Vehicle relay tracking method based on holographic road network

Abstract

The present application relates to the technical field of holographic road network, and in particular to a vehicle relay tracking method based on holographic road network. The method includes: in the process of tracking a vehicle based on a holographic road network, according to the real-time trajectory of the vehicle on the current monitored section, the driving trajectory of the vehicle on the next non-monitored section is predicted to obtain a predicted trajectory; the vehicle model is displayed on the holographic road network according to the spliced ​​trajectory, and the movement process from the first display area to the second display area; wherein the vehicle model is used to indicate the vehicle in the holographic road network; the first display area is used to indicate the current monitored section in the holographic road network, and the second display area is used to indicate the next non-monitored section in the holographic road network; the present application can solve the technical problem in the prior art that the coverage rate of monitoring equipment is still relatively low, resulting in the inability to achieve comprehensive and accurate vehicle relay tracking based on the holographic road network.

Description

Technical Field

[0001] This application relates to the technical field of holographic road networks, and in particular, to a vehicle relay tracking method based on a holographic road network. Background Art

[0002] In the prior art, a holographic road network is a technology that realizes all-weather accurate collection of information such as lane-level traffic flow, queue length, vehicle speed, driving trajectory, number of stops, and accidents at intersections by deploying monitoring devices at intersections and key sections, and then using data such as video data and radar data collected by the monitoring devices.

[0003] In practical applications, a holographic road network can achieve relay tracking of vehicles and display the moving process of the vehicles in real time. The relay tracking of vehicles relies on the video data and radar data provided by the monitoring devices.

[0004] In many existing areas, the coverage rate of monitoring devices is relatively low, resulting in the inability to achieve comprehensive and accurate vehicle relay tracking based on the holographic road network. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a vehicle relay tracking method based on a holographic road network to solve the technical problem in the prior art that due to the relatively low coverage rate of monitoring devices, it is impossible to achieve comprehensive and accurate vehicle relay tracking based on the holographic road network.

[0006] In a first aspect, this application provides a vehicle relay tracking method based on a holographic road network, and the method includes:

[0007] During the process of tracking a vehicle based on a holographic road network, according to the real-time trajectory of the vehicle on the current monitored section, predict the driving trajectory of the vehicle on the next non-monitored section to obtain a predicted trajectory;

[0008] Wherein, the next non-monitored section is the non-monitored section that the vehicle enters after driving away from the current monitored section;

[0009] In response to the vehicle driving from the current monitored section into the next non-monitored section, splice the real-time trajectory and the predicted trajectory to obtain a spliced trajectory;

[0010] On the holographic road network, display the moving process of the vehicle model driving from the first display area into the second display area according to the spliced trajectory;

[0011] Wherein, the vehicle model is used to indicate the vehicle in the holographic road network; the first display area is used to indicate the current monitored section in the holographic road network, and the second display area is used to indicate the next non-monitored section in the holographic road network.

[0012] In a second aspect, the present application provides a vehicle relay tracking device based on a holographic road network. The device includes: a trajectory prediction module, a trajectory splicing module, and a trajectory display module;

[0013] The trajectory prediction module is configured to predict a predicted trajectory of the vehicle on the next non-monitored road section according to the real-time trajectory of the vehicle on the current monitored road section during the process of tracking the vehicle based on the holographic road network;

[0014] Wherein, the next non-monitored road section is the non-monitored road section that the vehicle enters after driving away from the current monitored road section;

[0015] The trajectory splicing module is configured to splice the real-time trajectory and the predicted trajectory to obtain a spliced trajectory in response to the vehicle driving from the current monitored road section into the next non-monitored road section;

[0016] The trajectory display module is configured to display the moving process of the vehicle model driving from the first display area into the second display area on the holographic road network according to the spliced trajectory;

[0017] Wherein, the vehicle model is used to indicate the vehicle in the holographic road network; the first display area is used to indicate the current monitored road section in the holographic road network, and the second display area is used to indicate the next non-monitored road section in the holographic road network.

[0018] In a third aspect, the present application provides an electronic device. The electronic device includes a processor and a memory. The memory is used to store application programs. The processor runs or executes the software programs stored in the memory to enable the electronic device to implement the above-mentioned vehicle relay tracking method based on a holographic road network.

[0019] In a fourth aspect, the present application provides a computer-readable storage medium. The computer-readable storage medium is used to store program codes executed by a processor. The program codes are used to implement the above-mentioned vehicle relay tracking method based on a holographic road network.

[0020] In a fifth aspect, the present application provides a computer program product. The computer program product includes computer instructions. When the computer instructions run on an electronic device, the electronic device is enabled to implement the above-mentioned vehicle relay tracking method based on a holographic road network.

[0021] Advantageous effects:

[0022] The present application provides a vehicle relay tracking method based on a holographic road network. The method includes: during the process of tracking a vehicle based on the holographic road network, determining a predicted trajectory of the vehicle on a non-monitored road section according to the real-time trajectory of the vehicle on the monitored road section; wherein, the predicted trajectory is obtained by predicting according to the real-time trajectory; in response to the vehicle driving from the non-monitored road section into the monitored road section, determining a spliced trajectory according to the real-time trajectory and the predicted trajectory; displaying a vehicle model on the holographic road network according to the movement process of the vehicle model entering the second display area from the first display area along the spliced trajectory; wherein, the vehicle model is used to indicate the vehicle in the holographic road network; the first display area is used to indicate the monitored road section in the holographic road network, and the second display area is used to indicate the non-monitored road section in the holographic road network.

[0023] In summary, the vehicle relay tracking method of the holographic road network provided by the present application can determine the predicted trajectory of the non-monitored road section according to the real-time trajectory of the monitored road section, and splice the real-time trajectory and the predicted trajectory, so that the movement process of the vehicle model can be displayed in real time on the holographic road network according to the spliced trajectory. Therefore, the present application can solve the technical problem in the prior art that due to the relatively low coverage rate of monitoring devices, it is impossible to achieve comprehensive and accurate vehicle relay tracking based on the holographic road network. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the embodiments of the present application. The following drawings only show some embodiments of the present application, so they should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained according to these drawings.

[0025] Figure 1 It is a flowchart of the vehicle relay tracking method based on the holographic road network provided by the embodiment of the present application;

[0026] Figure 2 It is a schematic diagram of the setting of the monitored road section and the non-monitored road section provided by the embodiment of the present application;

[0027] Figure 3 It is a schematic diagram of the movement process of displaying the vehicle model in the holographic road network provided by the embodiment of the present application;

[0028] Figure 4 It is another schematic diagram of the movement process of displaying the vehicle model in the holographic road network provided by the embodiment of the present application;

[0029] Figure 5 It is a structural diagram of the vehicle relay tracking device based on the holographic road network provided by the embodiment of the present application. DETAILED DESCRIPTION

[0030] In the prior art, the holographic road network is a virtual traffic network based on Digital Twin technology. By using lidar-camera to fully cover and sense road sections and intersections, and converging and fusing vehicle trajectories at intersections; combined with 3D maps and equipment modeling, a digital base is constructed to complete the digitization and visualization of traffic elements such as vehicles, roads, and facilities for display.

[0031] Based on the holographic road network, vehicle tracking can be realized. The vehicle can be a vehicle with special uses that needs to be tracked, or a vehicle that needs to be tracked by certain public security, procuratorial, and judicial departments in accordance with regulations, etc. However, in many existing areas, the coverage rate of lidar-cameras is still relatively low, resulting in the inability to achieve comprehensive and accurate vehicle relay tracking based on the holographic road network.

[0032] To solve the above technical problems, the present application provides a technical solution for vehicle tracking based on a holographic road network. In this technical solution, through the real-time trajectory of the vehicle on the monitored road section, the driving trajectory of the vehicle on the non-monitored road section is predicted to obtain a predicted trajectory; when the vehicle enters the monitored road section from the non-monitored road section, the real-time trajectory and the predicted trajectory can be spliced to obtain a spliced trajectory; when the spliced trajectory is obtained, the vehicle model displayed on the holographic road network can be made to move from the first display area to the second display area according to the spliced trajectory.

[0033] Among them, the vehicle model is used to indicate the vehicle in the holographic road network; the first display area is used to indicate the monitored road section in the holographic road network, and the second display area is used to indicate the non-monitored road section in the holographic road network.

[0034] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0035] First, the present application provides a vehicle relay tracking method based on a holographic road network, as Figure 1 shown, Figure 1 is a flowchart of the vehicle relay tracking method based on the holographic road network provided by the embodiments of the present application. The method includes: S110~S130, details are as follows:

[0036] S110: During the process of tracking a vehicle based on the holographic road network, according to the real-time trajectory of the vehicle on the current monitored road section, predict the driving trajectory of the vehicle on the next non-monitored road section to obtain a predicted trajectory;

[0037] Among them, the next non-monitored section is the non-monitored section that the vehicle enters after leaving the current monitored section.

[0038] Specifically, the "current monitored section" is the monitored section that the vehicle is currently driving on, and the "next non-monitored section" is the non-monitored section that the vehicle enters immediately after leaving the current monitored section.

[0039] In actual operation, the "monitored sections" and "non-monitored sections" are arranged alternately. When the vehicle leaves a "monitored section", it immediately enters a "non-monitored section", and when the vehicle leaves a "non-monitored section", it immediately enters a "monitored section".

[0040] In the embodiments of the present application, early prediction can be performed on the vehicle in the "next non-monitored section". For example, when there are multiple sections between the vehicle and the "next non-monitored section", the driving trajectory of the vehicle in the "next non-monitored section" can be predicted. However, the number of vehicles and the traffic operation conditions on different sections are likely to be different. Therefore, the predicted trajectory obtained by predicting the driving trajectory of the vehicle in the "next non-monitored section" when the vehicle is driving on the "current monitored section" is the closest to the actual trajectory of the vehicle in the "next non-monitored section" because the "current monitored section" is the closest to the "next non-monitored section".

[0041] After determining the predicted trajectory, the vehicle model of the vehicle can be made to move on the road shown in the holographic road network according to the predicted trajectory on the holographic road network, so that the user can intuitively observe the driving state of the vehicle on the holographic road network, which is convenient for making decisions based on this driving state.

[0042] In actual operation, both the "real-time trajectory" and the "predicted trajectory" are trajectory data sets, and the elements in the trajectory data set are the real-time coordinates and timestamps in the coordinate system where the holographic road network is located.

[0043] In one implementation, before S110, the method further includes: steps (1) to (2), details are as follows:

[0044] Step (1): Collect real-time driving data for the vehicle through the radar-vision camera set on the current monitored section.

[0045] Among them, the monitored section is the monitoring and recognition area of the radar-vision camera, and the non-monitored section is the monitoring blind area of the radar-vision camera.

[0046] Specifically, in the embodiments of the present application, as Figure 2 shown Figure 2This is a schematic diagram of the setting of the monitored section and the non-monitored section provided by the embodiment of the present application. The monitored section is the section where the radar-vision camera is installed and within the monitoring and recognition area of the radar-vision camera, and the non-monitored section is the section where the radar-vision camera is not installed and within the monitoring blind area of the radar-vision camera.

[0047] In actual operation, when a vehicle is driving on the monitored section, the real-time driving data of the vehicle can be collected through the radar-vision camera, such as the specific position coordinates of the vehicle, the instantaneous speed and average speed, and the data of the vehicle type, color, size, and license plate of each vehicle; among them, the specific position coordinates of the vehicle are the coordinates of a certain feature point detected by the radar-vision camera, and this feature point is usually the position where the radar beam intersects the vehicle surface, and it may be located at the front, side, or rear of the vehicle, specifically depending on the installation position of the radar-vision camera, the detection angle, and the shape and size of the vehicle.

[0048] Step (2): Determine the real-time trajectory according to the real-time driving data.

[0049] Specifically, after determining the real-time driving data, the real-time trajectory of the vehicle can be determined according to the real-time driving data; the holographic road network can refer to this real-time trajectory and make the vehicle model in the holographic road network display the moving process of moving in the first display area according to the real-time trajectory, and its moving process is the process that the user can observe the vehicle model continuously moving on the holographic road network.

[0050] Among them, the model used to indicate the vehicle in the holographic road network is the vehicle model; the display area used to indicate the monitored section in the holographic road network is the first display area, and the display area used to indicate the non-monitored section in the holographic road network is the second display area.

[0051] Such as Figure 3 and Figure 4 shown, Figure 3 This is a schematic diagram of the moving process of the vehicle model displayed in the holographic road network provided by the embodiment of the present application. Figure 4 This is another schematic diagram of the moving process of the vehicle model displayed in the holographic road network provided by the embodiment of the present application. In Figure 3 when the moving process of the vehicle model is displayed on the holographic road network, the license plate of the vehicle will also be displayed. Figure 4 In addition to displaying the moving process of the vehicle model on the holographic road network, a display interface for performing other functions of the holographic road network is also displayed.

[0052] It should be emphasized that whether the holographic road network distinguishes and displays the first display area and the second display area can be determined according to the actual situation. It should also be emphasized that the background server is fully aware of the monitored section and the non-monitored section in the actual road network, because the above information can be determined through urban infrastructure data.

[0053] In one implementation, S110 includes: step (3), details are as follows:

[0054] Step (3): According to the real-time trajectory, use the trajectory prediction model to predict the trajectory of the vehicle on the next non-monitored section to obtain the predicted trajectory;

[0055] Among them, the trajectory prediction model is obtained by training a big data model.

[0056] Specifically, when applying the trajectory prediction model, it is also necessary to input the section length of the current monitored section corresponding to the real-time trajectory and the section length of the next non-monitored section, so that the trajectory prediction model outputs the predicted trajectory corresponding to the section length of the next non-monitored section.

[0057] In the embodiments of the present application, the big data model is an AI (Artificial Intelligence) big model; in actual operation, in order to enable the AI big model to have the prediction ability to determine the predicted trajectory, it is necessary to perform pre-training and fine-tuning on the AI big model. In the prior art, the traffic microscopic simulation model can be used to predict the driving trajectories of multiple vehicles on non-monitored sections to obtain the predicted trajectories of multiple vehicles. When the predicted trajectories of multiple vehicles are determined, the predicted trajectories of each vehicle among the multiple vehicles can be determined. However, the above operation method has application problems such as insufficient original data, high requirements for data processing technology, long data processing cycle, and very difficult calibration of parameters and variables, resulting in a very slow process of predicting the driving trajectories of multiple vehicles on non-monitored sections by the traffic microscopic simulation model, which is difficult to meet the real-time prediction requirements. Especially in actual operation, in many cases, it is not necessary to obtain the predicted trajectories of multiple vehicles, but only to determine the driving trajectory of the target vehicle among the multiple vehicles on the holographic road network, which further reduces the efficiency of obtaining the predicted trajectory of the target vehicle.

[0058] In the embodiments of the present application, in order to meet the real-time prediction requirements and real-time prediction efficiency and improve the prediction accuracy of the predicted trajectory of the target vehicle, the present application trains a trajectory prediction model that only predicts the predicted trajectory of the target vehicle.

[0059] In some implementations, step (3) includes: step (3.1) to step (3.4), details are as follows:

[0060] Step (3.1): Determine the first parameter according to the real-time trajectory of the current monitored section.

[0061] Among them, the first parameter indicates the influence of non-section factors on the driving state of the vehicle;

[0062] Specifically, in the embodiments of the present application, non-road-section factors refer to factors that generally do not change in real time, excluding the length, curvature, etc. of the road section. Non-road-section factors mainly include: weather conditions, traffic congestion conditions, etc.; in practical applications, non-road-section factors can be accurately obtained through the radar-vision cameras on the currently monitored road section.

[0063] In actual situations, real-time weather conditions will affect both the number of vehicles on the road section and the driving state of the target vehicle. Since the trajectory prediction model in the embodiments of the present application only focuses on the prediction of the driving trajectory of the target vehicle, and the number of vehicles on the road section will also affect the driving state of the target vehicle, therefore, the influence of weather conditions on the number of vehicles on the road section (and the influence on the driving state of the vehicle) is abstracted into the influence on the driving state of the target vehicle, so that the trajectory prediction model only considers the final influence of this weather condition on the predicted trajectory of the target vehicle; among them, the weather condition can be input into the trajectory prediction model by using meteorological professional terms as prompt words (i.e., the first parameter), so that the trajectory prediction model can perform efficient and accurate real-time prediction on the driving trajectory of the vehicle on the next non-monitored road section according to the weather condition of the currently monitored road section.

[0064] Traffic congestion may spread from the currently monitored road section to the next non-monitored road section. Therefore, it is necessary to consider the influence of this traffic accident situation on the driving state of the target vehicle. In the prior art, traffic congestion is used to describe the driving conditions of multiple vehicles in a road section. However, in the embodiments of the present application, this traffic congestion situation is abstracted into the influence on the driving state of the target vehicle, so that the trajectory prediction model only considers the final influence of this traffic congestion situation on the predicted trajectory of the target vehicle; among them, the traffic accident situation can be input into the trajectory prediction model by using the specific information of the traffic congestion situation as prompt words (i.e., the first parameter), so that the trajectory prediction model can perform efficient and accurate real-time prediction on the driving trajectory of the vehicle on the next non-monitored road section according to the traffic congestion situation of the currently monitored road section.

[0065] Step (3.2): According to the road-section type of the next non-monitored road section, determine the reference monitored road section among the currently monitored road section where the vehicle is traveling and at least one historical monitored road section that the vehicle has traveled through;

[0066] Among them, the road-section type includes: straight road section and curved road section; the road-section type of the reference monitored road section is the same as that of the next non-monitored road section.

[0067] Specifically, in the embodiments of the present application, to determine the predicted trajectory, it is necessary to retrieve the historical movement trajectories of the monitored road sections with the same road-section type as the next non-monitored road section.

[0068] In the embodiments of the present application, road sections are classified into two major categories: straight road sections and curved road sections. Among them, a straight road section refers to a section where a vehicle travels straight along the road center line or the main road without changing its driving direction; on a straight road section, the driver needs to keep the vehicle driving straight without deviating from the road center line or the main road; the accident risk is relatively low.

[0069] A curved road section refers to the curved part of the road, that is, the part where the route is not straight; on a curved road section, the driver needs to pay special attention to the line of sight and the driving stability of the vehicle; the accident risk is relatively high.

[0070] In the embodiments of the present application, the reason for selecting the reference monitoring road section is that the driving behaviors and trajectory patterns of vehicles on the same type of road section have high similarity. By analyzing the driving data of vehicles on the reference monitoring road section, typical driving habits and speed patterns of vehicles on this type of road section can be extracted, providing a strong basis for predicting the trajectory of the vehicle on the next non-monitored road section.

[0071] Step (3.3): Retrieve the current second parameter of the next non-monitored road section, the reference movement trajectory of the reference monitoring road section, and the reference second parameter from the database;

[0072] Among them, the second parameter indicates the influence of the road section on the driving state of the vehicle.

[0073] Specifically, in actual operation, due to the lack of real-time vehicle position and speed data, the prediction of the driving trajectory of vehicles on non-monitored road sections becomes particularly complex. Therefore, accurate prediction requires comprehensive consideration of the vehicle's historical driving behaviors, current state, and the influence of the second parameter on the vehicle's driving state.

[0074] In the embodiments of the present application, the second parameter should include a series of factors that affect the driving state of the vehicle, such as road section length, width, and speed limit, etc. This second parameter and the first parameter jointly determine the possibility of the vehicle's driving speed and direction change on this road section. In the embodiments of the present application, the trajectory prediction model is also connected to the database, and can retrieve the corresponding second parameter from the records in the database according to the fields and can store the second parameter into the records corresponding to the fields.

[0075] Step (3.4): Input the first parameter, the current second parameter, the reference movement trajectory, and the reference second parameter into the trajectory prediction model, and make the trajectory prediction model output the predicted trajectory.

[0076] Specifically, after inputting the first parameter, the current second parameter, the reference movement trajectory, and the reference second parameter into the trajectory prediction model, the trajectory prediction model can be made to adaptively adjust the reference movement trajectory according to the first parameter, the current second parameter, and the reference second parameter, so as to obtain the most likely driving trajectory of the target vehicle on the next non-monitored road section, and this most likely driving trajectory is the predicted trajectory.

[0077] It should be emphasized that although the specific structure of the trajectory prediction model is not limited in this application, the model structure of the trajectory prediction model should have a neural network structure for processing time series data and capturing time-dependent relationships.

[0078] In one implementation manner, between step (3.3) and step (3.4), the method further includes: steps (3.5) to (3.6), and the details are as follows:

[0079] Step (3.5): If the current number m of the reference monitored road sections is less than the preset number N, according to the current second parameter, (N - m) historical second parameters and the corresponding (N - m) historical movement trajectories are retrieved from the database;

[0080] Among them, the difference between each of the (N - m) historical second parameters and the current second parameter is less than or equal to a preset threshold.

[0081] Specifically, in actual operation, the accuracy and reliability of the predicted trajectory output by the trajectory prediction model depend on the reference second parameters and reference movement trajectories of rich reference monitored road sections to cover as comprehensive scenarios as possible. Therefore, when the current number m of the reference monitored road sections is less than the preset number N, it is necessary to supplement the number of the reference second parameters and reference movement trajectories of the reference monitored road sections to the preset number N; the preset number N can be determined according to actual needs, and this application does not make specific limitations on this.

[0082] In actual operation, the setting of the preset threshold can ensure that the retrieved historical second parameter and the current second parameter have sufficient similarity; the preset threshold can be determined according to actual needs, but if the threshold is set too high, it may lead to a large difference between the retrieved data and the target road section, affecting the prediction accuracy; if the threshold is set too low, it may limit the number of retrievable data.

[0083] Step (3.6): Determine the (N - m) historical movement trajectories and the (N - m) historical second parameters as (N - m) reference movement trajectories and (N - m) reference second parameters.

[0084] In one implementation, the trajectory prediction model includes: a first trajectory prediction model and a second trajectory prediction model; step (3.4) includes: step (3.4.1) to step (3.4.2), details are as follows:

[0085] Step (3.4.1): If the next non-monitored section is a straight section, input the first parameter, the current second parameter, the reference movement trajectory, and the reference second parameter into the first trajectory prediction model, and let the trajectory prediction model output the predicted trajectory of the vehicle on the straight section.

[0086] Specifically, on sections with different road section types, there may be significant differences in the driving states and behavior patterns of vehicles. Therefore, two specialized trajectory prediction models, namely the first trajectory prediction model and the second trajectory prediction model, are required to predict the driving trajectories of vehicles on straight sections and curved sections respectively; the first trajectory prediction model is specifically designed to predict the driving trajectory of vehicles on straight sections; the second trajectory prediction model is specifically designed to predict the driving trajectory of vehicles on curved sections.

[0087] Step (3.4.2): If the next non-monitored section is a curved section, input the first parameter, the current second parameter, the reference movement trajectory, and the reference second parameter into the second trajectory prediction model, and let the trajectory prediction model output the predicted trajectory of the vehicle on the curved section.

[0088] Specifically, in actual operation, compared with the fine-tuning samples of the first trajectory prediction model during the fine-tuning process, the fine-tuning samples of the second trajectory prediction model during the fine-tuning process also need to include the unique second parameters of the curved section, such as curvature and turning radius, etc.

[0089] S120: In response to the vehicle driving from the current monitored section into the next non-monitored section, splice the real-time trajectory and the predicted trajectory to obtain a spliced trajectory.

[0090] Specifically, in actual operation, by effectively splicing the predicted trajectory of the vehicle on the next non-monitored section with the real-time trajectory on the current monitored section, a complete movement trajectory can be formed, enabling users to visually observe the continuous movement process of the vehicle model on the holographic road network.

[0091] In actual operation, time synchronization processing needs to be performed on the real-time trajectory and the predicted trajectory. Specifically, the end timestamp of the predicted trajectory and the start timestamp of the real-time trajectory are determined as the splicing time point; on the basis of time synchronization, the last position point of the predicted trajectory and the first position point of the real-time trajectory need to be compared. If the distance between the two position points is within the preset distance, it is considered that the position matching is successful and the trajectory splicing can be performed. If the distance between the two position points is not within the preset distance, the predicted trajectory needs to be adjusted to splice it with the real-time trajectory.

[0092] S130: Display the moving process of the vehicle model driving from the first display area into the second display area on the holographic road network according to the splicing trajectory;

[0093] Among them, the vehicle model is used to indicate the vehicle in the holographic road network; the first display area is used to indicate the current monitored section in the holographic road network, and the second display area is used to indicate the next non-monitored section in the holographic road network.

[0094] Specifically, the holographic road network can accurately display the moving process of the vehicle model entering the second display area from the first display area, ensuring that users can clearly see the moving process of the vehicle model.

[0095] In one implementation, the method further includes: steps (4) to (7), the details are as follows:

[0096] Step (4): During the process of tracking the vehicle based on the holographic road network, if there is an intersection in the current non-monitored section where the vehicle is currently driving, in response to the vehicle entering the intersection, obtain the current signal light state of the signal lights in the intersection.

[0097] Specifically, the holographic road network provides real-time road condition information and vehicle dynamic display. In order to more accurately track the driving situation of the vehicle in the non-monitored section and adjust the predicted trajectory of the vehicle according to the signal light state when encountering an intersection; in actual operation, it is necessary to continuously judge whether there is an intersection in the non-monitored section where the vehicle is currently located during the vehicle tracking process; if there is an intersection, enter the next step; if there is no intersection, continue to track according to the current predicted trajectory. When the vehicle enters the intersection, obtain the current signal light state (such as red light, green light, yellow light, etc.) of the current intersection signal lights through the communication interface or data sharing platform with the intersection signal light system; it should be emphasized that ensure that the obtained signal light state data is accurate and timely to reflect the actual traffic situation at the intersection.

[0098] Step (5): Determine the predicted passing state of the vehicle at the intersection according to the current signal light state.

[0099] Specifically, in actual operation, the passing state of the vehicle within the current signal light cycle can be judged according to the obtained current signal light state; for example, if the signal light is green, the vehicle may pass through the intersection without stopping; if the signal light is red, the vehicle needs to stop and wait at the intersection.

[0100] In actual operation, the specific time for the vehicle to pass through the intersection or the waiting time for parking can be predicted by combining the signal light state and information such as the current position and speed of the vehicle. The confirmation of the above information helps to accurately adjust the vehicle trajectory subsequently.

[0101] Step (6): If the predicted traffic state indicates that the vehicle needs to stop at the intersection, adjust the current predicted trajectory of the vehicle to obtain an adjusted predicted trajectory.

[0102] Specifically, based on the predicted traffic state, determine whether the vehicle needs to stop at the intersection; if it needs to stop, proceed to the next step; if it does not need to stop, continue to track according to the current predicted trajectory.

[0103] In actual operation, when the vehicle needs to stop at the intersection, adjust the current predicted trajectory of the vehicle according to the stop position and stop time.

[0104] It should be emphasized that the adjusted predicted trajectory should reflect the process of the vehicle waiting at the intersection and then continuing to drive according to the change of the traffic signal. It should also be emphasized that optimize the adjusted predicted trajectory to ensure that it conforms to the actual road conditions and vehicle driving rules.

[0105] Step (7): In the second display area corresponding to the current non-monitored section in the holographic road network, display the moving process of the vehicle model moving according to the adjusted predicted trajectory.

[0106] Specifically, update the adjusted predicted trajectory in real time to the corresponding second display area in the holographic road network, that is, the current non-monitored section, to ensure that the updated trajectory is consistent with the actual driving situation of the vehicle. In the holographic road network, display the moving process of the vehicle model moving according to the adjusted predicted trajectory in a dynamic manner. Through the movement of the vehicle model, intuitively reflect the vehicle's stop waiting and subsequent driving situation at the intersection.

[0107] In one implementation, the first moment of the vehicle model at the target location is later than the second moment of receiving the real-time driving data of the vehicle at the target location on the holographic road network; the method further includes: Step (8) to Step (10), the details are as follows:

[0108] Step (8): During the process of tracking the vehicle based on the holographic road network, in response to the vehicle entering the current monitored section from the historical non-monitored section at the third moment, obtain the real-time driving data of the vehicle driving on the current monitored section.

[0109] Specifically, the meaning of "the first moment when the vehicle model is displayed on the holographic road network at the target location is later than the second moment when the real-time driving data of the vehicle at the target location is received" is that the first moment when the real-time driving data of the vehicle is received from the background server of the holographic road network and the second moment when the vehicle model is displayed on the holographic road network are not the same, and the first moment is earlier than the second moment. That is, there is a time delay between the "moment of displaying the trajectory" and the "moment of receiving the driving data". Therefore, in practical applications, when the vehicle model on the holographic road network is still in the second display area corresponding to the historical non-monitored section, the target vehicle has already entered the current monitored section.

[0110] In actual operation, after predicting the driving trajectory of the vehicle on the non-monitored section, the vehicle model in the holographic road network is made to move according to the predicted trajectory. However, if there is a deviation between the predicted trajectory and the actual trajectory of the vehicle, when the vehicle enters the monitored section and the real-time trajectory is determined based on the real-time driving data of the vehicle, the above-mentioned actual trajectory and the real-time trajectory are not completely aligned, resulting in the phenomenon of "jumping" of the vehicle model on the holographic road network and even causing the entire screen to freeze. This is very unfavorable for tracking the vehicle. Therefore, in the embodiment of the present application, the real-time trajectory of the vehicle in the current monitored section obtained by using the above time delay is used to correct the predicted trajectory on which the vehicle model on the holographic road network is moving to obtain a repaired trajectory, so that the vehicle model on the holographic road network can move smoothly according to the repaired trajectory.

[0111] Step (9): Input the remaining predicted trajectory of the vehicle on the historical non-monitored section and the real-time trajectory on the current monitored section into the trajectory repair model, and make the trajectory repair model repair the spliced trajectory of the remaining predicted trajectory and the real-time trajectory to obtain the repaired trajectory of the vehicle on the current monitored section.

[0112] Among them, the remaining predicted trajectory is the part of the predicted trajectory corresponding to the time period after the third moment in the predicted trajectory of the historical non-monitored section; the trajectory repair model is obtained by training a big data model.

[0113] Specifically, when applying the trajectory repair model, the road section length of the non-monitored section corresponding to the remaining predicted trajectory also needs to be input, so that the trajectory repair model outputs the repaired trajectory corresponding to the road section length of the remaining predicted trajectory.

[0114] In practical applications, since both the "real-time trajectory" and the "predicted trajectory" in the embodiments of the present application are trajectory datasets, and the elements in the trajectory dataset are the real-time coordinates and timestamps in the coordinate system where the holographic road network is located. Therefore, when the vehicle model in the holographic road network moves according to the repaired trajectory, it may exhibit behaviors such as "accelerating", "decelerating", or "changing the driving direction", so that it moves "smoothly" in the repaired trajectory corresponding to the remaining predicted trajectory, rather than showing phenomena such as "jumping".

[0115] Step (10): Display on the holographic road network the moving process of the vehicle model moving from the second display area corresponding to the historical non-monitored section to the first display area corresponding to the current monitored section according to the repaired trajectory at the fourth moment.

[0116] Among them, the fourth moment is later than the third moment.

[0117] Specifically, when the repaired trajectory is determined, the vehicle model in the holographic road network can be made to move from the second display area corresponding to the historical non-monitored section to the first display area corresponding to the current monitored section according to the repaired trajectory. However, the moment when the vehicle model enters the corresponding first display area in the holographic road network is the fourth moment, and this fourth moment is later than the third moment because there is a time delay between the "moment of displaying the trajectory" and the "moment of receiving the driving data".

[0118] Second, the present application provides a vehicle relay tracking device based on a holographic road network, as Figure 5 shown, Figure 5 is the structural diagram of the vehicle relay tracking device based on the holographic road network provided by the embodiments of the present application. The device includes: a trajectory prediction module 310, a trajectory splicing module 320, and a trajectory display module 330;

[0119] The trajectory prediction module 310 is used to predict the driving trajectory of the vehicle on the next non-monitored section to obtain a predicted trajectory according to the real-time trajectory of the vehicle on the current monitored section during the process of tracking the vehicle based on the holographic road network;

[0120] Among them, the next non-monitored section is the non-monitored section that the vehicle enters after leaving the current monitored section;

[0121] The trajectory splicing module 320 is used to splice the real-time trajectory and the predicted trajectory to obtain a spliced trajectory in response to the vehicle moving from the current monitored section to the next non-monitored section;

[0122] The trajectory display module 330 is used to display on the holographic road network the moving process of the vehicle model moving from the first display area to the second display area according to the spliced trajectory;

[0123] Among them, the vehicle model is used to indicate a vehicle in the holographic road network; the first display area is used to indicate the current monitored section in the holographic road network, and the second display area is used to indicate the next non-monitored section in the holographic road network.

[0124] In one implementation, the device further includes: a data acquisition module;

[0125] The data acquisition module is configured to collect real-time driving data for the vehicle through a radar-vision camera set on the current monitored section;

[0126] Among them, the monitored section is the monitoring and recognition area of the radar-vision camera, and the non-monitored section is the monitoring blind area of the radar-vision camera;

[0127] The data acquisition module is further configured to determine a real-time trajectory according to the real-time driving data.

[0128] In one implementation, the trajectory prediction module 310 is further configured to predict a predicted trajectory for the vehicle on the next non-monitored section through a trajectory prediction model according to the real-time trajectory;

[0129] Among them, the trajectory prediction model is obtained by training a big data model.

[0130] In one implementation, the trajectory prediction module 310 is further configured to determine a first parameter according to the real-time trajectory of the current monitored section;

[0131] Among them, the first parameter indicates the influence of non-section factors on the driving state of the vehicle;

[0132] The trajectory prediction module 310 is further configured to determine a reference monitored section among the current monitored section where the vehicle is traveling and at least one historical monitored section that the vehicle has traveled through according to the section type of the next non-monitored section;

[0133] Among them, the section types include: straight sections and curved sections; the section type of the reference monitored section is the same as that of the next non-monitored section;

[0134] The trajectory prediction module 310 is further configured to retrieve the current second parameter of the next non-monitored section, the reference movement trajectory and the reference second parameter of the reference monitored section from the database;

[0135] Among them, the second parameter indicates the influence of section factors on the driving state of the vehicle;

[0136] The trajectory prediction module 310 is further configured to input the first parameter, the current second parameter, the reference movement trajectory and the reference second parameter into the trajectory prediction model, and make the trajectory prediction model output a predicted trajectory.

[0137] In one implementation, the trajectory prediction module 310 is further configured to, if the current number m of the reference monitoring sections is less than the preset number N, retrieve (N - m) historical second parameters and the corresponding (N - m) historical movement trajectories from the database according to the current second parameter;

[0138] Wherein, the difference between each of the (N - m) historical second parameters and the current second parameter is less than or equal to the preset threshold;

[0139] The trajectory prediction module 310 is further configured to determine the (N - m) historical movement trajectories and the (N - m) historical second parameters as (N - m) reference movement trajectories and (N - m) reference second parameters.

[0140] In one implementation, the trajectory prediction model includes: a first trajectory prediction model and a second trajectory prediction model;

[0141] If the next non-monitoring section is a straight section, the trajectory prediction module 310 inputs the first parameter, the current second parameter, the reference movement trajectory, and the reference second parameter into the first trajectory prediction model, and enables the trajectory prediction model to output the predicted trajectory of the vehicle on the straight section;

[0142] The trajectory prediction module 310 is further configured to, if the next non-monitoring section is a curved section, input the first parameter, the current second parameter, the reference movement trajectory, and the reference second parameter into the second trajectory prediction model, and enable the trajectory prediction model to output the predicted trajectory of the vehicle on the curved section.

[0143] In one implementation, during the process of tracking the vehicle based on the holographic road network, if there is an intersection in the current non-monitoring section where the vehicle is traveling, the trajectory prediction module 310, in response to the vehicle entering the intersection, acquires the current signal state of the traffic lights in the intersection;

[0144] The trajectory prediction module 310 is further configured to determine the predicted passing state of the vehicle at the intersection according to the current signal state;

[0145] If the predicted passing state indicates that the vehicle needs to stop at the intersection, the trajectory prediction module 310 adjusts the current predicted trajectory of the vehicle to obtain an adjusted predicted trajectory;

[0146] The trajectory prediction module 310 is further configured to display the moving process of the vehicle model moving according to the adjusted predicted trajectory in the second display area corresponding to the current non-monitoring section in the holographic road network.

[0147] In one implementation, the display of the vehicle model at the target location on the holographic road network is later than the second moment when the real-time driving data of the vehicle at the target location is received; the trajectory prediction module 310 is further configured to, during the process of tracking the vehicle based on the holographic road network, in response to the vehicle driving into the current monitored section from the historical non-monitored section at the third moment, obtain the real-time trajectory of the vehicle driving on the current monitored section;

[0148] The trajectory prediction module 310 is further configured to input the remaining predicted trajectory of the vehicle on the historical non-monitored section and the real-time trajectory on the current monitored section into the trajectory repair model, and enable the trajectory repair model to repair the spliced trajectory of the remaining predicted trajectory and the real-time trajectory to obtain the repaired trajectory of the vehicle on the current monitored section;

[0149] Wherein, the remaining predicted trajectory is the partial predicted trajectory corresponding to the time period after the third moment in the predicted trajectory of the historical non-monitored section; the trajectory repair model is obtained by training for the big data model;

[0150] The trajectory prediction module 310 is further configured to display on the holographic road network the moving process of the vehicle model driving from the second display area corresponding to the historical non-monitored section into the first display area corresponding to the current monitored section according to the repaired trajectory at the fourth moment;

[0151] Wherein, the fourth moment is later than the third moment.

[0152] Third, the present application further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of S110~S130 provided in the above embodiments are implemented.

[0153] Fourth, the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by the processor, the steps of S110~S130 of the above embodiments are executed.

[0154] Fifth, the computer program product provided by the present application includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods in the foregoing method embodiments. For the specific implementation, reference can be made to the steps of S110~S130 of the method embodiments, which will not be elaborated herein.

[0155] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces. The indirect coupling or communication connection of devices or units can be in electrical, mechanical or other forms.

[0156] In addition, the units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0157] Furthermore, in each embodiment of the present application, the various functional modules can be integrated together to form an independent part, or each module can exist alone, or two or more modules can be integrated to form an independent part.

[0158] It should be noted that if the function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0159] In this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0160] The above are only embodiments of the present application and are not intended to limit the protection scope of the present application. For those skilled in the art, various modifications and changes can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A vehicle relay tracking method based on a holographic road network, characterized in that: The method comprises: In the process of tracking the vehicle based on the holographic road network, according to the real-time trajectory of the vehicle on the current monitored road section, the driving trajectory of the vehicle on the next non-monitored road section is predicted to obtain a predicted trajectory; The next non-monitored road section is the non-monitored road section that the vehicle enters after leaving the current monitored road section; In response to the vehicle driving from the current monitored section into the next non-monitored section, splicing the real-time trajectory and the predicted trajectory to obtain a spliced ​​trajectory; Displaying on the holographic road network the movement process of the vehicle model from the first display area to the second display area according to the splicing trajectory; The vehicle model is used to indicate the vehicle in the holographic road network; the first display area is used to indicate the current monitored road section in the holographic road network, and the second display area is used to indicate the next non-monitored road section in the holographic road network; The step of predicting the driving trajectory of the vehicle on the next non-monitored road section according to the real-time trajectory of the vehicle on the current monitored road section to obtain a predicted trajectory includes: According to the real-time trajectory, a trajectory of the vehicle on the next non-monitored road section is predicted by a trajectory prediction model to obtain the predicted trajectory; Wherein, the trajectory prediction model is obtained by training the big data model; The step of predicting the trajectory of the vehicle on the next non-monitored road section according to the real-time trajectory by using a trajectory prediction model to obtain the predicted trajectory includes: Determining a first parameter according to the real-time trajectory of the current monitored road section; Wherein, the first parameter indicates the influence of non-road section factors on the driving state of the vehicle; Determine, according to the road section type of the next non-monitored road section, a reference monitored road section from a current monitored road section traveled by the vehicle and at least one historical monitored road section traveled by the vehicle; The road section types include: straight road sections and curved road sections; the road section type of the reference monitoring road section is the same as the road section type of the next non-monitoring road section; Retrieving the current second parameter of the next non-monitored section, the reference moving trajectory and the reference second parameter of the reference monitored section from the database; Wherein, the second parameter indicates the influence of the road section factor on the driving state of the vehicle; The first parameter, the current second parameter, the reference moving trajectory and the reference second parameter are input into the trajectory prediction model, so that the trajectory prediction model outputs the predicted trajectory.

2. The method according to claim 1, characterized in that Before predicting the driving trajectory of the vehicle on the next non-monitored road section to obtain a predicted trajectory based on the real-time trajectory of the vehicle on the current monitored road section, the method further includes: The real-time driving data is obtained by collecting the driving data of the vehicle through a radar camera arranged on the currently monitored road section; The monitored road section is the monitoring identification area of ​​the radar camera, and the non-monitored road section is the monitoring blind area of ​​the radar camera; The real-time trajectory is determined according to the real-time driving data.

3. The method according to claim 2, characterized in that After determining the reference monitoring section, the method further includes: If the current number m of the reference monitoring sections is less than the preset number N, according to the current second parameter, (Nm) historical second parameters and corresponding (Nm) historical movement trajectories are retrieved from the database; Wherein, the difference between each of the (Nm) historical second parameters and the current second parameter is less than or equal to a preset threshold; The (Nm) historical movement trajectories and the (Nm) historical second parameters are determined as (Nm) reference movement trajectories and (Nm) reference second parameters.

4. The method according to claim 2, characterized in that: The trajectory prediction model includes: a first trajectory prediction model and a second trajectory prediction model; the first parameter, the current second parameter, the reference moving trajectory and the reference second parameter are input into the trajectory prediction model, and the trajectory prediction model outputs the predicted trajectory, including: If the next non-monitored road section is the straight road section, input the first parameter, the current second parameter, the reference moving trajectory and the reference second parameter into the first trajectory prediction model, so that the trajectory prediction model outputs the predicted trajectory of the vehicle on the straight road section; If the next non-monitored road section is the curved road section, the first parameter, the current second parameter, the reference moving trajectory and the reference second parameter are input into the second trajectory prediction model, so that the trajectory prediction model outputs the predicted trajectory of the vehicle in the curved road section.

5. The method according to claim 1, characterized in that The method further comprises: In the process of tracking the vehicle based on the holographic road network, if there is an intersection in the current non-monitored road section where the vehicle is traveling, in response to the vehicle entering the intersection, obtaining the current signal light state of the signal light in the intersection; Determining a predicted traffic state of the vehicle at the intersection according to the current traffic light state; If the predicted traffic status indicates that the vehicle needs to stop at the intersection, adjusting the current predicted trajectory of the vehicle to obtain an adjusted predicted trajectory; In the second display area corresponding to the current non-monitored road section in the holographic road network, the movement process of the vehicle model moving according to the adjusted predicted trajectory is displayed.

6. The method according to claim 2, characterized in that The first time when the vehicle model is displayed at the target location on the holographic road network is later than the second time when the real-time driving data of the vehicle at the target location is received; the method further comprises: In the process of tracking the vehicle based on the holographic road network, in response to the vehicle entering the current monitored road section from the historical non-monitored road section at a third moment, obtaining a real-time trajectory of the vehicle traveling on the current monitored road section; Inputting the remaining predicted trajectory of the vehicle on the historical non-monitored section and the real-time trajectory on the current monitored section into a trajectory repair model, allowing the trajectory repair model to repair the vehicle on the current monitored section according to the spliced ​​trajectory of the remaining predicted trajectory and the real-time trajectory; The remaining predicted trajectory is a portion of the predicted trajectory of the historical non-monitored road section corresponding to the period after the third moment; the trajectory repair model is obtained by training the big data model; Displaying on the holographic road network a moving process of the vehicle model according to the repair trajectory, at a fourth moment, driving from the second display area corresponding to the historical non-monitored road section into the first display area corresponding to the current monitored road section; The fourth moment is later than the third moment.

7. A vehicle relay tracking device based on a holographic road network, characterized in that: For implementing the method of claim 1, the device comprises: a trajectory prediction module, a trajectory splicing module and a trajectory display module; The trajectory prediction module is used to predict the driving trajectory of the vehicle on the next non-monitored road section according to the real-time trajectory of the vehicle on the current monitored road section to obtain a predicted trajectory during the process of tracking the vehicle based on the holographic road network; The next non-monitored road section is the non-monitored road section that the vehicle enters after leaving the current monitored road section; The trajectory splicing module is used for splicing the real-time trajectory and the predicted trajectory to obtain a spliced ​​trajectory in response to the vehicle entering the next non-monitored road section from the current monitored road section; The track display module is used to display the movement process of the vehicle model from the first display area to the second display area according to the splicing track on the holographic road network; The vehicle model is used to indicate the vehicle in the holographic road network; the first display area is used to indicate the current monitored section in the holographic road network, and the second display area is used to indicate the next non-monitored section in the holographic road network.

8. An electronic device, characterized in that: The electronic device includes a processor and a memory, the memory is used to store an application, and the processor runs or executes a software program stored in the memory so that the electronic device implements the vehicle relay tracking method based on a holographic road network as described in any one of claims 1 to 6.

Images