A wide-area radar-based holographic traffic perception method and system

Abstract

The application discloses a kind of holographic traffic perception method and system based on wide-area radar, belong to vehicle traffic control technical field, for solving the technical problems of existing traffic perception system, it is difficult to provide real-time, holographic traffic information service for mass vehicle, not conducive to optimizing traffic flow space-time distribution, and relieving congestion and reducing accident rate.Method includes: the coordinate data uniform processing of initial tracking data and initial driving data about latitude and longitude offset, respectively corresponding to obtain radar tracking data and target driving data;Radar tracking data and target driving data are calculated about the trajectory topology of target reference system, respectively to obtain first state estimation information and second state estimation information;And generate minimum topology distance optimization model;Based on minimum topology distance optimization model, relative position of multi-vehicle target is distinguished and identified, to obtain vehicle holographic information;Realize holographic traffic perception to global vehicle target.

Description

Technical Field

[0001] This application relates to the field of vehicle traffic control, and in particular to a holographic traffic perception method and system based on wide-area radar. Background Technology

[0002] With socio-economic development, the number of motor vehicles in cities continues to increase, and existing road resources are unable to meet the ever-growing demand for vehicle traffic, leading to severe traffic congestion and accidents, and significantly impacting national economic development. Implementing intelligent transportation technology is one of the effective means to solve current traffic problems. By sensing road traffic conditions in real time and formulating effective traffic control strategies, it provides travelers with real-time travel information services, thereby optimizing the spatiotemporal distribution of traffic flow, alleviating congestion, and reducing accident rates. Traffic perception is a prerequisite for the effective functioning of intelligent transportation systems; accurate, comprehensive, and reliable traffic perception is a prerequisite for ensuring the effectiveness of traffic control.

[0003] Existing traffic perception systems mostly employ cross-sectional detectors, video equipment, floating car technology, or internet-based trajectory detection technology. Due to limitations in equipment detection range or low penetration rates of vehicle-mounted terminals, they cannot achieve holographic traffic perception of all vehicle targets within a road segment. Alternatively, holographic traffic perception can also be achieved based on vehicle-mounted LiDAR and V2V communication. However, intelligent vehicle-mounted terminal equipment such as LiDAR is expensive and can only identify targets in the area surrounding the vehicle, limiting its detection range. V2V communication has low penetration rates and cannot achieve full coverage of vehicles on the road.

[0004] Currently, wide-area millimeter-wave radar can continuously track the trajectories of multiple vehicles within a 200-300 meter range on a road. However, the vehicle position and motion information collected by roadside wide-area millimeter-wave radar is based on a coordinate system established at the radar's location, i.e., the radar's own relative coordinate system. This cannot directly provide vehicle target position services for vehicles on the road, and it is also difficult to provide real-time, holographic traffic information services for the general public. Summary of the Invention

[0005] This application provides a holographic traffic perception method and system based on wide-area radar to solve the following technical problems: Existing traffic perception systems cannot achieve holographic traffic perception of all vehicle targets within a road segment, making it difficult to provide real-time, holographic traffic information services for the public, which is not conducive to optimizing the spatiotemporal distribution of traffic flow, alleviating congestion, and reducing the accident rate.

[0006] The embodiments of this application adopt the following technical solutions:

[0007] On one hand, this application provides a holographic traffic perception method based on wide-area radar, including: acquiring frame-by-frame trajectory data of multiple vehicle targets within a specific frequency coverage area using a roadside millimeter-wave radar to obtain initial tracking data of the multiple vehicle targets; acquiring frame-by-frame real-time driving trajectory data of the multiple vehicle targets using a GNSS dedicated data terminal to obtain initial driving data of the multiple vehicle targets; uniformly processing the initial tracking data and the initial driving data with coordinate data related to latitude and longitude offsets to obtain radar tracking data and target driving data respectively; performing trajectory topology calculations on the radar tracking data and the target driving data with respect to a target reference frame to obtain first state estimation information based on the radar tracking data and second state estimation information based on the target driving data respectively; generating a minimum topological distance optimization model based on the topological statistical distance between the first state estimation information and the second state estimation information; and distinguishing and identifying the relative positions of the multiple vehicle targets based on the minimum topological distance optimization model to obtain vehicle holographic information about the relative positions of the vehicles; and performing data parsing and calibration on the vehicle holographic information to achieve holographic traffic perception of vehicle targets across the entire area.

[0008] This application embodiment collects the motion trajectory data of all vehicles within a road segment using a roadside wide-area millimeter-wave radar, and simultaneously collects GNSS driving trajectory data of vehicles pushed by traffic information services using an in-vehicle mobile terminal. Preprocessing of both types of trajectory data improves data quality and achieves coordinate unification. Based on this, according to the correlation between the vehicle motion state sequences collected from the two types of trajectory data, consistent target matching is performed between the GNSS-based target driving data and the radar tracking data, thereby enabling the identification and differentiation of vehicles pushing information services and surrounding neighboring vehicles. Finally, the information dissemination module pushes the identified holographic traffic information to vehicles, providing real-time, holographic traffic information services to the public, facilitating the management of the spatiotemporal distribution of traffic flow, alleviating traffic congestion, and reducing vehicle accident rates.

[0009] In one feasible implementation, a millimeter-wave radar mounted on a roadside is used to collect trajectory data frame by frame from multiple vehicle targets within a specific frequency coverage area to obtain initial tracking data for the multiple vehicle targets. Specifically, this includes: using a millimeter-wave radar mounted on a roadside pole to collect video frames one by one from multiple vehicle targets within a specific frequency coverage area to obtain a set of vehicle tracking video frames for the multiple vehicle targets; according to [d x (i),d y (i),v x (i),v y (i),r(i)] kThe initial tracking data of the i-th vehicle target in the k-th frame of the vehicle tracking video frame set is obtained; where d x (i) represents the lateral distance between vehicle target i and the vertical line of the radar antenna; d y (i) represents the radial distance between vehicle target i and radar antenna; v x (i) represents the lateral velocity component of vehicle target i relative to the radar detection direction; v y (i) represents the radial velocity component of vehicle target i relative to the radar detection direction; r(i) represents the RCS return energy value of vehicle target i.

[0010] In one feasible implementation, the initial driving data of the multiple vehicle targets is obtained by frame-by-frame acquisition of real-time driving trajectory data of the vehicles themselves using a GNSS data terminal. Specifically, this includes: acquiring video frames of real-time driving trajectory data of the multiple vehicle targets one by one using a GNSS data terminal to obtain a set of vehicle driving video frames of the multiple vehicle targets; according to... The initial driving data concerning the latitude and longitude of the vehicle target is obtained in the k-th frame of the vehicle driving video frame set; where p x Indicates the longitude of the vehicle target pushed by the GNSS information service; p y Indicates the latitude of the vehicle target pushed by the GNSS information service; v g Indicates the target speed of the vehicle; The heading angle represents the direction of the vehicle's movement relative to true north.

[0011] In one feasible implementation, the initial tracking data and the initial driving data are subjected to unified coordinate data processing related to latitude and longitude offsets to obtain radar tracking data and target driving data respectively. Specifically, this includes: based on... The latitude and longitude offset Δp based on the coordinate data is obtained. x and Δp y ;in, The GNSS information service pushes the latitude and longitude coordinates of the vehicle target in the initial driving data, (d x d y ) represents the latitude and longitude coordinates acquired by the millimeter-wave radar antenna in the initial tracking data, and α represents the angle between the perpendicular vector and the true north vector in the radar antenna; according to Obtain the standard target coordinates (p) x ,p yThe standard target coordinates are standard coordinate templates used to perform position fusion transformation of the initial tracking data and the initial driving data in different coordinate systems. Based on the standard target coordinates, the initial tracking data and the initial driving data are uniformly standardized and transformed to obtain the radar tracking data and the target driving data respectively.

[0012] In one feasible implementation, according to The radar trajectory U in the radar tracking data is obtained. r and the driving trajectory U in the target driving data g ;in, Let m be the radar trajectory of the i-th vehicle target, and m be the total number of vehicles collected within the detection range of the millimeter-wave radar. Let m be the driving trajectory of the i-th vehicle target, n be the number of vehicles receiving GNSS information services, and m ≥ n; according to Obtain the transfer matrix based on radar tracking data Where, θ v For the radar trajectory U r The estimated velocity azimuth angle; based on A state estimate is obtained based on the radar tracking data. in, For the state estimation of the i-th radar trajectory at time k; This is for state estimation of a new target v based on the target reference system; the target reference system is a coordinate reference system with target v as the origin, transformed from m target data in the millimeter-wave radar coordinate system; according to Obtain the error covariance based on the radar trajectory in, This is the covariance estimate corresponding to the i-th radar trajectory at time k; This is the covariance estimate based on the new target v in the target reference frame; T is... The transpose of the target driving data is used to perform trajectory topology calculations with respect to the target reference frame, thereby determining the state estimate based on the target driving data. and error covariance in, This is a state estimation based on the new target s as the origin in the target reference frame. This is a covariance estimation based on the target reference frame with the new target s as the origin; wherein, the first state estimation information includes: a state estimate based on the radar tracking data. and error covariance The second state estimation information includes: a state estimate based on the target driving data. and error covariance

[0013] In one feasible implementation, before generating a minimum topological distance optimization model based on the topological statistical distance between the first state estimation information and the second state estimation information, the method further includes: based on State estimation of the radar tracking data obtained State estimation of the target driving data The state estimation difference Δ between ij (v,s); where v is the target coordinate system with the new target v as the origin, and s is the target coordinate system with the new target s as the origin; state estimation This is the state estimate of vehicle target i in the first state estimation information. This is the state estimate of vehicle target j in the second state estimation information; based on The difference Δ between the obtained state estimate and the state estimate is obtained. ij The error covariance B corresponding to (v,s) ij ; Let be the error covariance of vehicle target i in the first state estimation information. The error covariance of vehicle target j in the second state estimation information; according to The topological statistical distance γ is obtained. ij ; where T is the transpose operation.

[0014] In one feasible implementation, a minimum topological distance optimization model is generated based on the topological statistical distance between the first state estimation information and the second state estimation information, specifically including: based on... The minimum topological distance optimization model is determined; where γ ij Let η be the topological statistical distance. ij It is a binary variable; and The total number of vehicles collected within the detection range of the m-millimeter-wave radar, where n is the number of vehicles receiving GNSS information services, and η is the total number of vehicles in the entire area. ij =1 indicates that i and j come from the same vehicle target, η ij =0 indicates that the targets are from different vehicle targets. i is vehicle target i in the first state estimation information, j is vehicle target j in the second state estimation information, v is the target coordinate system based on the origin of the new target v, and s is the target coordinate system based on the origin of the new target s.

[0015] In one feasible implementation, based on the minimum topological distance optimization model, the relative positions of multiple vehicle targets are distinguished and identified to obtain vehicle holographic information about their relative positions. Specifically, this includes: associating the topological statistical distance in the minimum topological distance optimization model with a specific threshold; if the topological statistical distance is less than or equal to the specific threshold, then the relative position information of multiple vehicle targets in the first state estimation information and the second state estimation information are determined to be the same vehicle information; if the topological statistical distance is greater than the specific threshold, then the relative position information of multiple vehicle targets in the first state estimation information and the second state estimation information are determined to be adjacent vehicle information, so as to distinguish and identify the relative positions of each vehicle target based on the adjacent vehicle information; wherein, the vehicle holographic information includes: the same vehicle information and adjacent vehicle information based on the vehicle target.

[0016] In one feasible implementation, the vehicle holographic information is analyzed and calibrated, specifically including: analyzing the vehicle holographic information to determine the relative position information of the multiple vehicle targets within the entire acquisition area; based on the relative position information, marking the information service push vehicle as its own target, and marking the adjacent vehicles corresponding to the information service push vehicle as neighbor targets; wherein, the information service push vehicle is any vehicle among the multiple vehicle targets; and updating the own target and the neighbor targets in real time with relevant time frequencies to achieve holographic traffic perception of vehicle targets across the entire area.

[0017] On the other hand, this application also provides a holographic traffic perception system based on wide-area radar. The system includes: a roadside subsystem for acquiring frame-by-frame trajectory data of multiple vehicle targets within a specific frequency coverage area using a roadside millimeter-wave radar, obtaining initial tracking data of the multiple vehicle targets; further, for uniformly processing the initial tracking data and the initial driving data with coordinate data related to latitude and longitude offsets, respectively obtaining radar tracking data and target driving data; and for performing trajectory topology calculations on the radar tracking data and the target driving data with respect to a target reference frame, respectively obtaining first state estimation information based on the radar tracking data and... The system uses a second state estimation information based on the target driving data; generates a minimum topological distance optimization model based on the topological statistical distance between the first and second state estimation information; and, based on the minimum topological distance optimization model, distinguishes and identifies the relative positions of multiple vehicle targets to obtain vehicle holographic information about the relative positions of the vehicles; an onboard subsystem is used to collect frame-by-frame real-time driving trajectory data of the multiple vehicle targets through a dedicated GNSS data terminal to obtain the initial driving data of the multiple vehicle targets; and a cloud service subsystem is used to parse and calibrate the vehicle holographic information to achieve holographic traffic perception of vehicle targets across the entire domain.

[0018] This application provides a holographic traffic perception method and system based on wide-area radar. It collects the motion trajectory data of all vehicles within a road segment using a roadside wide-area millimeter-wave radar, and simultaneously collects GNSS driving trajectory data of vehicles pushed to traffic information services using an in-vehicle mobile terminal. Preprocessing of both types of trajectory data improves data quality and achieves coordinate unification. Based on this, according to the correlation between the vehicle motion state sequences collected from the two types of trajectory data, consistent target matching is performed between the GNSS-based target driving data and the radar tracking data, thereby enabling the identification and differentiation of vehicles pushing information services and surrounding neighboring vehicles. Finally, the information dissemination module pushes the identified holographic traffic information to vehicles, providing real-time, holographic traffic information services to the public, facilitating the management of the spatiotemporal distribution of traffic flow, alleviating traffic congestion, and reducing vehicle accident rates. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0020] Figure 1A flowchart of a holographic traffic perception method based on wide-area radar provided for embodiments of this application;

[0021] Figure 2 A holographic traffic perception system architecture based on wide-area radar is provided for embodiments of this application;

[0022] Figure 3 This is a schematic diagram of a holographic traffic perception system based on wide-area radar, provided as an embodiment of this application. Detailed Implementation

[0023] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0024] This application provides a holographic traffic perception method based on wide-area radar, such as... Figure 1 As shown, the holographic traffic perception method based on wide-area radar specifically includes steps S101-S106:

[0025] It should be noted that existing holographic traffic information service technologies mainly rely on intelligent vehicle location information collection terminals to detect continuous trajectory data of all vehicles, or on vehicle-mounted LiDAR to perceive vehicle targets in the surrounding road environment. However, for location collection terminal-based solutions, in the early stages of vehicle-to-everything (V2X) technology application, the penetration rate of intelligent vehicle terminals is low, making it impossible to achieve full coverage of vehicle targets on the road. For LiDAR-based solutions, only the status data of neighboring vehicles within a small range of the target vehicle can be collected, and LiDAR equipment is expensive, making it difficult to provide real-time, holographic traffic information services for the general public.

[0026] S101. Using roadside millimeter-wave radar, the trajectory data of multiple vehicle targets within a specific frequency coverage area is collected frame by frame to obtain the initial tracking data of the multiple vehicle targets.

[0027] Specifically, by using millimeter-wave radar installed on roadside poles, video frames of multiple vehicle targets within a specific frequency coverage area are collected one by one to obtain a set of vehicle tracking video frames of multiple vehicle targets.

[0028] Furthermore, according to [d] x (i),d y (i),v x (i),v y(i),r(i)] k This yields the initial tracking data for the i-th vehicle target in the k-th frame of the vehicle tracking video frame set. Where d x (i) represents the lateral distance between vehicle target i and the vertical line of the radar antenna. d y (i) represents the radial distance between vehicle target i and radar antenna. x (i) represents the lateral velocity component of vehicle target i relative to the radar detection direction. y (i) represents the radial velocity component of vehicle target i relative to the radar detection direction. r(i) represents the RCS return energy value of vehicle target i.

[0029] S102. Using a dedicated GNSS data terminal, collect frame-by-frame real-time driving trajectory data of multiple vehicle targets to obtain initial driving data of multiple vehicle targets.

[0030] Specifically, through a GNSS data terminal, video frames of the real-time driving trajectory data of multiple vehicle targets are collected one by one to obtain a set of vehicle driving video frames of multiple vehicle targets.

[0031] Furthermore, according to The initial driving data of the vehicle target's latitude and longitude in the k-th frame of the vehicle driving video frame set is obtained. Where p x This indicates the longitude of the vehicle target being pushed by the GNSS information service. y This indicates the latitude of the vehicle target being pushed by the GNSS information service. g Indicates the target speed of the vehicle. The heading angle represents the direction of the vehicle's movement relative to true north.

[0032] S103. Perform unified processing on the coordinate data of latitude and longitude offsets of the initial tracking data and initial driving data to obtain radar tracking data and target driving data respectively.

[0033] Specifically, according to Obtain the latitude and longitude offset Δp based on coordinate data x and Δp y .in, To push the latitude and longitude coordinates of the vehicle target to the GNSS information service in the initial driving data, (d x d y ) represents the latitude and longitude coordinates collected by the radar antenna of the millimeter-wave radar in the initial tracking data, and α represents the angle between the perpendicular vector of the radar antenna and the true north direction vector.

[0034] Furthermore, according to Obtain the standard target coordinates (p) x ,py The standard target coordinates are standard coordinate templates used to perform position fusion transformations of initial tracking data and initial driving data in different coordinate systems.

[0035] Furthermore, based on the standard target coordinates, the initial tracking data and initial driving data are uniformly standardized and transformed to obtain radar tracking data and target driving data respectively.

[0036] As a feasible implementation method, since millimeter-wave radar acquires coordinates as relative coordinates with the radar antenna as the origin, while GNSS terminal equipment acquires absolute latitude and longitude coordinates in the geodetic coordinate system, a unified transformation is performed on the two different coordinate systems contained in the initial tracking data acquired by millimeter-wave radar and the initial driving data acquired by GNSS, to achieve coordinate system unification processing for the information service push vehicle, i.e., to obtain unified and standardized radar tracking data and target driving data.

[0037] S104. Perform trajectory topology calculations on the radar tracking data and target driving data in relation to the target reference frame to obtain first state estimation information based on radar tracking data and second state estimation information based on target driving data, respectively.

[0038] Specifically, according to The radar trajectory U in the radar tracking data is obtained. r And the driving trajectory U in the target driving data g .in, Let m be the radar trajectory of the i-th vehicle target, and m be the total number of vehicles collected within the detection range of the millimeter-wave radar. Let m be the trajectory of the i-th vehicle target, n be the number of vehicles receiving GNSS information services, and m ≥ n.

[0039] Furthermore, according to Obtain the transfer matrix based on radar tracking data Where, θ v For radar trajectory U r The estimated velocity azimuth angle.

[0040] Furthermore, according to Obtain state estimation based on radar tracking data in, This is the state estimate of the i-th radar trajectory at time k. This is for state estimation of a new target v based on the target reference frame. The target reference frame is a coordinate reference frame with target v as the origin, transformed from the millimeter-wave radar coordinate system to the coordinate data of m targets.

[0041] Furthermore, according to Obtain the error covariance based on radar trajectory in, This is the covariance estimate corresponding to the i-th radar trajectory at time k. This is a covariance estimate based on the new target v in the target reference frame. T is... The transpose of the target data is then used. Similarly, the target driving data from n GNSS information services is transformed into a target coordinate system with target s as the origin. Then, trajectory topology calculations are performed on the target driving data in relation to the target reference system to determine the state estimate based on the target driving data. and error covariance in, This is a state estimation based on the new target s as the origin in the target reference frame. This is a covariance estimate based on the target reference frame with the new target s as the origin.

[0042] The first state estimation information includes: a state estimate based on radar tracking data. and error covariance The second state estimation information includes: state estimation based on target driving data. and error covariance

[0043] S105. Based on the topological statistical distance between the first state estimation information and the second state estimation information, a minimum topological distance optimization model is generated. Then, based on the minimum topological distance optimization model, the relative positions of multiple vehicle targets are distinguished and identified to obtain vehicle holographic information regarding the relative positions of the vehicles.

[0044] Specifically, according to State estimation based on radar tracking data State estimation with target driving data The state estimation difference Δ between ij (v,s). Where v is the target coordinate system with the new target v as the origin, and s is the target coordinate system with the new target s as the origin. State estimation. This is the state estimate of vehicle target i in the first state estimation information. This is the state estimate of vehicle target j in the second state estimation information.

[0045] Furthermore, according to The difference between the state estimate and the actual state estimate is Δ. ij The error covariance B corresponding to (v,s) ij . Let be the error covariance of vehicle target i in the first-state estimation information. Let be the error covariance of vehicle target j in the second-state estimation information.

[0046] Furthermore, according to Obtain the topological statistical distance γ ij Where T is the transpose operation.

[0047] Furthermore, based on The minimum topological distance optimization model was determined. Wherein, γ ij Let η be the topological statistical distance. ij It is a binary variable.

[0048] and The total number of vehicles collected within the detection range of the m-millimeter-wave radar, where n is the number of vehicles receiving GNSS information services, and η is the total number of vehicles in the entire area. ij =1 indicates that i and j come from the same vehicle target, η ij =0 indicates that the targets are from different vehicle targets. i is vehicle target i in the first state estimation information, j is vehicle target j in the second state estimation information, v is the target coordinate system based on the origin of the new target v, and s is the target coordinate system based on the origin of the new target s.

[0049] Furthermore, the topological statistical distance in the minimum topological distance optimization model is correlated with a specific threshold. If the topological statistical distance is less than or equal to the specific threshold, the relative position information of multiple vehicle targets in the first state estimation information and the second state estimation information is determined to be the same vehicle information. If the topological statistical distance is greater than the specific threshold, the relative position information of multiple vehicle targets in the first state estimation information and the second state estimation information is determined to be adjacent vehicle information, so as to distinguish and identify the relative position of each vehicle target based on the adjacent vehicle information. The vehicle holographic information includes: the same vehicle information based on the vehicle target and the adjacent vehicle information.

[0050] In one embodiment, after calculating the topological statistical distance between the first state estimation information and the second state estimation information, the radar tracking data and the target driving data are matched to the same target. By associating the topological statistical distance in the minimum topological distance optimization model with a specific threshold, the vehicle's own target and the surrounding adjacent vehicle targets are distinguished and identified.

[0051] S106. Perform data analysis and calibration on the vehicle holographic information to achieve holographic traffic perception of vehicle targets across the entire area.

[0052] Specifically, the vehicle holographic information is analyzed to determine the relative positions of multiple vehicle targets within the entire acquisition area.

[0053] Furthermore, based on relative location information, the vehicle providing the information service is marked as its own target, and adjacent vehicles corresponding to the vehicle providing the information service are marked as neighbor targets. The vehicle providing the information service can be any vehicle among multiple vehicle targets. Finally, real-time data updates are performed on the own target and neighbor targets regarding time frequency to achieve holographic traffic perception of all vehicle targets across the entire area.

[0054] In one embodiment, assume the w-th target is the vehicle itself, i.e., the information service push vehicle. The system marks the w-th target and wirelessly transmits the motion states of its m-1 surrounding neighboring targets to the GNSS mobile terminal device of the information service push vehicle. The GNSS mobile terminal device receives the data packets, parses them, and obtains information such as the position and speed of all vehicle targets in the entire domain. Centering on the location of the marked w-th vehicle target (its own target), all neighboring vehicles (neighboring targets) are marked at corresponding positions around it, and the relative position data is updated at a certain time frequency to realize holographic traffic information service.

[0055] In one embodiment, Figure 2 A holographic traffic perception system architecture based on wide-area radar is provided for embodiments of this application, as shown in the diagram. Figure 2 As shown:

[0056] 1. Vehicle Subsystem

[0057] The vehicle-mounted subsystem primarily handles GNSS data acquisition and holographic traffic information dissemination. It utilizes the GNSS sensor on the vehicle-mounted mobile terminal to collect continuous trajectory data and then transmits this data wirelessly to the cloud service subsystem via a communication interface module.

[0058] On the other hand, through the communication interface module, it receives holographic traffic data or other management data from the roadside subsystem and cloud service subsystem, and uses the information publishing module, including in-vehicle mobile terminal APP or dedicated in-vehicle display device, to display the holographic traffic information of the road and provide drivers with real-time information push services.

[0059] The communication interface module integrates multiple wireless communication modes, including 4G / 5G, LTE-V, 802.11p or other V2X communication methods, to achieve effective data interaction between the vehicle subsystem and the cloud service subsystem.

[0060] 2. Roadside subsystem

[0061] Based on the radar coverage area, the road network is divided into multiple roadside areas. Each radar-covered area is equipped with a separate roadside subsystem responsible for processing the data collected by the corresponding radar within the road area.

[0062] Continuous trajectory data of all vehicles within the coverage area is collected using a roadside millimeter-wave radar module. Simultaneously, GNSS trajectory data collected by the onboard subsystem of the information service push vehicle is received via a communication interface module. The two types of trajectory data undergo coordinate system unification in the data preprocessing module to achieve spatial calibration of the coordinate trajectory.

[0063] The trajectory data after coordinate unification is sent to the target matching and recognition module, where the target recognition and matching algorithm is executed to complete the identification of the vehicle's own target and neighboring targets in the information service push, thus obtaining holographic traffic data.

[0064] Holographic traffic data is sent to the cloud service subsystem via the communication interface module of the roadside subsystem, and then further transmitted to the vehicle subsystem to complete the holographic data release.

[0065] The algorithm processing modules and communication interface modules in the roadside subsystem constitute the edge computing unit of the subsystem.

[0066] 3. Cloud Service Subsystem

[0067] The cloud service subsystem primarily implements wireless data forwarding and service management functions. Utilizing the communication interface module, it receives vehicle GNSS trajectory data from the vehicle-mounted subsystem and matches it against intersections based on latitude and longitude to locate the vehicle's radar coverage area and corresponding roadside subsystems. The specific method is as follows:

[0068] Each roadside subsystem is assigned a latitude and longitude range based on the radar's coverage area and is maintained and managed within the cloud service subsystem. The system employs a real-time monitoring strategy; when the information service pushes vehicle GNSS data location data into a specific area, the system identifies the corresponding roadside subsystem and associates it with that subsystem. When a vehicle leaves the coverage area of ​​a roadside subsystem, the association is severed.

[0069] The backbone network between the cloud service subsystem and the roadside subsystem is used to forward vehicle GNSS data to the associated roadside subsystem.

[0070] In addition, the cloud service subsystem also has a data storage module to store dynamically collected trajectory data, management data, and business application data; a business application module to realize basic management, dynamic monitoring, data mining and other business functions of the roadside subsystem and vehicle subsystem; and a user interaction module to realize user operations for cloud management.

[0071] In addition, embodiments of this application also provide a holographic traffic perception system based on wide-area radar, such as... Figure 3 As shown, the holographic traffic perception system 300 based on wide-area radar specifically includes:

[0072] The roadside subsystem 310 is used to acquire trajectory data of multiple vehicle targets within a specific frequency coverage area using a roadside millimeter-wave radar, obtaining initial tracking data for the multiple vehicle targets. It is also used to uniformly process the initial tracking data and initial driving data for coordinate data with latitude and longitude offsets, respectively obtaining radar tracking data and target driving data. The radar tracking data and target driving data are then used to perform trajectory topology calculations based on a target reference frame, obtaining first state estimation information based on the radar tracking data and second state estimation information based on the target driving data. A minimum topology distance optimization model is generated based on the topology statistical distance between the first and second state estimation information. Based on the minimum topology distance optimization model, the relative positions of the multiple vehicle targets are distinguished and identified, obtaining vehicle holographic information regarding the relative positions of the vehicles.

[0073] The vehicle subsystem 320 is used to collect frame-by-frame real-time driving trajectory data of multiple vehicle targets through a dedicated GNSS data terminal to obtain the initial driving data of the multiple vehicle targets.

[0074] The cloud service subsystem 330 is used to analyze and calibrate vehicle holographic information to achieve holographic traffic perception of vehicle targets across the entire domain.

[0075] This application utilizes roadside wide-area millimeter-wave radar to collect trajectory data of all vehicles within a road segment, while simultaneously employing an in-vehicle mobile terminal to collect GNSS driving trajectory data of vehicles used for traffic information service delivery. Preprocessing of both types of trajectory data improves data quality and unifies coordinates. Based on this, and according to the correlation between vehicle motion state sequences collected from the two trajectory data sets, consistent target matching is performed between the GNSS-based target driving data and radar tracking data, thereby enabling the identification and differentiation of vehicles delivering information services and neighboring vehicles. Finally, the information dissemination module pushes the identified holographic traffic information to vehicles, providing real-time, holographic traffic information services to the public, facilitating the management of the spatiotemporal distribution of traffic flow, alleviating traffic congestion, and reducing vehicle accident rates.

[0076] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and system embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0077] The foregoing has described specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0078] The above description is merely an embodiment of this application and is not intended to limit this application. For those skilled in the art, various modifications and variations can be made to the embodiments of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the embodiments of this application should be included within the scope of the claims of this application.

Claims

1. A holographic traffic perception method based on wide-area radar, characterized in that, The method includes: By using a roadside millimeter-wave radar, the trajectory data of multiple vehicle targets within a specific frequency coverage area is collected frame by frame to obtain the initial tracking data of the multiple vehicle targets. The initial driving data of the multiple vehicle targets is obtained by frame-by-frame acquisition of the real-time driving trajectory data of the vehicles themselves through a dedicated GNSS data terminal. The initial tracking data and the initial driving data are processed by a unified coordinate data method involving latitude and longitude offsets to obtain radar tracking data and target driving data respectively. The radar tracking data and the target driving data are used to perform trajectory topology calculations with respect to the target reference frame to obtain first state estimation information based on the radar tracking data and second state estimation information based on the target driving data, respectively. Based on the topological statistical distance between the first state estimation information and the second state estimation information, a minimum topological distance optimization model is generated; and based on the minimum topological distance optimization model, the relative positions of multiple vehicle targets are distinguished and identified to obtain vehicle holographic information about the relative positions of the vehicles. The vehicle holographic information is analyzed and calibrated to achieve holographic traffic perception of vehicle targets across the entire area.

2. The holographic traffic perception method based on wide-area radar according to claim 1, characterized in that, By using a roadside millimeter-wave radar, frame-by-frame trajectory data of multiple vehicle targets within a specific frequency coverage area is collected to obtain initial tracking data for the multiple vehicle targets, specifically including: By using a millimeter-wave radar installed on a roadside pole, video frames of multiple vehicle targets within a specific frequency acquisition coverage area are collected one by one to obtain a set of vehicle tracking video frames of the multiple vehicle targets. according to The first video frame in the set of vehicle tracking videos is obtained. k Frame 1 i The initial tracking data for each vehicle target; wherein... Indicates vehicle target i The lateral distance between the radar antenna and the vertical line; This represents the radial distance between vehicle target i and the radar antenna; Indicates vehicle target i The lateral velocity component relative to the radar detection direction; Indicates vehicle target i The radial velocity component relative to the radar detection direction; Indicates vehicle target i The RCS returns a strong energy value.

3. The holographic traffic perception method based on wide-area radar according to claim 1, characterized in that, Using a GNSS data terminal, frame-by-frame acquisition of the real-time driving trajectory data of the multiple vehicle targets is performed to obtain the initial driving data of the multiple vehicle targets, specifically including: Using a GNSS data terminal, video frames of the real-time driving trajectory data of the multiple vehicle targets are collected one by one to obtain a set of vehicle driving video frames of the multiple vehicle targets. according to The first video frame in the set of vehicle driving video frames is obtained. k The initial driving data relating to the latitude and longitude of the vehicle target within the frame; wherein, Indicates the longitude of the vehicle target pushed by the GNSS information service; Indicates the latitude of the vehicle target pushed by the GNSS information service; Indicates the target speed of the vehicle; The heading angle represents the direction of the vehicle's movement relative to true north.

4. The holographic traffic perception method based on wide-area radar according to claim 1, characterized in that, The initial tracking data and the initial driving data are processed by unifying the coordinate data related to latitude and longitude offsets to obtain radar tracking data and target driving data, respectively, specifically including: according to The latitude and longitude offsets based on the coordinate data are obtained. as well as ;in, The GNSS information service pushes the latitude and longitude coordinates of the vehicle target in the initial driving data. d x , d y The coordinates () are the latitude and longitude coordinates collected by the radar antenna of the millimeter-wave radar in the initial tracking data. This represents the angle between the perpendicular bisector vector and the true north vector of the radar antenna; according to Obtain the standard target coordinates The standard target coordinates are standard coordinate templates used to perform position fusion transformations of the initial tracking data and the initial driving data in different coordinate systems. Based on the standard target coordinates, the initial tracking data and the initial driving data are subjected to a unified standardization transformation of coordinate data to obtain the radar tracking data and the target driving data, respectively.

5. A holographic traffic perception method based on wide-area radar according to claim 1, characterized in that, The radar tracking data and the target driving data are used to perform trajectory topology calculations with respect to the target reference frame to obtain first state estimation information based on the radar tracking data and second state estimation information based on the target driving data, specifically including: according to The radar trajectory in the radar tracking data is obtained. and the driving trajectory in the target driving data ;in, The radar trajectory of the i-th vehicle target. m This represents the total number of vehicles collected across the entire detection range of the millimeter-wave radar. For the first i The driving trajectory of the target vehicle. n To receive the number of vehicles in the GNSS information service, and ; according to The transfer matrix based on radar tracking data is obtained. ;in, For the radar trajectory The estimated velocity azimuth angle; according to A state estimate is obtained based on the radar tracking data. ;in, In the first k At that moment, the i State estimation of radar trajectory; For a new target based on the target reference frame v State estimation; the target reference system is the millimeter-wave radar coordinate system. m Transform target data into target v A coordinate reference system with the origin as its point; according to The error covariance based on the radar trajectory is obtained. ;in, In the first k At that moment, with the first i Covariance estimation for each radar trajectory; For a new target based on the target reference frame v Covariance estimation; T for transpose; The target driving data is used to perform trajectory topology calculations with respect to the target reference frame to determine the state estimate based on the target driving data. and error covariance ;in, To use a new target in the target reference frame s State estimation for the origin, To use a new target in the target reference frame s Covariance estimation for the origin; The first state estimation information includes: a state estimate based on the radar tracking data. and error covariance The second state estimation information includes: a state estimate based on the target driving data. and error covariance .

6. The holographic traffic perception method based on wide-area radar according to claim 1, characterized in that, Before generating a minimum topological distance optimization model based on the topological statistical distance between the first state estimation information and the second state estimation information, the method further includes: according to The state estimate of the radar tracking data is obtained. State estimation of the target driving data State estimation difference between ;in, v To be based on new goals v The target coordinate system is the origin. s To be based on new goals s The target coordinate system is defined by the origin; state estimation. For the vehicle target in the first state estimation information i State estimation, state estimation For the vehicle target in the second state estimation information j State estimation; according to The difference between the obtained state estimate and the state estimate is obtained. Corresponding error covariance ; Let be the error covariance of vehicle target i in the first state estimation information. For the vehicle target in the second state estimation information j The error covariance; according to The topological statistical distance is obtained. ;in, T This is for the transpose operation.

7. A holographic traffic perception method based on wide-area radar according to claim 6, characterized in that, Based on the topological statistical distance between the first state estimation information and the second state estimation information, a minimum topological distance optimization model is generated, specifically including: based on The minimum topological distance optimization model is determined; wherein, The topological statistical distance, It is a binary variable; and ; m The total number of vehicles collected across the entire area within the detection range of the millimeter-wave radar. n To receive the number of vehicles from GNSS information services, express i and j From the same vehicle target, This indicates that the vehicles are from different vehicle targets. i For the vehicle target in the first state estimation information i , j For the vehicle target in the second state estimation information j , v To be based on new goals v The target coordinate system of the origin s To be based on new goals s The target coordinate system of the origin.

8. A holographic traffic perception method based on wide-area radar according to claim 1, characterized in that, Based on the minimum topological distance optimization model, the relative positions of multiple vehicle targets are distinguished and identified to obtain vehicle holographic information regarding their relative positions, specifically including: The topological statistical distance in the minimum topological distance optimization model is correlated with a specific threshold. If the topological statistical distance is less than or equal to the specific threshold, then the relative position information of multiple vehicle targets in the first state estimation information and the second state estimation information are determined to be the same vehicle information; If the topological statistical distance is greater than the specific threshold, the relative position information of multiple vehicle targets in the first state estimation information and the second state estimation information is determined as adjacent vehicle information, so as to distinguish and identify the relative position of each vehicle target based on the adjacent vehicle information. The vehicle holographic information includes: information about the same vehicle and information about adjacent vehicles based on the vehicle target.

9. A holographic traffic perception method based on wide-area radar according to claim 1, characterized in that, The holographic information of the vehicle is analyzed and calibrated, specifically including: The holographic information of the vehicles is analyzed to determine the relative position information of the multiple vehicle targets within the entire acquisition area; Based on the relative position information, the information service push vehicle is marked as its own target, and the adjacent vehicles corresponding to the information service push vehicle are marked as neighbor targets; wherein, the information service push vehicle is any vehicle among the multiple vehicle targets; Real-time data updates on the target itself and its neighboring targets are performed at relevant time frequencies to achieve holographic traffic perception of all vehicle targets in the entire domain.

10. A holographic traffic perception system based on wide-area radar, characterized in that, The system includes: The vehicle-mounted subsystem is used to collect frame-by-frame real-time driving trajectory data of multiple vehicle targets through a dedicated GNSS data terminal to obtain the initial driving data of the multiple vehicle targets. The roadside subsystem is used to acquire trajectory data of multiple vehicle targets within a specific frequency coverage area frame by frame using a roadside millimeter-wave radar, obtaining initial tracking data for the multiple vehicle targets; it is also used to uniformly process the initial tracking data and the initial driving data with coordinate data related to latitude and longitude offsets, respectively obtaining radar tracking data and target driving data; it performs trajectory topology calculations on the radar tracking data and the target driving data with respect to a target reference frame, respectively obtaining first state estimation information based on the radar tracking data and second state estimation information based on the target driving data; it generates a minimum topology distance optimization model based on the topology statistical distance between the first state estimation information and the second state estimation information; and based on the minimum topology distance optimization model, it distinguishes and identifies the relative positions of the multiple vehicle targets, obtaining vehicle holographic information about the relative positions of the vehicles; The cloud service subsystem is used to perform data analysis and calibration on the vehicle holographic information in order to achieve holographic traffic perception of vehicle targets across the entire domain.

Images