehicle side and the vehicle latitude data in the second vehicle estimation and recognition data on the roadside. The second data difference corresponding to the vehicle latitude data is the difference between the latitude intermediate value and the latitude average value;

[0123] Obtain a vehicle length intermediate value based on the vehicle length data in the second vehicle estimation and recognition data on the vehicle side, the vehicle length data in the second vehicle estimation and recognition data on the roadside, the vehicle side weight value, and the roadside weight value. Obtain a vehicle length average value based on the vehicle length data in the second vehicle estimation and recognition data on the vehicle side and the vehicle length data in the second vehicle estimation and recognition data on the roadside. The second data difference corresponding to the vehicle length is the difference between the vehicle length intermediate value and the vehicle length average value;

[0124] Obtain a vehicle width intermediate value based on the vehicle width data in the second vehicle estimation and recognition data on the vehicle side, the vehicle width data in the second vehicle estimation and recognition data on the roadside, the vehicle side weight value, and the roadside weight value. Obtain a vehicle width average value based on the vehicle width data in the second vehicle estimation and recognition data on the vehicle side and the vehicle width data in the second vehicle estimation and recognition data on the roadside. The second data difference corresponding to the vehicle width is the difference between the vehicle width intermediate value and the vehicle width average value;

[0125] Obtain a vehicle height intermediate value based on the vehicle height data in the second vehicle estimation and recognition data on the vehicle side, the vehicle height data in the second vehicle estimation and recognition data on the roadside, the vehicle side weight value, and the roadside weight value. Obtain a vehicle height average value based on the vehicle height data in the second vehicle estimation and recognition data on the vehicle side and the vehicle height data in the second vehicle estimation and recognition data on the roadside. The second data difference corresponding to the vehicle height is the difference between the vehicle height intermediate value and the vehicle height average value.

[0126] Obtain the recognition data of the second vehicle based on the second vehicle estimation and recognition data on the vehicle side and the second data difference, or obtain the recognition data of the second vehicle based on the second vehicle estimation and recognition data on the roadside and the second data difference;

[0127] Specifically, if the second data difference (including the second data difference corresponding to the vehicle longitude data described above, the second data difference corresponding to the vehicle latitude data described above, the second data difference corresponding to the vehicle length described above, the second data difference corresponding to the vehicle width described above, and the first data difference corresponding to the vehicle height described above) is greater than or equal to the corresponding threshold, the second vehicle estimated recognition data of the corresponding vehicle side is deleted; if it is less than the threshold, the second vehicle estimated recognition data of the corresponding vehicle side is retained to obtain the recognition data of the second vehicle. Or, if the second data difference (including the second data difference corresponding to the vehicle longitude data described above, the second data difference corresponding to the vehicle latitude data described above, the second data difference corresponding to the vehicle length described above, the second data difference corresponding to the vehicle width described above, and the first data difference corresponding to the vehicle height described above) is greater than or equal to the corresponding threshold, the second vehicle estimated recognition data of the corresponding roadside is deleted; if it is less than the threshold, the second vehicle estimated recognition data of the corresponding roadside is retained to obtain the recognition data of the second vehicle.

[0128] By fusing the vehicle side data and the roadside data through the above process, the accuracy of the recognition data of the first vehicle and the second vehicle obtained by processing is further improved.

[0129] Further, after obtaining the recognition data of the first vehicle and the second vehicle, the parameters in the recognition data of the first vehicle and the second vehicle are assigned to the output parameters, that is, the output parameters of the first vehicle and the output parameters of the second vehicle. Among them, the output parameters include the data number as output_idα, the vehicle longitude data as output_lonα, the vehicle latitude data as output_latα, the vehicle type as output_xα, the vehicle model as output_yα, the body color as output_colorα, the vehicle length data as output_lα, the vehicle width data as output_wα, the vehicle height data as output_hα, the license plate number as output_numbα, the vehicle speed as output_speedα, the vehicle heading angle as output_headingα, the main vehicle flag bit as output_hvflag, and the PC5 broadcast flag bit as output_pc5flag.

[0130] The assignment formula is as follows:

[0131] output_idα = road_idα; road_idα represents the data number in the recognition data of the first vehicle or the data number in the recognition data of the second vehicle;

[0132] output_lonα = Med_lonα; Med_lonα represents the vehicle longitude data in the recognition data of the first vehicle or the vehicle longitude data in the recognition data of the second vehicle;

[0133] output_latα = Med_latα; Med_latα represents the vehicle latitude data in the identification data of the first vehicle or the vehicle latitude data in the identification data of the second vehicle;

[0134] output_xα = road_xα; road_xα represents the vehicle type in the identification data of the first vehicle or the vehicle type in the identification data of the second vehicle;

[0135] output_yα = road_yα; road_yα represents the vehicle model in the identification data of the first vehicle or the vehicle model in the identification data of the second vehicle;

[0136] output_colorα = roadcolorα; roadcolorα represents the body color in the identification data of the first vehicle or the body color in the identification data of the second vehicle;

[0137] output_lα = Med_lα; Med_lα represents the vehicle length data in the identification data of the first vehicle or the vehicle length data in the identification data of the second vehicle;

[0138] output_wα = Med_wα; Med_wα represents the vehicle width data in the identification data of the first vehicle or the vehicle width data in the identification data of the second vehicle;

[0139] output_hα = Med_hα; Med_hα represents the vehicle height data in the identification data of the first vehicle or the vehicle height data in the identification data of the second vehicle;

[0140] output_numbα = road_numbα; road_numbα represents the license plate number in the identification data of the first vehicle or the license plate number in the identification data of the second vehicle;

[0141] output_speedα = road_speedα; road_speedα represents the vehicle speed in the identification data of the first vehicle or the vehicle speed in the identification data of the second vehicle;

[0142] output_headingα = road_headingα; road_headingα represents the vehicle heading angle in the identification data of the first vehicle or the vehicle heading angle in the identification data of the second vehicle.

[0143] After that, based on the identification data of the first vehicle and the identification data of the second vehicle above, it is possible to determine which vehicles are the first vehicles, that is, the connected vehicles carrying the OBU. On this basis, the main vehicle flag bit, namely "veh_hvflag", in the identification data of the first vehicle is 1. Also, according to the above steps, for the first vehicle and the second vehicle, in the output parameters of the first vehicle, the special broadcast flag bit, namely "Pc5_Flag", is set to 0, and setting this "Pc5_Flag" to 0 is the first identification information. In the output parameters of the second vehicle, the special broadcast flag bit, namely "Pc5_Flag", is set to 1, and setting this "Pc5_Flag" to 1 is the second identification information.

[0144] Add the first identification information to the identification data of the first vehicle to obtain the output parameters of the first vehicle, and add the second identification information to the identification data of the second vehicle to obtain the output parameters of the second vehicle;

[0145] After the "Pc5_Flag" flag bit is generated, send the output parameters of the second vehicle to the OBU of the first vehicle, or send the output parameters of the first vehicle and the output parameters of the second vehicle to the OBU of the first vehicle; wherein, in the case of sending the output parameters of the first vehicle and the output parameters of the second vehicle to the OBU of the first vehicle, the OBU of the first vehicle is used to display the vehicle output parameters according to the first identification information and the second identification information; the vehicle output parameters include the output parameters of the first vehicle and / or the output parameters of the second vehicle.

[0146] Specifically, the "Pc5_Flag" flag bit is applied as follows:

[0147] If the structure of the output parameters includes: data transmission timestamp, data number, vehicle longitude data, vehicle latitude data, vehicle type, vehicle model, body color, vehicle length data, vehicle width data, vehicle height data, license plate number, vehicle speed, vehicle heading angle, HvFlag, Pc5_Flag, other information (temporary ID, acceleration, altitude of the target, etc.), since the output parameters already carry "Pc5_Flag" to determine whether it is the connected main vehicle (the first vehicle), therefore, from an application perspective, there are two ways to achieve deduplication of the connected main vehicle itself. The first is that the target RSU with "Pc5_Flag" being 1 will not perform PC5 and uu interface broadcasts, and the target RSU with "Pc5_Flag" being 0 will normally perform PC5 and uu interface broadcasts. There are no requirements for the target data (output parameters of the second vehicle) sent to the vehicle-road collaborative cloud visualization display. The second is that the RSU does not perform any processing, broadcasts the target with "Pc5_Flag" being 1 for PC5 and uu interfaces to all connected vehicles, and the connected main vehicle with OBU processes and filters this target data (output parameters of the first vehicle). In this way, it can be achieved that the connected main vehicle with OBU does not display itself at the HMI presentation end, and at the same time, it does not affect the presentation requirements of the vehicle-road collaborative cloud control platform for all detected targets.

[0148] The following combines Figure 4 , and specifically describes the process of the vehicle identification data processing method provided by the embodiments of the present invention:

[0149] This specific process includes three processes: vehicle-side and roadside data input, flag bit generation process, and flag bit generation and application;

[0150] In the vehicle-side and roadside data input process, the input of vehicle-side vehicle data and the input of roadside perception data are received, and data integrity judgment is performed. If the data is complete, the data enters the flag bit generation process; if the data is incomplete, the data is discarded.

[0151] In the flag bit generation process, it includes feature attribute matching, confirming the main vehicle sub-process and the model calculation sub-process. In the feature attribute matching and main vehicle confirmation sub-process, parameters such as target longitude and latitude, relative position, speed, and heading angle are traversed and matched for each target. If the match is successful, suspected main vehicle determination and other vehicle determination are performed. If the match is unsuccessful, it directly enters the process of determining the main vehicle and other vehicles. After the suspected main vehicle determination and other vehicle determination, secondary feature attribute matching of parameters such as target size and license plate number is performed. If the match is successful, it enters the process of determining the main vehicle and other vehicles. If the match is unsuccessful, the data is discarded. In the weighted calculation unit of the model calculation sub-process, relative position confirmation with the suspected main vehicle, weighted average, and weight model matching calculation are performed. Then, difference calculation and threshold comparison are carried out. If it exceeds the threshold, the data is discarded. If it is less than the threshold, new fused data is generated;

[0152] In the flag bit generation and application process, a PC5 broadcast flag bit is generated in the output parameters, and the target information (output parameters) is output. The OBU receives the target flag bit information and performs deduplication and display.

[0153] The vehicle recognition data processing method provided by the embodiments of the present invention can solve the problem of the repeated display of the vehicle with an OBU on the HMI presentation end in existing vehicle-road collaborative applications, and at the same time does not affect the presentation requirements of all detected target objects by the vehicle-road collaborative cloud control platform. This method can be applied to high-speed and urban road scenarios with complex target types by generating a PC5 broadcast flag bit in combination with the recognized and matched target objects. Moreover, the generation of the PC5 broadcast flag bit is not limited to the types and quantities of vehicle-side data and roadside perception data sensors, and can be applied to various sensor configuration schemes. Compared with a pure vehicle-side or pure roadside fusion perception system, the data comes from both the vehicle side and the roadside, improving the diversity of the data structure; feature matching is performed through timestamps, longitude and latitude, relative position of the target object, speed, and heading angle, and at the same time, secondary feature attribute matching is performed through information such as vehicle type, body color, body size, and license plate number to further ensure the accuracy of target recognition; the output logic of the PC5 broadcast flag bit is placed at the roadside end, which is more cost-effective and simpler than placing it at the vehicle end.

[0154] The vehicle identification data processing method provided by the embodiment of the present invention solves the problem of the phenomenon that the connected host vehicle carrying the OBU repeatedly displays its own vehicle on the HMI presentation end in the existing vehicle-road collaboration application, and at the same time does not affect the presentation requirements of all detected targets by the vehicle-road collaboration cloud control platform; the generation of the PC5 broadcast flag bit is not limited to the types and quantities of vehicle-side data and roadside perception data sensors, and can be applied to various sensor configuration schemes. Compared with the pure vehicle-side or pure roadside fusion perception system, the data comes from the vehicle side and the roadside, and the multi-target data input method is adopted to improve the diversity of the data structure and the accuracy of the data; under the condition of unified timestamp, based on the longitude and latitude information, after confirming the host vehicle, the target object information is confirmed through the group traversal matching method based on the relative position of the target object, and the secondary feature attributes are repeatedly matched to further ensure the accuracy of target recognition; the application of vehicle-side data and roadside perception data adopts the method of the relative position and weight relationship model between the target object and the host vehicle to ensure the accuracy of the output data of the target object.

[0155] As Figure 5 shown, the embodiment of the present invention further provides a vehicle identification data processing device, including:

[0156] A data acquisition module 501, configured to acquire the first vehicle-side vehicle data of the first vehicle sent by the first vehicle and the second vehicle-side vehicle data of the second vehicle, and acquire the roadside perception data of each roadside vehicle sent by the roadside unit RSU;

[0157] A first processing module 502, configured to obtain the estimated identification data of the second vehicle on the vehicle side, the estimated identification data of the first vehicle on the roadside, and the estimated identification data of the second vehicle on the roadside according to the first vehicle-side vehicle data, the second vehicle-side vehicle data, and the roadside perception data;

[0158] A second processing module 503, configured to obtain the identification data of the first vehicle and the identification data of the second vehicle according to the first vehicle-side vehicle data, the estimated identification data of the second vehicle on the vehicle side, the estimated identification data of the first vehicle on the roadside, the estimated identification data of the second vehicle on the roadside, the vehicle-side weight value, and the roadside weight value;

[0159] Wherein, the first vehicle is a vehicle carrying an on-vehicle unit OBU, and the second vehicle is a vehicle within a first preset range of the first vehicle.

[0160] Optionally, the vehicle-side vehicle data, the roadside perception data, the estimated identification data, or the identification data includes the following sub-data:

[0161] Data sending timestamp, data number, vehicle longitude data, vehicle latitude data, vehicle type, vehicle model, body color, vehicle size data, license plate number, vehicle speed, vehicle heading angle;

[0162] Among them, the vehicle-side vehicle data further includes an OBU carrying flag bit.

[0163] Optionally, the device further includes:

[0164] A first deletion module, configured to delete the first target vehicle-side vehicle data and the first target roadside perception data when neither the first target vehicle-side vehicle data nor the first target roadside perception data includes the first target sub-data;

[0165] Among them, the first target vehicle-side vehicle data is the second vehicle-side vehicle data of any second vehicle, and the first target roadside perception data is the roadside perception data of any roadside vehicle;

[0166] The first target sub-data is one or more of the sub-data.

[0167] Optionally, the first processing module 502 includes:

[0168] A first determination unit, configured to use the first vehicle-side vehicle data and each roadside perception data to determine a third vehicle in the roadside vehicles that is most similar to the first vehicle;

[0169] A second determination unit, configured to determine a first distance between the first vehicle and each second vehicle according to the vehicle longitude data and vehicle latitude data in the first vehicle-side vehicle data and the vehicle longitude data and vehicle latitude data in the second vehicle-side vehicle data;

[0170] A third determination unit, configured to determine a second distance between the third vehicle and each fourth vehicle according to the vehicle longitude data and vehicle latitude data in the roadside perception data of the third vehicle and the vehicle longitude data and vehicle latitude data in the roadside perception data of the fourth vehicle; the fourth vehicle is a vehicle in the roadside vehicles that is within a second preset range of the third vehicle;

[0171] A first processing unit, configured to obtain initial estimated identification data of the second vehicle on the vehicle side, initial estimated identification data of the first vehicle on the roadside, and initial estimated identification data of the second vehicle on the roadside according to the first distance, the second distance, the second vehicle-side vehicle data, the roadside perception data of the third vehicle, and the roadside perception data of the fourth vehicle;

[0172] A second processing unit, configured to obtain estimated identification data of the second vehicle on the vehicle side, estimated identification data of the first vehicle on the roadside, and estimated identification data of the second vehicle on the roadside according to the initial estimated identification data of the second vehicle on the vehicle side, the initial estimated identification data of the first vehicle on the roadside, the initial estimated identification data of the second vehicle on the roadside, and the first vehicle-side vehicle data.

[0173] Optionally, the first processing unit is specifically configured to:

[0174] Use the roadside perception data of the third vehicle as the first vehicle estimation and identification data of the roadside;

[0175] In the case where the vehicle data on the second side of the first target vehicle matches the roadside perception data of the second target vehicle, use the vehicle data on the second side of the first target vehicle as the second vehicle estimation and identification data of the vehicle side, and use the roadside perception data of the second target vehicle as the second vehicle estimation and identification data of the roadside;

[0176] Wherein, the first target vehicle is one vehicle among the second vehicles, the second target vehicle is one vehicle among the fourth vehicles, and the absolute value of the difference between the first distance corresponding to the first target vehicle and the second distance corresponding to the second target vehicle is less than or equal to a first preset distance.

[0177] Optionally, the first processing unit is further specifically configured to:

[0178] In the case where a first condition is satisfied, determine that the vehicle data on the second side of the first target vehicle matches the roadside perception data of the second target vehicle;

[0179] Wherein, the first condition includes one or more of the following:

[0180] The absolute value of the difference between the vehicle longitude data in the vehicle data on the second side of the first target vehicle and the vehicle longitude data in the roadside perception data of the second target vehicle is less than a second preset distance;

[0181] The absolute value of the difference between the vehicle latitude data in the vehicle data on the second side of the first target vehicle and the vehicle latitude data in the roadside perception data of the second target vehicle is less than a second preset distance;

[0182] The absolute value of the difference between the vehicle heading angle in the vehicle data on the second side of the first target vehicle and the vehicle heading angle in the roadside perception data of the second target vehicle is less than a first preset angle;

[0183] The absolute value of the difference between the vehicle speed in the vehicle data on the second side of the first target vehicle and the vehicle speed in the roadside perception data of the second target vehicle is less than a first preset speed;

[0184] The absolute value of the difference between the vehicle size data in the vehicle data on the second side of the first target vehicle and the vehicle size data in the roadside perception data of the second target vehicle is less than a first preset size.

[0185] Optionally, the second processing unit is specifically configured to:

[0186] When the second target sub - data in the second vehicle initial estimated recognition data on the vehicle side matches successfully with the second target sub - data in the second vehicle initial estimated recognition data on the roadside, use the second vehicle initial estimated recognition data on the vehicle side as the second vehicle estimated recognition data on the vehicle side, and use the second vehicle initial estimated recognition data on the roadside as the second vehicle estimated recognition data on the roadside;

[0187] When the second target sub - data in the first vehicle initial estimated recognition data on the roadside matches successfully with the second target sub - data in the first vehicle - side vehicle data, use the first vehicle initial estimated recognition data on the roadside as the first vehicle estimated recognition data on the roadside;

[0188] Wherein, the second target sub - data includes at least one of the following:

[0189] Vehicle type; vehicle model; body color; license plate number.

[0190] Optionally, the device further includes:

[0191] A first determination module, configured to determine the vehicle - side weight value and the roadside weight value according to the first distance between the first vehicle and each second vehicle.

[0192] Optionally, the second processing module 503 includes:

[0193] A third processing unit, configured to obtain a first data difference according to the third target data in the first vehicle - side vehicle data, the corresponding third target data in the first vehicle estimated recognition data on the roadside, the vehicle - side weight value, and the roadside weight value;

[0194] A fourth processing unit, configured to obtain the recognition data of the first vehicle according to the first vehicle - side vehicle data and the first data difference, or obtain the recognition data of the first vehicle according to the first vehicle estimated recognition data on the roadside and the first data difference;

[0195] A fifth processing unit, configured to obtain a second data difference according to the third target data in the second vehicle estimated recognition data on the vehicle side, the corresponding third target data in the second vehicle estimated recognition data on the roadside, the vehicle - side weight value, and the roadside weight value;

[0196] A sixth processing unit, configured to obtain the recognition data of the second vehicle according to the second vehicle estimated recognition data on the vehicle side and the second data difference, or obtain the recognition data of the second vehicle according to the second vehicle estimated recognition data on the roadside and the second data difference;

[0197] Among them, the third target data is one of the following:

[0198] Vehicle longitude data; vehicle latitude data; vehicle size data.

[0199] Optionally, the method further includes:

[0200] An adding module, configured to add first identification information to the identification data of the first vehicle to obtain the output parameters of the first vehicle, and add second identification information to the identification data of the second vehicle to obtain the output parameters of the second vehicle;

[0201] A sending module, configured to send the output parameters of the second vehicle to the OBU of the first vehicle, or send the output parameters of the first vehicle and the output parameters of the second vehicle to the OBU of the first vehicle;

[0202] Among them, when sending the output parameters of the first vehicle and the output parameters of the second vehicle to the OBU of the first vehicle, the OBU of the first vehicle is configured to display the vehicle output parameters according to the first identification information and the second identification information; the vehicle output parameters include the output parameters of the first vehicle and / or the output parameters of the second vehicle.

[0203] It should be noted that the vehicle identification data processing device provided in the embodiment of the present invention is a device capable of executing the above vehicle identification data processing method. All embodiments of the above vehicle identification data processing method are applicable to this device and can achieve the same or similar technical effects.

[0204] As Figure 6 shown, the embodiment of the present invention further provides an edge computing node, including: a processor 601, a memory 602, and a program stored on the memory 602 and executable on the processor 601. When the program is executed by the processor 601, the above signal lamp control method is implemented.

[0205] Optionally, it further includes: a transceiver 603, and the transceiver 603 is configured to receive and send data under the control of the processor 601.

[0206] Specifically, the processor 601 performs the following processes:

[0207] Obtain the first vehicle side vehicle data of the first vehicle and the second vehicle side vehicle data of the second vehicle sent by the first vehicle, and obtain the roadside perception data of each roadside vehicle sent by the roadside unit RSU;

[0208] Based on the first vehicle-side vehicle data, the second vehicle-side vehicle data, and the roadside perception data, obtain the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, and the second vehicle estimation and recognition data on the roadside;

[0209] Based on the first vehicle-side vehicle data, the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, the second vehicle estimation and recognition data on the roadside, the vehicle-side weight value, and the roadside weight value, obtain the recognition data of the first vehicle and the recognition data of the second vehicle;

[0210] Wherein, the first vehicle is a vehicle carrying an on-board unit OBU, and the second vehicle is a vehicle within a first preset range of the first vehicle.

[0211] Optionally, the vehicle-side vehicle data, the roadside perception data, the estimation and recognition data, or the recognition data includes the following sub-data:

[0212] Data transmission timestamp, data number, vehicle longitude data, vehicle latitude data, vehicle type, vehicle model, body color, vehicle size data, license plate number, vehicle speed, vehicle heading angle;

[0213] Wherein, the vehicle-side vehicle data further includes an OBU carrying flag bit.

[0214] Optionally, the processor 601 is further configured to:

[0215] In the case where neither the first target vehicle-side vehicle data nor the first target roadside perception data includes the first target sub-data, delete the first target vehicle-side vehicle data and the first target roadside perception data;

[0216] Wherein, the first target vehicle-side vehicle data is the second vehicle-side vehicle data of any second vehicle, and the first target roadside perception data is the roadside perception data of any roadside vehicle;

[0217] The first target sub-data is one or more of the sub-data.

[0218] Optionally, the processor 601 is specifically configured to:

[0219] Use the first vehicle-side vehicle data and each roadside perception data to determine the third vehicle among the roadside vehicles that is most similar to the first vehicle;

[0220] According to the vehicle longitude data and vehicle latitude data in the first vehicle-side vehicle data and the vehicle longitude data and vehicle latitude data in the second vehicle-side vehicle data, determine the first distance between the first vehicle and each second vehicle;

[0221] Determine a second distance between the third vehicle and each of the fourth vehicles based on the vehicle longitude data and vehicle latitude data in the roadside perception data of the third vehicle and the vehicle longitude data and vehicle latitude data in the roadside perception data of the fourth vehicle; the fourth vehicle is a vehicle among the roadside vehicles that is within a second preset range of the third vehicle;

[0222] Based on the first distance, the second distance, the second vehicle data on the vehicle side, the roadside perception data of the third vehicle, and the roadside perception data of the fourth vehicle, obtain the initial estimated identification data of the second vehicle on the vehicle side, the initial estimated identification data of the first vehicle on the roadside, and the initial estimated identification data of the second vehicle on the roadside;

[0223] Based on the initial estimated identification data of the second vehicle on the vehicle side, the initial estimated identification data of the first vehicle on the roadside, the initial estimated identification data of the second vehicle on the roadside, and the first vehicle data on the vehicle side, obtain the estimated identification data of the second vehicle on the vehicle side, the estimated identification data of the first vehicle on the roadside, and the estimated identification data of the second vehicle on the roadside.

[0224] Optionally, the processor 601 is specifically configured to:

[0225] Use the roadside perception data of the third vehicle as the initial estimated identification data of the first vehicle on the roadside;

[0226] When the second vehicle data on the vehicle side of the first target vehicle matches successfully with the roadside perception data of the second target vehicle, use the second vehicle data on the vehicle side of the first target vehicle as the initial estimated identification data of the second vehicle on the vehicle side, and use the roadside perception data of the second target vehicle as the initial estimated identification data of the second vehicle on the roadside;

[0227] Wherein, the first target vehicle is one of the second vehicles, the second target vehicle is one of the fourth vehicles, and the absolute value of the difference between the first distance corresponding to the first target vehicle and the second distance corresponding to the second target vehicle is less than or equal to a first preset distance.

[0228] Optionally, the processor 601 is further specifically configured to:

[0229] When a first condition is satisfied, determine that the second vehicle data on the vehicle side of the first target vehicle matches successfully with the roadside perception data of the second target vehicle;

[0230] Wherein, the first condition includes one or more of the following:

[0231] The absolute value of the difference between the vehicle longitude data in the second vehicle - side vehicle data of the first target vehicle and the vehicle longitude data in the roadside perception data of the second target vehicle is less than a second preset distance;

[0232] The absolute value of the difference between the vehicle latitude data in the second vehicle - side vehicle data of the first target vehicle and the vehicle latitude data in the roadside perception data of the second target vehicle is less than a second preset distance;

[0233] The absolute value of the difference between the vehicle heading angle in the second vehicle - side vehicle data of the first target vehicle and the vehicle heading angle in the roadside perception data of the second target vehicle is less than a first preset angle;

[0234] The absolute value of the difference between the vehicle speed in the second vehicle - side vehicle data of the first target vehicle and the vehicle speed in the roadside perception data of the second target vehicle is less than a first preset speed;

[0235] The absolute value of the difference between the vehicle size data in the second vehicle - side vehicle data of the first target vehicle and the vehicle size data in the roadside perception data of the second target vehicle is less than a first preset size.

[0236] Optionally, the processor 601 is specifically configured to:

[0237] When the second target sub - data in the second vehicle initial estimated identification data on the vehicle side matches the second target sub - data in the second vehicle initial estimated identification data on the roadside, taking the second vehicle initial estimated identification data on the vehicle side as the second vehicle estimated identification data on the vehicle side, and taking the second vehicle initial estimated identification data on the roadside as the second vehicle estimated identification data on the roadside;

[0238] When the second target sub - data in the first vehicle initial estimated identification data on the roadside matches the second target sub - data in the first vehicle - side vehicle data, taking the first vehicle initial estimated identification data on the roadside as the first vehicle estimated identification data on the roadside;

[0239] Wherein, the second target sub - data includes at least one of the following:

[0240] Vehicle type; vehicle model; body color; license plate number.

[0241] Optionally, the processor 601 is further configured to:

[0242] Determine the vehicle - side weight value and the roadside weight value according to the first distance between the first vehicle and each second vehicle.

[0243] Optionally, the processor 601 is specifically configured to:

[0244] Obtain a first data difference based on the third target data in the first vehicle-side vehicle data, the corresponding third target data in the first vehicle estimated recognition data on the roadside, the vehicle-side weight value, and the roadside weight value;

[0245] Obtain the recognition data of the first vehicle based on the first vehicle-side vehicle data and the first data difference, or obtain the recognition data of the first vehicle based on the first vehicle estimated recognition data on the roadside and the first data difference;

[0246] Obtain a second data difference based on the third target data in the second vehicle estimated recognition data on the vehicle side, the corresponding third target data in the second vehicle estimated recognition data on the roadside, the vehicle-side weight value, and the roadside weight value;

[0247] Obtain the recognition data of the second vehicle based on the second vehicle estimated recognition data on the vehicle side and the second data difference, or obtain the recognition data of the second vehicle based on the second vehicle estimated recognition data on the roadside and the second data difference;

[0248] Wherein, the third target data is one of the following:

[0249] Vehicle longitude data; Vehicle latitude data; Vehicle size data.

[0250] Optionally, the processor 601 is further configured to:

[0251] Add first identification information to the recognition data of the first vehicle to obtain the output parameter of the first vehicle, and add second identification information to the recognition data of the second vehicle to obtain the output parameter of the second vehicle;

[0252] The transceiver 603 is further configured to:

[0253] Send the output parameter of the second vehicle to the OBU of the first vehicle, or send the output parameter of the first vehicle and the output parameter of the second vehicle to the OBU of the first vehicle;

[0254] Wherein, when sending the output parameter of the first vehicle and the output parameter of the second vehicle to the OBU of the first vehicle, the OBU of the first vehicle is configured to display the vehicle output parameter according to the first identification information and the second identification information; The vehicle output parameter includes the output parameter of the first vehicle and / or the output parameter of the second vehicle.

[0255] Wherein, in Figure 6Among them, the bus architecture may include any number of interconnected buses and bridges, specifically, various circuits represented by one or more processors represented by processor 601 and a memory represented by memory 602 are linked together. The bus architecture may also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and thus will not be further described herein. The bus interface provides a user interface 604. The transceiver 603 may be a plurality of components, that is, including a transmitter and a receiver, and provides a unit for communicating with various other devices on a transmission medium. The processor 601 is responsible for managing the bus architecture and general processing, and the memory 602 may store data used by the processor 601 when performing operations.

[0256] In addition, a specific embodiment of the present invention further provides a readable storage medium, on which a program is stored, and when the program is executed by a processor, the steps in any one of the above vehicle identification data processing methods are implemented.

[0257] The above are the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, several improvements and refinements can be made without departing from the principles described in the present invention, and these improvements and refinements are also within the protection scope of the present invention.

Claims

1. A method for processing vehicle recognition data, characterized in that, it includes: obtaining the first vehicle-side vehicle data of the first vehicle and the second vehicle-side vehicle data of the second vehicle sent by the first vehicle, and obtaining the roadside perception data of each roadside vehicle sent by the roadside unit RSU; obtaining the second vehicle estimated recognition data on the vehicle side, the first vehicle estimated recognition data on the roadside, and the second vehicle estimated recognition data on the roadside according to the first vehicle-side vehicle data, the second vehicle-side vehicle data, and the roadside perception data; obtaining the recognition data of the first vehicle and the recognition data of the second vehicle according to the first vehicle-side vehicle data, the second vehicle estimated recognition data on the vehicle side, the first vehicle estimated recognition data on the roadside, the second vehicle estimated recognition data on the roadside, the vehicle-side weight value, and the roadside weight value; wherein, the first vehicle is a vehicle carrying an on-vehicle unit OBU, and the second vehicle is a vehicle located within a first preset range of the first vehicle; wherein, the step of obtaining the second vehicle estimated recognition data on the vehicle side, the first vehicle estimated recognition data on the roadside, and the second vehicle estimated recognition data on the roadside according to the first vehicle-side vehicle data, the second vehicle-side vehicle data, and the roadside perception data includes: using the first vehicle-side vehicle data and each roadside perception data to determine the third vehicle among the roadside vehicles that is most similar to the first vehicle; determining the first distance between the first vehicle and each second vehicle according to the vehicle longitude data and vehicle latitude data in the first vehicle-side vehicle data and the vehicle longitude data and vehicle latitude data in the second vehicle-side vehicle data; determining the second distance between the third vehicle and each fourth vehicle according to the vehicle longitude data and vehicle latitude data in the roadside perception data of the third vehicle and the vehicle longitude data and vehicle latitude data in the roadside perception data of the fourth vehicle; the fourth vehicle is a vehicle among the roadside vehicles that is located within a second preset range of the third vehicle; obtaining the second vehicle initial estimated recognition data on the vehicle side, the first vehicle initial estimated recognition data on the roadside, and the second vehicle initial estimated recognition data on the roadside according to the first distance, the second distance, the second vehicle-side vehicle data, the roadside perception data of the third vehicle, and the roadside perception data of the fourth vehicle; obtaining the second vehicle estimated recognition data on the vehicle side, the first vehicle estimated recognition data on the roadside, and the second vehicle estimated recognition data on the roadside according to the second vehicle initial estimated recognition data on the vehicle side, the first vehicle initial estimated recognition data on the roadside, the second vehicle initial estimated recognition data on the roadside, and the first vehicle-side vehicle data; wherein, the vehicle-side vehicle data, the roadside perception data, the estimated recognition data, or the recognition data includes the following sub-data: data sending timestamp, data number, vehicle longitude data, vehicle latitude data, vehicle type, vehicle model, body color, vehicle size data, license plate number, vehicle speed, vehicle heading angle; wherein, the vehicle-side vehicle data further includes an OBU carrying flag bit.

2. The vehicle identification data processing method according to claim 1, characterized in that, after obtaining the first vehicle side vehicle data of the first vehicle and the second vehicle side vehicle data of the second vehicle sent by the first vehicle, and obtaining the roadside perception data of each roadside vehicle sent by the roadside unit RSU, the method further includes: when the first target vehicle side vehicle data and the first target roadside perception data do not include the first target sub-data, deleting the first target vehicle side vehicle data and the first target roadside perception data; wherein, the first target vehicle side vehicle data is the second vehicle side vehicle data of any second vehicle, and the first target roadside perception data is the roadside perception data of any roadside vehicle; the first target sub-data is one or more of the sub-data.

3. The vehicle identification data processing method according to claim 1, characterized in that, the obtaining the second vehicle estimated identification data on the vehicle side, the first vehicle estimated identification data on the roadside, and the second vehicle estimated identification data on the roadside according to the first distance, the second distance, the second vehicle side vehicle data, the roadside perception data of the third vehicle, and the roadside perception data of the fourth vehicle includes: using the roadside perception data of the third vehicle as the first vehicle estimated identification data on the roadside; when the second vehicle side vehicle data of the first target vehicle matches successfully with the roadside perception data of the second target vehicle, using the second vehicle side vehicle data of the first target vehicle as the second vehicle estimated identification data on the vehicle side, and using the roadside perception data of the second target vehicle as the second vehicle estimated identification data on the roadside; wherein, the first target vehicle is one vehicle among the second vehicles, the second target vehicle is one vehicle among the fourth vehicles, and the absolute value of the difference between the first distance corresponding to the first target vehicle and the second distance corresponding to the second target vehicle is less than or equal to a first preset distance.

4. The vehicle identification data processing method according to claim 3, characterized in that, the method further includes: when a first condition is satisfied, determining that the second vehicle side vehicle data of the first target vehicle matches successfully with the roadside perception data of the second target vehicle; wherein, the first condition includes one or more of the following: the absolute value of the difference between the vehicle longitude data in the second vehicle side vehicle data of the first target vehicle and the vehicle longitude data in the roadside perception data of the second target vehicle is less than a second preset distance; the absolute value of the difference between the vehicle latitude data in the second vehicle side vehicle data of the first target vehicle and the vehicle latitude data in the roadside perception data of the second target vehicle is less than a second preset distance; the absolute value of the difference between the vehicle heading angle in the second vehicle side vehicle data of the first target vehicle and the vehicle heading angle in the roadside perception data of the second target vehicle is less than a first preset angle; the absolute value of the difference between the vehicle speed in the second vehicle side vehicle data of the first target vehicle and the vehicle speed in the roadside perception data of the second target vehicle is less than a first preset speed; The absolute value of the difference between the vehicle size data in the second vehicle side vehicle data of the first target vehicle and the vehicle size data in the roadside perception data of the second target vehicle is less than a first preset size.

5. The vehicle identification data processing method according to claim 1, wherein, the obtaining of the second vehicle estimated identification data on the vehicle side, the first vehicle estimated identification data on the roadside, and the second vehicle estimated identification data on the roadside according to the second vehicle initial estimated identification data on the vehicle side, the first vehicle initial estimated identification data on the roadside, the second vehicle initial estimated identification data on the roadside, and the first vehicle side vehicle data includes: when the second target sub-data in the second vehicle initial estimated identification data on the vehicle side matches the second target sub-data in the second vehicle initial estimated identification data on the roadside, taking the second vehicle initial estimated identification data on the vehicle side as the second vehicle estimated identification data on the vehicle side, and taking the second vehicle initial estimated identification data on the roadside as the second vehicle estimated identification data on the roadside; when the second target sub-data in the first vehicle initial estimated identification data on the roadside matches the second target sub-data in the first vehicle side vehicle data, taking the first vehicle initial estimated identification data on the roadside as the first vehicle estimated identification data on the roadside; wherein, the second target sub-data includes at least one of the following: vehicle type; vehicle model; body color; license plate number.

6. The vehicle identification data processing method according to claim 1, wherein, before obtaining the identification data of the first vehicle and the identification data of the second vehicle according to the first vehicle side vehicle data, the second vehicle estimated identification data on the vehicle side, the first vehicle estimated identification data on the roadside, the second vehicle estimated identification data on the roadside, the vehicle side weight value, and the roadside weight value, the method further includes: determining the vehicle side weight value and the roadside weight value according to the first distance between the first vehicle and each of the second vehicles.

7. The vehicle identification data processing method according to claim 1, wherein, the obtaining of the identification data of the first vehicle and the identification data of the second vehicle according to the first vehicle side vehicle data, the second vehicle estimated identification data on the vehicle side, the first vehicle estimated identification data on the roadside, the second vehicle estimated identification data on the roadside, the vehicle side weight value, and the roadside weight value includes: obtaining a first data difference according to the third target data in the first vehicle side vehicle data, the corresponding third target data in the first vehicle estimated identification data on the roadside, the vehicle side weight value, and the roadside weight value; obtaining the identification data of the first vehicle according to the first vehicle side vehicle data and the first data difference, or obtaining the identification data of the first vehicle according to the first vehicle estimated identification data on the roadside and the first data difference; Obtain a second data difference according to the third target data in the second vehicle estimation and recognition data on the vehicle side, the corresponding third target data in the second vehicle estimation and recognition data on the roadside, the vehicle side weight value, and the roadside weight value; Obtain the recognition data of the second vehicle according to the second vehicle estimation and recognition data on the vehicle side and the second data difference, or obtain the recognition data of the second vehicle according to the second vehicle estimation and recognition data on the roadside and the second data difference; Wherein, the third target data is one of the following: Vehicle longitude data; vehicle latitude data; vehicle size data.

8. The vehicle recognition data processing method according to claim 1, characterized in that the method further includes: Adding first identification information to the recognition data of the first vehicle to obtain the output parameter of the first vehicle, and adding second identification information to the recognition data of the second vehicle to obtain the output parameter of the second vehicle; Sending the output parameter of the second vehicle to the OBU of the first vehicle, or sending the output parameter of the first vehicle and the output parameter of the second vehicle to the OBU of the first vehicle; Wherein, in the case of sending the output parameter of the first vehicle and the output parameter of the second vehicle to the OBU of the first vehicle, the OBU of the first vehicle is used to display the vehicle output parameter according to the first identification information and the second identification information; the vehicle output parameter includes the output parameter of the first vehicle and / or the output parameter of the second vehicle.

9. A vehicle recognition data processing device, characterized in that it includes: A data acquisition module, configured to acquire the first vehicle side vehicle data of the first vehicle and the second vehicle side vehicle data of the second vehicle sent by the first vehicle, and acquire the roadside perception data of each roadside vehicle sent by the roadside unit RSU; A first processing module, configured to obtain the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, and the second vehicle estimation and recognition data on the roadside according to the first vehicle side vehicle data, the second vehicle side vehicle data, and the roadside perception data; A second processing module, configured to obtain the recognition data of the first vehicle and the recognition data of the second vehicle according to the first vehicle side vehicle data, the second vehicle estimation and recognition data on the vehicle side, the first vehicle estimation and recognition data on the roadside, the second vehicle estimation and recognition data on the roadside, the vehicle side weight value, and the roadside weight value; Wherein, the first vehicle is a vehicle carrying an on-board unit OBU, and the second vehicle is a vehicle within a first preset range of the first vehicle; Wherein, the first processing module includes: A first determination unit, configured to use the first vehicle side vehicle data and each roadside perception data to determine the third vehicle among the roadside vehicles that is most similar to the first vehicle; A second determination unit, configured to determine the first distance between the first vehicle and each second vehicle according to the vehicle longitude data and vehicle latitude data in the first vehicle side vehicle data and the vehicle longitude data and vehicle latitude data in the second vehicle side vehicle data; A third determination unit, configured to determine a second distance between the third vehicle and each of the fourth vehicles according to the vehicle longitude data and vehicle latitude data in the roadside perception data of the third vehicle and the vehicle longitude data and vehicle latitude data in the roadside perception data of the fourth vehicle; the fourth vehicle is a vehicle in the roadside vehicles that is within a second preset range of the third vehicle; A first processing unit, configured to obtain initial estimated identification data of a second vehicle on the vehicle side, initial estimated identification data of a first vehicle on the roadside, and initial estimated identification data of a second vehicle on the roadside according to the first distance, the second distance, the vehicle side vehicle data, the roadside perception data of the third vehicle, and the roadside perception data of the fourth vehicle; A second processing unit, configured to obtain estimated identification data of a second vehicle on the vehicle side, estimated identification data of a first vehicle on the roadside, and estimated identification data of a second vehicle on the roadside according to the initial estimated identification data of the second vehicle on the vehicle side, the initial estimated identification data of the first vehicle on the roadside, the initial estimated identification data of the second vehicle on the roadside, and the vehicle side vehicle data of the first vehicle; Wherein, the vehicle side vehicle data, the roadside perception data, the estimated identification data, or the identification data includes the following sub-data: Data transmission timestamp, data number, vehicle longitude data, vehicle latitude data, vehicle type, vehicle model, body color, vehicle size data, license plate number, vehicle speed, vehicle heading angle; Wherein, the vehicle side vehicle data further includes an OBU carrying flag bit.

10. An edge computing node, Characterized in that, Comprising: A processor, a memory, and a program stored on the memory and executable on the processor, where when the program is executed by the processor, the steps of the vehicle identification data processing method according to any one of claims 1 to 8 are implemented.

11. A readable storage medium, Characterized in that, A program is stored on the readable storage medium, and when the program is executed by a processor, the steps in the vehicle identification data processing method according to any one of claims 1 to 8 are implemented.

Images