Vehicle driving track monitoring method based on digital twinning and application thereof
Through digital twin technology combining image monitoring and deep learning models, the problem of operating vehicles' driving trajectory monitoring and review is solved, intuitive monitoring and efficient review of vehicle driving conditions is achieved, and reliable data support is provided.
Patent Information
- Application Number
- CN202411325099.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-23
- Publication Date
- 2025-01-14
AI Technical Summary
The prior art is difficult to effectively monitor and review the driving trajectory of operating vehicles, especially when there are many vehicles, it is difficult to judge the vehicle situation by combining a single positioning data and an electronic map. Recalling the image data requires a lot of manpower or computing resources, and Not suitable for large-scale applications.
Using a vehicle driving trajectory monitoring method based on digital twins, the vehicle in the preset public area is image monitoring and positioning data acquisition, and a digital twin model is constructed in combination with terrain and building information, and the vehicle trajectory data is generated, and the vehicle is predicted using deep learning models. The stay time and discrete situations at designated locations provide real-time monitoring and post-review functions.
It realizes intuitive, convenient monitoring and efficient review of the driving conditions of operating vehicles, improves monitoring efficiency and simulation, and provides reliable data support for background management, suitable for real-time monitoring and future prediction.
Smart Images
Figure CN119312666A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of digital twin technology and vehicle driving monitoring technology, and in particular to a vehicle driving trajectory monitoring method based on digital twin and its application. Background Art
[0002] As an important means of transportation for public travel and economic activities, most operating vehicles travel according to preset route plans when they are working. During the operation of operating vehicles, relevant back-end personnel also monitor their driving and operation conditions to avoid abnormal operations. With the increasing number of operating vehicles, how to monitor the driving trajectory of operating vehicles and how to invert the driving conditions of operating vehicles afterwards have become difficult problems for back-end supervisors to manage in real time and review afterwards. At present, some literature has disclosed the combination of electronic maps and positioning units deployed on vehicles to realize the monitoring of vehicle driving conditions. However, due to the single The combination of positioning data and electronic maps is often difficult to make a good judgment on the actual situation of the vehicle. In this case, some literature has disclosed the use of driving recorders to collect images to review the vehicle's driving conditions. However, due to the large storage space occupied by image data, content extraction and analysis requires a lot of manpower or computer computing resources. Therefore, reviewing through image data is not suitable for situations with a large number of operating vehicles. Therefore, how to optimize the data structure and integrate multiple data of operating vehicles at work to improve the monitoring efficiency and post-review simulation during the driving process of operating vehicles is a very positive and practical research topic. Summary of the invention
[0003] In view of this, the purpose of the present invention is to propose a vehicle driving trajectory monitoring method based on digital twins and its application, which is reliable in implementation, flexible in application, and can integrate multiple data of operating vehicles at work and improve the intuitiveness of on-site monitoring and the degree of post-review simulation.
[0004] In order to achieve the above technical objectives, the technical solution adopted by the present invention is:
[0005] A vehicle driving trajectory monitoring method based on digital twins, comprising:
[0006] Perform image monitoring on the vehicle driving area within the preset public area and generate image monitoring data;
[0007] In response to the start signal of the target vehicle, the position of the target vehicle is monitored according to preset conditions to generate vehicle positioning data;
[0008] Based on the vehicle positioning data and combined with the image monitoring data, the driving trajectory of the target vehicle in the preset public area is determined to generate vehicle trajectory data;
[0009] Construct a target vehicle database, collect vehicle trajectory data, vehicle positioning data and image monitoring data corresponding to the target vehicle to form a target vehicle data set, and store multiple target vehicle data sets in the target vehicle database;
[0010] Build a digital twin model to map the terrain and building information within the preset public area;
[0011] According to the vehicle positioning data or vehicle trajectory data corresponding to the target vehicle, the driving trajectory of the target vehicle is monitored in the digital twin model.
[0012] As a possible implementation method, further, the solution also includes:
[0013] In response to the start signal of the target vehicle, the image monitoring is performed in the target vehicle according to the preset conditions to collect the image viewing angle of the target vehicle in the road direction and generate driving image data in time series;
[0014] The target vehicle data set also includes driving image data corresponding to the target vehicle.
[0015] As a preferred implementation option, preferably, the target vehicles described in this solution are multiple, and the basic information of the multiple target vehicles is collected;
[0016] The basic information includes: one or more of the vehicle model, appearance specifications, exterior color, and license plate number;
[0017] The image monitoring data, vehicle positioning data and vehicle trajectory data are all generated in time series.
[0018] As a preferred implementation option, preferably, this solution also includes:
[0019] Marking the vehicle positioning data of all target vehicle data sets in the target vehicle database in combination with the electronic map to identify the location where the target vehicle appears on the electronic map and the corresponding date and time point, generating marking information, and then associating the marking information with the target vehicle data set;
[0020] The tag information and / or the license plate number of the target vehicle is used as one of the index information of the target vehicle data set.
[0021] As a preferred implementation option, preferably, this solution also includes:
[0022] The corresponding target vehicle data set is retrieved from the target vehicle database according to preset conditions, and then the driving trajectory of the target vehicle is inverted in the digital twin model based on the target vehicle data set.
[0023] As a preferred implementation option, preferably, this solution includes:
[0024] A demonstration location and a demonstration time period are designated within a preset public area, where the demonstration time period is a past time period, and a target vehicle that appears at the demonstration location within the demonstration time period is determined based on the demonstration location and the demonstration time period in combination with the marking information.
[0025] When the target vehicle is not present, "No target vehicle" is returned as the confirmation result, and then the 3D view of the demonstration location is directly displayed in the digital twin model.
[0026] When there is more than one target vehicle, the vehicle trajectory data and / or driving image data of the target vehicle appearing at the demonstration location during the demonstration time period are retrieved from the target vehicle data set in the target vehicle database, and then mapped in the digital twin model to invert the driving trajectory of the target vehicle at the specified demonstration location and demonstration time period.
[0027] As a preferred implementation option, preferably, this solution also provides multi-view data for display when inverting the driving trajectory of the target vehicle in the digital twin model;
[0028] The multi-view data includes at least one of the following:
[0029] (1) Invert the target vehicle’s driving trajectory in the digital twin model from a bird’s-eye view;
[0030] (2) The driving image data is used as the main display data, and the position change of the target vehicle in the digital twin model is used as the auxiliary display data to invert the driving trajectory of the target vehicle.
[0031] As a preferred implementation option, preferably, this solution also includes:
[0032] Based on the vehicle positioning data and corresponding marking information in multiple target vehicle data sets in the target vehicle database, a deep learning model is constructed or trained to predict the target vehicle's stay time and / or discrete trajectory at a preset location and a preset time. The prediction results of the deep learning model are then demonstrated in the digital twin model.
[0033] Based on the above, the present invention also provides an operating vehicle monitoring method, which includes the above-mentioned vehicle driving trajectory monitoring method based on digital twins, and the target vehicle is an operating vehicle.
[0034] Based on the above, the present invention also provides a vehicle driving trajectory monitoring system based on digital twins, which includes:
[0035] There are multiple image monitoring units, which are arranged within the preset public area and perform image monitoring on the vehicle driving area within the preset public area to generate image monitoring data;
[0036] A vehicle positioning unit is used to respond to a start signal of a target vehicle, monitor the position of the target vehicle according to preset conditions, and generate vehicle positioning data;
[0037] A data association unit is used to determine the driving trajectory of the target vehicle in a preset public area based on the vehicle positioning data and in combination with the image monitoring data, and generate vehicle trajectory data;
[0038] A data storage unit is used to construct a target vehicle database, collect vehicle trajectory data, vehicle positioning data and image monitoring data corresponding to the target vehicle, form a target vehicle data set, and store multiple target vehicle data sets in the target vehicle database;
[0039] Digital twin unit, used to build a digital twin model to map the terrain and building information within the preset public area;
[0040] The vehicle monitoring unit is used to monitor the driving trajectory of the target vehicle in the digital twin model according to the vehicle positioning data or vehicle trajectory data corresponding to the target vehicle.
[0041] By adopting the above-mentioned technical scheme, the present invention has the following beneficial effects compared with the prior art: the present scheme cleverly establishes a digital twin model so that the terrain and building information of the target vehicle driving area can be mapped into the model, and then the vehicle driving area within the preset public area is image-monitored, and the position of the target vehicle after starting is detected, and the image monitoring data and vehicle positioning data are combined to form vehicle trajectory data, and then the vehicle trajectory data is associated with the digital twin model, so that the back-end management personnel can more intuitively and conveniently monitor the vehicle driving conditions in real time or conduct a later inversion review, and by collecting the vehicle trajectory data, vehicle positioning data and image monitoring data corresponding to the target vehicle to form a target vehicle data set, the target vehicle data set is used to train the deep learning model to obtain a prediction model for predicting the discrete situation of the target vehicle at a specified location at a future time point, and provide data prediction reference for the back-end management personnel to manage operating vehicles or road traffic management. The present scheme is not only reliable and flexible to implement, but can also be used for real-time monitoring or post-event review of operating vehicles, providing convenient and reliable data and mechanism support for management personnel. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0043] Figure 1 This is a brief implementation flow chart of the monitoring method of this program;
[0044] Figure 2 It is a brief illustration of the inversion of the driving trajectories of target vehicles A, B, and C at a certain intersection in the monitoring method of this scheme;
[0045] Figure 3 This is a schematic diagram of the unit module connection of the monitoring system of this scheme. DETAILED DESCRIPTION
[0046] The present invention will be further described in detail below in conjunction with the accompanying drawings and examples. It is particularly noted that the following examples are only used to illustrate the present invention, but are not intended to limit the scope of the present invention. Similarly, the following examples are only partial embodiments of the present invention rather than all embodiments, and all other embodiments obtained by those of ordinary skill in the art without creative work are within the scope of protection of the present invention.
[0047] like Figure 1 As shown, this embodiment provides a vehicle driving trajectory monitoring method based on digital twins, which includes:
[0048] S010, performing image monitoring on the vehicle driving area within the preset public area to generate image monitoring data;
[0049] S020, responding to a start signal of the target vehicle, monitoring the position of the target vehicle according to preset conditions, and generating vehicle positioning data;
[0050] S030, determining the driving trajectory of the target vehicle in a preset public area based on the vehicle positioning data and in combination with the image monitoring data, and generating vehicle trajectory data;
[0051] S040, constructing a target vehicle database, collecting vehicle trajectory data, vehicle positioning data and image monitoring data corresponding to the target vehicle to form a target vehicle data set, and storing multiple target vehicle data sets in the target vehicle database;
[0052] S050. Build a digital twin model to map the terrain and building information within the preset public area;
[0053] S060. Monitor the driving trajectory of the target vehicle in the digital twin model according to the vehicle positioning data or vehicle trajectory data corresponding to the target vehicle.
[0054] Since the scenes of the digital twin model are mostly idealized, when the target vehicle is moving, it may be disturbed by other vehicles, pedestrians and other factors and make actions such as changing lanes and slowing down during driving. In order to facilitate the understanding of the actual on-site situation when monitoring the driving trajectory of the target vehicle, this solution can also obtain the road conditions from the perspective of the inside of the target vehicle through the driving record image. As a possible implementation method, further, when responding to the start signal of the target vehicle, this solution also includes:
[0055] S021. Respond to a start signal of the target vehicle, perform image monitoring in the target vehicle according to preset conditions, collect image viewing angles of the target vehicle in the direction of the road, and generate driving image data in a time series.
[0056] Wherein, in S040, the target vehicle data set also includes driving image data corresponding to the target vehicle.
[0057] In this solution, the recording of driving image data can help to verify the driving conditions of the target vehicle from the first perspective when the driving conditions of the target vehicle are subsequently reviewed. Especially when some accident scene situations are simulated, the driving image data can provide a better display of the actual situation.
[0058] When there are multiple target vehicles, in order to facilitate data matching and judgment, as a better implementation option, preferably, there are multiple target vehicles in this scheme, and basic information of the multiple target vehicles is collected; the basic information includes: one or more of vehicle model, appearance specifications, appearance color, and license plate number.
[0059] In order to facilitate data retrieval, the image monitoring data, vehicle positioning data and vehicle trajectory data described in this solution are all generated in time series.
[0060] Combination Figure 2 As shown, through the above method, when the driving trajectory of multiple target vehicles is monitored or inverted in the digital twin model, the position of the target vehicle in the model can be quickly and conveniently displayed on the map of the digital twin model by simply generating vehicle positioning data or vehicle trajectory data in time series. Figure 2 The image shows the positions of target vehicles A, B, and C at a certain intersection at the same time point, and the dotted lines show their driving trajectories. On this basis, the driving image data can also be displayed as the first-person perspective data of the target vehicle in the secondary display screen of the digital twin model to facilitate backend management personnel to intuitively understand the driving conditions of the target vehicle.
[0061] Since there are a large number of target vehicles, the data generated by them may be inconvenient to retrieve or difficult to temporarily process the data content. Therefore, in order to facilitate data retrieval, as a better implementation option, preferably, this solution also includes:
[0062] A01. Marking the vehicle positioning data of all target vehicle data sets in the target vehicle database in combination with the electronic map to identify the location where the target vehicle appears on the electronic map and the corresponding date and time point, generating marking information, and then associating the marking information with the target vehicle data set;
[0063] A02. Use the marking information and / or the license plate number of the target vehicle as one of the index information of the target vehicle data set.
[0064] When the above scheme displays the target vehicle based on the digital twin model, it is done through the location of the target vehicle. Therefore, by reusing the vehicle positioning data and combining the marking information to mark the date and time points corresponding to the different locations of the target vehicle on the electronic map, it is helpful to display the target vehicle in the digital twin model in real time, or to invert the driving conditions of the target vehicle in the digital twin model afterwards.
[0065] As one of the cases of data inversion, as a better implementation option, preferably, this solution also includes:
[0066] S070. Retrieve a corresponding target vehicle data set from a target vehicle database according to preset conditions, and then invert the driving trajectory of the target vehicle in the digital twin model based on the target vehicle data set.
[0067] For example, a certain area (electronic fence) on the electronic map is designated as the demonstration location to demonstrate the entry and exit of target vehicles within 24 hours before the specified time point. The target vehicle may refer to an operating vehicle. The basic information of the target vehicles can be collected to form a basic database, which is then mapped to the digital twin model for easy retrieval. In this way, the requirement to demonstrate the driving trajectories of all target vehicles can be achieved. For example, clicking on a vehicle on the interface of the digital twin model can demonstrate the route of the vehicle from the forward perspective (through the driving image data of the target vehicle), as well as reproduce the map scene with 3D simulation.
[0068] In addition, as a better implementation option, preferably, in the demonstration of the digital twin model on the target vehicle, this solution includes:
[0069] A demonstration location and a demonstration time period are designated within a preset public area, where the demonstration time period is a past time period, and a target vehicle that appears at the demonstration location within the demonstration time period is determined based on the demonstration location and the demonstration time period in combination with the marking information.
[0070] When the target vehicle is not present, "No target vehicle" is returned as the confirmation result, and then the 3D view of the demonstration location is directly displayed in the digital twin model.
[0071] When there is more than one target vehicle, the vehicle trajectory data and / or driving image data of the target vehicle appearing at the demonstration location during the demonstration time period are retrieved from the target vehicle data set in the target vehicle database, and then mapped in the digital twin model to invert the driving trajectory of the target vehicle at the specified demonstration location and demonstration time period.
[0072] Among them, as a better implementation option, preferably, this solution also provides multi-view data for display when inverting the driving trajectory of the target vehicle in the digital twin model;
[0073] The multi-view data includes at least one of the following:
[0074] (1) Invert the target vehicle’s driving trajectory in the digital twin model from a bird’s-eye view;
[0075] (2) The driving image data is used as the main display data, and the position change of the target vehicle in the digital twin model is used as the auxiliary display data to invert the driving trajectory of the target vehicle.
[0076] In addition to the above-mentioned real-time monitoring or post-inversion of the target vehicle through the digital twin model, as a better implementation option, preferably, this solution also includes:
[0077] Based on the vehicle positioning data and corresponding marking information in multiple target vehicle data sets in the target vehicle database, a deep learning model is constructed or trained to predict the target vehicle's stay time and / or discrete trajectory at a preset location and a preset time. The prediction results of the deep learning model are then demonstrated in the digital twin model.
[0078] The above scheme uses the data of multiple target vehicle data sets in the target vehicle database as the basis to train a deep learning model, which is used to predict the departure or stay time of the target vehicle at a certain location in the future, so as to facilitate the background supervision personnel to provide data reference for the future scheduling of the target vehicle, wherein the deep learning model can be a neural network model or other learning models.
[0079] As an example, in the application of neural network model to predict vehicle stay, this solution can extract vehicle positioning data and corresponding tag information from multiple target vehicle data sets, and then establish a data set of the target vehicle's stay time at location A on different dates and times, which can be expressed as: D iA ={(W1, T1), (W2, T2)......(W n , T n )}, where D iA is the target vehicle i’s residence time dataset at location A, W n Indicates date and time, T n Represents the length of stay; after the stay time data sets corresponding to all target vehicles are collected, they are divided into training sets and validation sets, and then imported into the neural network for training until the model converges, a prediction neural network for predicting the stay market of the target vehicle at a certain location is obtained, which is then used to predict the stay time of the target vehicle at a certain location at a future time point, and this result is combined with the historical driving trajectory of the target vehicle at a certain location (usually, the trajectory of vehicles traveling on public roads has a high similarity) as a driving trajectory parameter and mapped into the digital twin model to predict the discrete situation of vehicles at a certain location in a specific time period.
[0080] In order to better improve the reliability of data prediction, when constructing the residence time dataset, we can also combine the date and time W n The weather conditions, working days, non-working days, holidays, non-holidays and other factors of a certain location are added to the data set as one of the parameters and imported into the neural network for learning, so that the neural network can obtain more comprehensive environmental factors and internal factors for training, and obtain a more reliable and accurate prediction neural network.
[0081] As another example of a solution for predicting the duration of a target vehicle's stay at a certain location (set as location A), it may include the following:
[0082] (1) Collect the historical data set of the target vehicle staying at location A in the target vehicle data set, and set it to D = {(t1, d1), (t2, d2) ... (t i , d i )}, where t i is the starting time of the target vehicle’s i-th stop at location A, d i is the duration that the target vehicle stays at location A for the i-th time;
[0083] (2) The time data in the dataset D is converted into a timestamp format, and then the time features are extracted, which include hours (h i =hour(t i )), day of the week (w i=weekday(t i ))、Month(m i =month(t i ));
[0084] (3) Create a feature matrix, combining the extracted features into a feature matrix, which includes:
[0085] X=[X1,X2,……,X i ]
[0086] X i =[h i ,w i ,m i ]
[0087] Y=[Y1,Y2,……,Y i ]
[0088] Y i =d i
[0089] (4) Support vector regression (SVR) or random forest regression (RFR) is used as the machine learning model for training.
[0090] Among them, the formula of the support vector regression SVR model is defined as follows:
[0091]
[0092] Among them, K(X i ,X) is the kernel function, which is used to map the input features to a high-dimensional space; α i , is the Lagrange multiplier, represents the support vector, and b is the bias top.
[0093] The formula of random forest regression RFR is defined as follows:
[0094] The RFR model performs regression by constructing multiple decision trees and taking the average prediction results. The formula is as follows:
[0095]
[0096] Where T represents the number of decision trees, f t (X) represents the prediction result of the t-th decision tree for the input feature X.
[0097] (5) Model evaluation and prediction: Split the historical data set into a training set and a test set; import the training set into SVR or RFR for training, which is expressed as follows:
[0098] Model Fit: f(X)=SVR / RFR(X,Y)
[0099] Using the trained model for prediction, it is expressed as follows:
[0100]
[0101] Model evaluation by comparing predicted values With the actual value Y test The performance of the evaluation model can be evaluated by the mean square error (MSE) and the root mean square error (RMSE).
[0102] The formulas are expressed as:
[0103]
[0104]
[0105] (6) Model updating can be achieved by adding new data to the data set and then updating the feature matrix X and target variable Y; then the model is trained again to make the model better meet the preset requirements.
[0106] Based on the above, the vehicle driving trajectory monitoring method based on digital twins in this embodiment can also be applied to the operating vehicle monitoring method, and the target vehicle is an operating vehicle.
[0107] Combination Figure 3 As shown, based on the above, this embodiment also provides a vehicle driving trajectory monitoring system based on digital twins, which includes:
[0108] There are multiple image monitoring units, which are arranged within the preset public area and perform image monitoring on the vehicle driving area within the preset public area to generate image monitoring data;
[0109] A vehicle positioning unit is used to respond to a start signal of a target vehicle, monitor the position of the target vehicle according to preset conditions, and generate vehicle positioning data;
[0110] A data association unit is used to determine the driving trajectory of the target vehicle in a preset public area based on the vehicle positioning data and in combination with the image monitoring data, and generate vehicle trajectory data;
[0111] A data storage unit is used to construct a target vehicle database, collect vehicle trajectory data, vehicle positioning data and image monitoring data corresponding to the target vehicle, form a target vehicle data set, and store multiple target vehicle data sets in the target vehicle database;
[0112] Digital twin unit, used to build a digital twin model to map the terrain and building information within the preset public area;
[0113] The vehicle monitoring unit is used to monitor the driving trajectory of the target vehicle in the digital twin model according to the vehicle positioning data or vehicle trajectory data corresponding to the target vehicle.
[0114] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0115] If the integrated unit is implemented in the form of a software functional unit 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 invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0116] The above descriptions are only some embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any equivalent device or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.
Claims
1. A vehicle driving trajectory monitoring method based on digital twins, characterized in that: It includes: Perform image monitoring on the vehicle driving area within the preset public area and generate image monitoring data; In response to the start signal of the target vehicle, the position of the target vehicle is monitored according to preset conditions to generate vehicle positioning data; Based on the vehicle positioning data and combined with the image monitoring data, the driving trajectory of the target vehicle in the preset public area is determined to generate vehicle trajectory data; Construct a target vehicle database, collect vehicle trajectory data, vehicle positioning data and image monitoring data corresponding to the target vehicle to form a target vehicle data set, and store multiple target vehicle data sets in the target vehicle database; Build a digital twin model to map the terrain and building information within the preset public area; According to the vehicle positioning data or vehicle trajectory data corresponding to the target vehicle, the driving trajectory of the target vehicle is monitored in the digital twin model.
2. The vehicle driving trajectory monitoring method based on digital twin according to claim 1 is characterized in that: It also includes: In response to the start signal of the target vehicle, the image monitoring is performed in the target vehicle according to the preset conditions to collect the image viewing angle of the target vehicle in the road direction and generate driving image data in time series; The target vehicle data set also includes driving image data corresponding to the target vehicle.
3. The vehicle driving trajectory monitoring method based on digital twins according to claim 2 is characterized in that: There are multiple target vehicles, and basic information of the multiple target vehicles is collected; The basic information includes: one or more of the vehicle model, appearance specifications, exterior color, and license plate number; The image monitoring data, vehicle positioning data and vehicle trajectory data are all generated in time series.
4. The vehicle driving trajectory monitoring method based on digital twin according to claim 2 or 3, characterized in that: It also includes: Marking the vehicle positioning data of all target vehicle data sets in the target vehicle database in combination with the electronic map to identify the location where the target vehicle appears on the electronic map and the corresponding date and time point, generating marking information, and then associating the marking information with the target vehicle data set; The tag information and / or the license plate number of the target vehicle is used as one of the index information of the target vehicle data set.
5. The vehicle driving trajectory monitoring method based on digital twin according to claim 4 is characterized in that: It also includes: The corresponding target vehicle data set is retrieved from the target vehicle database according to preset conditions, and then the driving trajectory of the target vehicle is inverted in the digital twin model based on the target vehicle data set.
6. The vehicle driving trajectory monitoring method based on digital twins according to claim 5 is characterized in that: It includes: A demonstration location and a demonstration time period are designated within a preset public area, where the demonstration time period is a past time period, and a target vehicle that appears at the demonstration location within the demonstration time period is determined based on the demonstration location and the demonstration time period in combination with the marking information. When the target vehicle is not present, "No target vehicle" is returned as the confirmation result, and then the 3D view of the demonstration location is directly displayed in the digital twin model. When there is more than one target vehicle, the vehicle trajectory data and / or driving image data of the target vehicle appearing at the demonstration location during the demonstration time period are retrieved from the target vehicle data set in the target vehicle database, and then mapped in the digital twin model to invert the driving trajectory of the target vehicle at the specified demonstration location and demonstration time period.
7. The vehicle driving trajectory monitoring method based on digital twin according to claim 5 or 6, characterized in that: When inverting the target vehicle’s driving trajectory in the digital twin model, multi-view data is also provided for display; The multi-view data includes at least one of the following: (1) Invert the target vehicle’s driving trajectory in the digital twin model from a bird’s-eye view; (2) The driving image data is used as the main display data, and the position change of the target vehicle in the digital twin model is used as the auxiliary display data to invert the driving trajectory of the target vehicle.
8. The vehicle driving trajectory monitoring method based on digital twins according to claim 4 is characterized in that: It also includes: Based on the vehicle positioning data and corresponding marking information in multiple target vehicle data sets in the target vehicle database, a deep learning model is constructed or trained to predict the target vehicle's stay time and / or discrete trajectory at a preset location and a preset time. The prediction results of the deep learning model are then demonstrated in the digital twin model.
9. A method for monitoring an operating vehicle, characterized in that: It includes the vehicle driving trajectory monitoring method based on digital twins as described in one of claims 1 to 8, wherein the target vehicle is an operating vehicle.
10. A vehicle driving trajectory monitoring system based on digital twins, characterized in that: It includes: There are multiple image monitoring units, which are arranged within the preset public area and perform image monitoring on the vehicle driving area within the preset public area to generate image monitoring data; A vehicle positioning unit is used to respond to a start signal of a target vehicle, monitor the position of the target vehicle according to preset conditions, and generate vehicle positioning data; A data association unit is used to determine the driving trajectory of the target vehicle in a preset public area based on the vehicle positioning data and in combination with the image monitoring data, and generate vehicle trajectory data; A data storage unit is used to construct a target vehicle database, collect vehicle trajectory data, vehicle positioning data and image monitoring data corresponding to the target vehicle, form a target vehicle data set, and store multiple target vehicle data sets in the target vehicle database; Digital twin unit, used to build a digital twin model to map the terrain and building information within the preset public area; The vehicle monitoring unit is used to monitor the driving trajectory of the target vehicle in the digital twin model according to the vehicle positioning data or vehicle trajectory data corresponding to the target vehicle.