Vehicle trajectory determination method and device, electronic equipment, medium and program product

Abstract

The application discloses a vehicle trajectory determination method and device, an electronic device, a medium and a program product. The method comprises the following steps: acquiring visual identification data corresponding to a vehicle passing through a monitoring point, and an associated mobile terminal satisfying a space-time association condition with the vehicle, and establishing an association relationship between the vehicle and the associated mobile terminal; the space-time association condition comprises: there is position update communication signaling in a preset range within a preset time window corresponding to the monitoring point, and the preset range comprises the monitoring point; determining that a continuous trajectory data stream of the vehicle satisfies a trigger condition of visual failure detection, generating virtual trajectory data corresponding to the vehicle based on communication signaling of the associated mobile terminal; and fusing the virtual trajectory data and visual trajectory data to obtain target vehicle trajectory corresponding to the vehicle. The embodiment of the application can improve the accuracy of vehicle trajectory determination in a complex environment.

Description

Technical Field

[0001] This application relates to the field of vehicle positioning technology, specifically to a vehicle trajectory determination method and device, electronic equipment, medium, and program product. Background Technology

[0002] Existing vehicle trajectory tracking technologies primarily rely on cameras at checkpoints or electronic police systems. However, real-world applications often encounter situations like heavy rain, dense fog, and backlighting that degrade image quality, leading to interruptions in vehicle trajectory tracking. Furthermore, blind spots in surveillance coverage and equipment malfunctions further contribute to the loss of vehicle trajectory data. Therefore, accurately locating and reconstructing vehicle trajectories has become a pressing technical challenge. Summary of the Invention

[0003] In view of the above problems, this application provides a vehicle trajectory determination method and apparatus, electronic device, medium and program product, to at least solve the technical problem of low accuracy of vehicle trajectory determination in complex environments in related technologies.

[0004] According to a first aspect of the embodiments of this application, a vehicle trajectory determination method is provided. The method includes: acquiring visual recognition data corresponding to a vehicle passing through a monitoring point, and an associated mobile terminal that satisfies spatiotemporal association conditions with the vehicle; establishing an association relationship between the vehicle and the associated mobile terminal; the spatiotemporal association conditions include: communication signaling for location updates existing within a preset time window corresponding to the vehicle entering the monitoring point and within a preset range, the preset range including the monitoring point; determining that the continuous trajectory data stream of the vehicle satisfies the triggering conditions for visual failure detection; generating virtual trajectory data corresponding to the vehicle based on the communication signaling of the associated mobile terminal; and fusing the virtual trajectory data with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle.

[0005] According to a second aspect of the embodiments of this application, a vehicle trajectory determination device is provided. The device includes: an acquisition unit, configured to acquire visual recognition data corresponding to a vehicle passing through a monitoring point, and an associated mobile terminal that satisfies spatiotemporal association conditions with the vehicle, and establish an association relationship between the vehicle and the associated mobile terminal; the spatiotemporal association conditions include: communication signaling for location updates existing within a preset time window corresponding to the vehicle entering the monitoring point, and the preset range including the monitoring point; a generation unit, configured to determine that the continuous trajectory data stream of the vehicle satisfies the triggering conditions for visual failure detection, and generate virtual trajectory data corresponding to the vehicle based on the communication signaling of the associated mobile terminal; and a fusion unit, configured to fuse the virtual trajectory data with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle.

[0006] According to a third aspect of the embodiments of this application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the vehicle trajectory determination method of the first aspect described above.

[0007] According to a fourth aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer program, and the computer program is configured to execute the vehicle trajectory determination method of the first aspect described above when running.

[0008] According to a fifth aspect of the embodiments of this application, a computer program product is also provided, including a computer program that is executed by a processor to implement the vehicle trajectory determination method of the first aspect described above.

[0009] In this embodiment, visual recognition data of vehicles passing through monitoring points and associated mobile terminals that meet spatiotemporal association conditions with the vehicles are acquired, and an association relationship between the vehicles and the associated mobile terminals is established. The spatiotemporal association conditions include: communication signaling for location updates within a preset time window when the vehicle enters the monitoring point and within a preset range, wherein the preset range includes the monitoring point; determining that the continuous trajectory data stream of the vehicle meets the triggering conditions for visual failure detection, generating virtual trajectory data corresponding to the vehicle based on the communication signaling of the associated mobile terminals; and fusing the virtual trajectory data with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle. This application can proactively identify visual tracking data interruptions caused by adverse environments such as heavy rain, dense fog, and equipment power outages. It generates virtual trajectory points using continuous mobile terminal signaling data and connects them with previous and subsequent valid visual trajectory nodes to fit a temporally continuous and spatially complete vehicle driving trajectory. This avoids inaccurate vehicle trajectory analysis caused by trajectory interruptions and achieves reliable and continuous vehicle trajectory reconstruction in complex real-world environments. It greatly improves the reliability and usability of vehicle trajectory reconstruction results and provides a high-quality data foundation for subsequent applications such as traffic analysis and event diagnosis. Attached Figure Description

[0010] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiments below. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a schematic diagram of an application environment for an optional vehicle trajectory determination method according to an embodiment of this application; Figure 2 This is a schematic diagram of an application environment for another optional vehicle trajectory determination method according to an embodiment of this application; Figure 3 This is a flowchart illustrating a vehicle trajectory determination method according to an embodiment of this application; Figure 4 This is a flowchart illustrating another vehicle trajectory determination method according to an embodiment of this application; Figure 5 This is a flowchart illustrating another vehicle trajectory determination method according to an embodiment of this application; Figure 6 This is a flowchart illustrating another vehicle trajectory determination method according to an embodiment of this application; Figure 7 This is a flowchart illustrating another vehicle trajectory determination method according to an embodiment of this application; Figure 8 This is a flowchart illustrating another vehicle trajectory determination method according to an embodiment of this application; Figure 9 This is a schematic diagram of the structure of a vehicle trajectory determination device provided in an embodiment of this application; Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0011] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0012] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0013] According to one aspect of the present invention, a vehicle trajectory determination method is provided. Optionally, as an alternative implementation, the above-described vehicle trajectory determination method may be applied to, but is not limited to, [examples of other methods]. Figure 1 The application environment shown includes: a terminal device 102 for human-computer interaction, a network 104, and a server 106. User 108 can interact with terminal device 102, which runs a vehicle trajectory determination application. Terminal device 102 includes a human-computer interaction screen 1022, a processor 1024, and a memory 1026. The human-computer interaction screen 1022 displays vehicle trajectories; the processor 1024 acquires visual recognition data corresponding to vehicles at monitoring points; and the memory 1026 stores the visual recognition data corresponding to vehicles at the monitoring points.

[0014] Furthermore, server 106 includes database 1062 and processing engine 1064. Database 1062 stores visual recognition data corresponding to vehicles at monitoring points. Processing engine 1064 acquires visual recognition data corresponding to vehicles passing through monitoring points, as well as associated mobile terminals that meet spatiotemporal association conditions with the vehicles, and establishes an association relationship between the vehicles and the associated mobile terminals. The spatiotemporal association conditions include: communication signaling for location updates within a preset time window when the vehicle enters the monitoring point, and within a preset range, where the preset range includes the monitoring point; determining that the continuous trajectory data stream of the vehicle meets the triggering conditions for visual failure detection; generating virtual trajectory data corresponding to the vehicle based on the communication signaling of the mobile terminals; and fusing the virtual trajectory data with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle.

[0015] In one or more embodiments, the vehicle trajectory determination method described above in this application can be applied to... Figure 2 The application environment shown. For example... Figure 2 As shown, user 202 and user device 204 can interact. User device 204 includes memory 206 and processor 208. In this embodiment, user device 204 can, but is not limited to, referencing the operations performed by terminal device 102 to obtain the target vehicle trajectory corresponding to the vehicle.

[0016] Optionally, the terminal device 102 and user device 204 mentioned above include, but are not limited to, mobile phones, televisions, tablets, laptops, PCs, in-vehicle electronic devices, wearable devices, and other terminals. The network 104 may include, but is not limited to, wireless networks or wired networks. The wireless network includes Wi-Fi and other networks that enable wireless communication. The wired network may include, but is not limited to, wide area networks (WANs), metropolitan area networks (MANs), and local area networks (LANs). The server 106 may include, but is not limited to, any hardware device capable of computation. The server 106 may be a single server, a server cluster consisting of multiple servers, or a cloud server. The above is merely an example, and no limitations are imposed in this embodiment.

[0017] Existing vehicle trajectory tracking technologies primarily rely on cameras at checkpoints or electronic police systems. However, real-world applications often encounter situations like heavy rain, dense fog, and backlighting that degrade image quality, leading to interruptions in vehicle trajectory tracking. Furthermore, blind spots in surveillance coverage and equipment malfunctions further contribute to the loss of vehicle trajectory data. Therefore, accurately locating and reconstructing vehicle trajectories has become a pressing technical challenge.

[0018] To address the aforementioned technical problems, as an optional implementation method, such as Figure 3 As shown in the figure, this application provides a method for determining vehicle trajectory, the method including the following steps: S302, acquire visual recognition data of vehicles passing through monitoring points, and associated mobile terminals that meet the spatiotemporal association conditions with the vehicles, and establish an association relationship between the vehicles and the associated mobile terminals; the spatiotemporal association conditions include: communication signaling for location updates exists within a preset time window when the vehicle enters the monitoring point and within a preset range, the preset range including the monitoring point. S304, determine that the continuous trajectory data stream of the vehicle meets the triggering conditions for visual failure detection, and generate virtual trajectory data corresponding to the vehicle based on the communication signaling of the mobile terminal; S306, the virtual trajectory data and the visual trajectory data are fused to obtain the target vehicle trajectory corresponding to the vehicle.

[0019] Specifically, in this embodiment, the monitoring point can be a camera on a traffic road or a traffic enforcement camera. When a vehicle passes through a high-definition checkpoint, and its visual recognition confidence (such as the accuracy of license plate recognition) exceeds a preset threshold (such as 0.9), a dynamic "vehicle-mobile terminal" anchoring process is triggered. The mobile terminal can be a mobile phone. When a mobile phone and a vehicle's trajectory are matched successfully multiple times with high confidence, a stable "vehicle-mobile phone" association model is established. For example, the system acquires signaling data (such as location updates, handover events, etc.) of all mobile phones within the geographical range of the monitoring point within a time window (such as 2 minutes) before and after the vehicle's passage time. By calculating the matching degree between the vehicle's passage time and location and the mobile phone's signaling time and location, the most likely candidate associated mobile phones are selected. In one example, this application can also generate and dynamically update a dynamic association mapping table with the structure {vehicle ID, mobile phone IMSI hash value, association confidence}. The aforementioned associated mobile terminal refers to a terminal device with mobile communication function. This terminal device may include a television, tablet computer, laptop computer, PC, wearable device, or vehicle terminal with mobile communication function, etc. The above is only an example, and no limitation is made in this embodiment.

[0020] Based on the road network topology, if a vehicle enters road segment A and is not captured in downstream road segment B within a reasonable maximum travel time, this is considered a trigger condition for visual failure detection. Additionally, directly receiving alarm information indicating offline status or low image quality (e.g., due to heavy rain or dense fog) from downstream cameras can also be considered a trigger condition for visual failure detection. This application pre-establishes a database of precise mappings between base station sector handover boundaries and road station numbers through historical data mining. Utilizing cellular network control signaling (such as handover events and location area update events), when a mobile phone bound to a vehicle experiences such signaling events, the system maps them to virtual trajectory data at a specific road station number (e.g., K120+500) and a precise time point.

[0021] In the normal visual area, high-precision image positioning data is used; when vision fails, continuous mobile phone signaling data is used for interpolation and completion. By assigning different confidence weights to visual image data and non-line-of-sight perception signal data, and performing weighted fusion, an optimal continuous trajectory is fitted and generated. Specifically, by suppressing the inherent noise of the signaling data, discrete visual capture points and virtual trajectory points are fused into a temporally continuous and spatially smooth vehicle trajectory.

[0022] In this embodiment, visual recognition data of vehicles passing through monitoring points and associated mobile terminals that meet spatiotemporal association conditions with the vehicles are acquired, and an association relationship between the vehicles and the associated mobile terminals is established. The spatiotemporal association conditions include: communication signaling for location updates within a preset time window when the vehicle enters the monitoring point and within a preset range, where the preset range includes the monitoring point; determining that the continuous trajectory data stream of the vehicle meets the triggering conditions for visual failure detection, generating virtual trajectory data corresponding to the vehicle based on the communication signaling of the mobile terminals; and fusing the virtual trajectory data with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle. This application can proactively identify visual tracking data interruptions caused by adverse environments such as heavy rain, dense fog, and equipment power outages. It generates virtual trajectory points using continuous mobile terminal signaling data and connects them with previous and subsequent valid visual trajectory nodes to fit a temporally continuous and spatially complete vehicle driving trajectory. This avoids inaccurate vehicle trajectory analysis caused by trajectory interruptions and achieves reliable and continuous vehicle trajectory reconstruction in complex real-world environments. It greatly improves the reliability and usability of vehicle trajectory reconstruction results and provides a high-quality data foundation for subsequent applications such as traffic analysis and event diagnosis.

[0023] In one or more embodiments, the visual recognition data includes vehicle attribute information, and the acquisition of visual recognition data corresponding to vehicles passing through monitoring points, as well as associated mobile terminals that meet spatiotemporal correlation conditions with the vehicles, includes: When the vehicle passes the monitoring point, the visual recognition data corresponding to the vehicle is acquired; If the visual recognition confidence level of the visual recognition data exceeds the first threshold, then the vehicle is determined to meet the dynamic accompaniment anchoring condition. The vehicle's passage time is matched with the update time of the communication signaling of each mobile terminal within the preset range, and the vehicle's location information is matched with the location information of the mobile terminal to filter out at least one associated mobile terminal that meets the spatiotemporal association condition.

[0024] Specifically, in this embodiment, when a vehicle passes through a preset checkpoint monitoring point, its visual recognition data is acquired. If the confidence level of the vehicle's visual recognition exceeds a first threshold, a dynamic accompaniment anchoring condition is triggered, and a spatiotemporal collision association algorithm is initiated. The range of the first threshold can be, for example, [0.85, 0.9]. Signaling data of all mobile terminals whose locations are updated within a preset geographical area of ​​the checkpoint monitoring point are acquired within a preset time window before and after the vehicle passes through the checkpoint monitoring point. Based on the spatiotemporal collision association algorithm, the vehicle's passage time and location information are matched and calculated with the update time and location information of each mobile terminal's signaling to filter out at least one candidate associated mobile terminal.

[0025] In one or more embodiments, establishing the association between the vehicle and the associated mobile terminal includes: Obtain the vehicle identifier and the terminal identifier of the associated mobile terminal from the visual recognition data; Based on the frequency of trajectory matching and spatiotemporal consistency between the vehicle and the associated mobile terminal, a confidence score is obtained for the credibility of the association between the associated mobile terminal and the vehicle. Based on the vehicle identifier, terminal identifier, and confidence score, an association mapping table is created between the vehicle and the associated mobile terminal.

[0026] Specifically, in this embodiment of the application, the vehicle's passage time and location information are first matched with the update time and location information of the signaling of each associated mobile terminal based on the spatiotemporal collision association algorithm to filter out at least one candidate associated mobile terminal. Then, a dynamic association mapping table is generated and maintained according to the calculation results of the association matching step.

[0027] In one embodiment, the dynamic association mapping table includes at least a vehicle identifier (Vehicle_ID), a hashed International Mobile Subscriber Identity (IMSI) code (IMSI_Hash), and a confidence score (Confidence_Score) that characterizes the credibility of the association between the vehicle and the mobile terminal. The structure of the dynamic association mapping table can be represented as {Vehicle_ID, IMSI_Hash, Confidence_Score}.

[0028] In one example, the structure of the above association mapping table can be as shown in Table 1: Table 1

[0029] The maintenance mechanism for the aforementioned association mapping table includes: 1. Addition: Insert when the probability of a new association is higher than the threshold.

[0030] 2. Update: Increase the confidence level when the same vehicle-phone pair is successfully linked again.

[0031] 3. Decay: If no co-occurrence occurs for a long period of time, the confidence level will automatically decay over time.

[0032] 4. Cleanup: Remove a data point from the mapping table when the confidence level is below the failure threshold.

[0033] In one or more embodiments, the triggering conditions for the visual failure detection include at least one of: a visual trajectory breakage condition and a visual sensing device abnormality condition; The visual trajectory breakage condition includes: after the vehicle enters the target road segment, it is not detected by the downstream visual perception device of the target road segment within a preset time period; Abnormal conditions for visual perception devices include: the visual perception device that identifies the vehicle is offline, or the image quality score acquired by the visual perception device is lower than the confidence threshold.

[0034] Specifically, in this embodiment of the application, the continuous trajectory data stream of the target vehicle is continuously monitored, and it is determined whether a trigger condition for visual failure detection occurs; the trigger condition includes: 1. Visual Trajectory Logic Break: Based on the preset road network topology, when a target vehicle is detected entering a specific road segment A, it is not captured or appears within the visual monitoring range of its downstream adjacent road segment B within the preset maximum tolerable passage time threshold T_max. The value of T_max can be set according to actual needs.

[0035] 2. Abnormal alarm of sensing device: Receive and process the operating status information of downstream visual sensing devices. When it is determined that the downstream visual sensing device is offline or the quality score of the image it collects is lower than the preset confidence threshold, an abnormal state is triggered.

[0036] In one example, when any of the above-mentioned abnormal triggering conditions are detected, the system automatically triggers a safety switching mechanism, switching the trajectory tracking and reconstruction mode of the target vehicle from the "active fusion mode," which mainly relies on visual data, to the shadow mode. In the shadow mode, the system still runs trajectory estimation algorithms based on non-visual data sources such as mobile phone signaling in parallel and generates a predicted trajectory. The predicted trajectory is not directly used as trajectory output or for executing critical control commands, but is mainly used for internal evaluation, alarms, or as auxiliary reference information. The system continuously monitors the recovery of visual perception conditions, and automatically switches back to the "active fusion mode" when preset recovery conditions are met.

[0037] In one or more embodiments, generating virtual trajectory data corresponding to the vehicle based on the communication signaling of the mobile terminal includes: Obtain a pre-built database of mapping relationships between the spatial locations of communication network base stations and the physical locations of roads; The communication signaling data stream corresponding to the associated mobile terminal is tracked, and network events representing location changes of the associated mobile terminal are captured from the communication signaling data stream; the network events include handover events or location area update events; Based on the network events and the mapping database, virtual trajectory data corresponding to the vehicle is generated.

[0038] Specifically, in this embodiment, when visual perception data is unavailable or missing, auxiliary virtual vehicle trajectory data is generated using communication signaling from the associated mobile terminal. First, a mapping database is pre-constructed that associates the spatial location of communication network infrastructure (e.g., communication base stations) with the physical location of roads. In one example, based on geographic mapping and historical data analysis, a correspondence table is established between signal handover boundaries between cellular network base station sectors and the location of highway / highway chainages. For example, a mapping relationship is established: the typical handover boundary between base station sectors S_1 and S_2 corresponds to the geographical area near highway chainage K120+500.

[0039] The signaling data stream of associated mobile terminals that have established a connection with the target vehicle is tracked in real time. From the signaling data stream, specific network events characterizing significant location changes of the associated mobile terminals are detected and captured in real time. These specific network events include at least: handover events and location area update events; wherein, a handover event refers to a handover event where the mobile terminal switches service connections between different cellular base station sectors. A location area update event refers to a registration update event initiated by the mobile terminal when moving between different tracking areas.

[0040] In response to the capture of any of the aforementioned network events, a virtual trajectory generation algorithm is executed. This algorithm, based on the aforementioned mapping database, converts the captured network event (including event type Handover_Event or update zone TAU, and event occurrence time T_signal) into a virtual vehicle passage record (virtual trajectory data) corresponding to the target vehicle. In one example, the generated virtual vehicle passage record data structure includes at least the following fields: Plate: The associated target vehicle identifier (such as license plate number).

[0041] Loc: The road station location (e.g., K120+500) that best matches the network event, as determined by the mapping database.

[0042] Time: The timestamp of the network event, T_signal.

[0043] Type: Record type identifier, marked as Virtual, to distinguish it from the real records generated by visual recognition (HD checkpoints / electronic police).

[0044] In one or more embodiments, generating virtual trajectory data corresponding to the vehicle based on the network events and the mapping relationship database includes: Obtain the handover boundary or tracking area boundary of the communication network base station corresponding to the network event, as well as the time of occurrence of the network event; Based on the mapping relationship database, determine the road information corresponding to the switching boundary or tracking area boundary; Based on the road information and the time of occurrence, virtual trajectory data corresponding to the vehicle is generated.

[0045] Specifically, in this embodiment, when a handover event or location area update event occurs on the associated mobile terminal bound to the vehicle, the network generates corresponding communication signaling. Based on this signaling, the handover boundary or tracking area boundary of the communication network base station can be obtained. According to the mapping database, the road information corresponding to the handover boundary or tracking area boundary is determined. Upon receiving the network event, the system immediately queries this mapping database to find the road location information that best matches the event identifier. In this way, a communication topology concept is directly converted into a precise road marker or geographical coordinate. For example, the handover boundary between base station sectors S1 and S2 has a mapping relationship with highway marker K120+500. Finally, the system combines the matched road information and the time of the event to generate a standardized virtual trajectory record. This record may contain fields such as vehicle identifier, location, time, and trajectory type.

[0046] In one or more embodiments, fusing the virtual trajectory data with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle includes: According to the timestamp order corresponding to the virtual trajectory data, the virtual trajectory data is inserted into the original trajectory sequence formed by the visual trajectory data to construct a multi-source heterogeneous trajectory node sequence; A trajectory smoothing model with the vehicle's motion state as the state vector is constructed. Based on the trajectory smoothing model, an iterative operation is performed on the multi-source heterogeneous trajectory node sequence for prediction and correction to obtain the target vehicle trajectory that satisfies the spatiotemporal continuity condition.

[0047] Specifically, in this embodiment of the application, the generated virtual trajectory record is inserted sequentially into the original trajectory sequence formed by visual perception data (such as high-definition checkpoint capture records) according to its timestamp, thus forming a multi-source heterogeneous trajectory node sequence containing real nodes and virtual nodes.

[0048] A Kalman filter is used to construct a trajectory smoothing model with the vehicle's motion state (including position and velocity) as the state vector. Based on this trajectory smoothing model, iterative prediction and correction processing is performed on the multi-source heterogeneous trajectory node sequence. Prediction phase: Based on the vehicle's optimal estimated state at the previous moment, combined with a preset motion model (such as a uniform speed or uniform acceleration model), predict its state at the current moment.

[0049] Correction Phase: When a new trajectory node (whether a real visual node or a virtual node) is received, it is input as an observation into the trajectory smoothing model. The trajectory smoothing model performs weighted correction on the predicted state based on the confidence level of the observation itself (assigning different observation noise covariances to visual nodes and virtual nodes, for example, virtual nodes have higher initial uncertainty), to obtain the optimal state estimate for the current time step.

[0050] Eliminating signaling drift: The core function of this process is to use the physical continuity constraint of vehicle movement itself to smooth and filter the inherent time lag and spatial drift (such as the "sawtooth" positioning jump caused by the base station handover boundary effect) in the virtual nodes from signaling, thereby suppressing multi-source noise.

[0051] After iterative processing by the trajectory smoothing model, a complete vehicle trajectory that remains continuous and smooth in both time and space is finally output. In sections where visual data is available, the trajectory is dominated by visual data; in sections where visual data is missing, smoothed virtual nodes effectively complete and bridge the gaps.

[0052] In one application embodiment, this application also provides a vehicle trajectory determination method, including: 1. Dynamic Anchoring Steps: Establish a high-confidence temporary binding between "vehicle and mobile terminal" in the visually normal zone. The mobile terminal can be a mobile phone. When the trajectory of a mobile phone and a vehicle are matched successfully multiple times with high confidence, a stable association model between "vehicle and mobile phone" is established.

[0053] 2. Visual failure detection and mode switching steps: When visual failure occurs, automatically switch to signaling tracing; after visual recovery, automatically revert and verify.

[0054] 3. Virtual trajectory generation steps based on mobile communication signaling handover: Utilize the boundary effect of base station sector handover to map and generate "virtual capture points" on the highway.

[0055] 4. Multimodal trajectory data fusion steps based on prediction correction: In the normal visual area, high-precision image positioning data is used; when vision fails, continuous mobile phone signaling data is used for interpolation and completion. By assigning different confidence weights to visual image data and non-line-of-sight perception signal data, and performing weighted fusion, an optimal continuous trajectory is fitted and generated.

[0056] Specifically, in one embodiment, combined with Figure 4 As shown, step 1 above includes: 1.1 Triggering steps: When a vehicle passes through a preset high-definition checkpoint monitoring point, its visual recognition data is acquired; if the confidence level of the visual recognition of the vehicle exceeds the first threshold (preferably 0.9), the dynamic accompanying anchoring condition is triggered, and the spatiotemporal collision association algorithm is started. 1.2 Association Matching Step: Obtain the signaling data of all mobile terminals whose locations are updated within a preset geographical range of the high-definition checkpoint within a preset time window before and after the vehicle passes through the high-definition checkpoint; based on the spatiotemporal collision association algorithm, match and calculate the vehicle's passage time and location information with the update time and location information of each mobile terminal signaling, and filter out at least one candidate associated mobile terminal.

[0057] 1.3 Mapping Relationship Construction and Maintenance Steps: Based on the calculation results of the association matching step, a dynamic association mapping table is generated and maintained; the mapping table includes at least the vehicle identifier Vehicle_ID, the hashed International Mobile Subscriber Identity (IMSI) code of the mobile terminal IMSI_Hash, and the confidence score item Confidence_Score, which represents the credibility of the association between the vehicle and the mobile terminal, and its structure is represented as {Vehicle_ID, IMSI_Hash, Confidence_Score}.

[0058] In one embodiment, combined with Figure 5 As shown, step 2 above includes: Continuously monitor the continuous trajectory data stream of the target vehicle and determine whether a preset abnormal trigger condition occurs; the abnormal trigger condition includes: 1. Visual trajectory logic break: According to the preset road network topology, when a target vehicle is detected to enter a specific road segment A, it is not captured or appears within the visual monitoring range of its downstream adjacent road segment B within the preset maximum tolerable passage time threshold T_{max}.

[0059] II. Sensing Device Anomaly Alarm: Receives and processes the operating status information of downstream visual sensing devices. When it is determined that a downstream device is offline, or the quality score of its acquired images is lower than a preset confidence threshold, an anomaly is triggered. When any of the above anomaly triggering conditions are detected, the system automatically triggers a safety switching mechanism, switching the trajectory tracking and reconstruction mode of the target vehicle from the "active fusion mode" that mainly relies on visual data to the shadow mode. In the shadow mode: the system still runs trajectory estimation algorithms based on non-visual data sources such as mobile phone signaling in parallel and generates inferred trajectories. These inferred trajectories are not directly used as official trajectory outputs or for executing critical control commands, but are primarily used for internal evaluation, alarms, or as auxiliary reference information. The system continuously monitors the recovery of visual perception conditions, and automatically switches back to "active fusion mode" when preset recovery conditions are met.

[0060] In one embodiment, combined with Figure 6 and Figure 7 As shown, step 3 above includes: 3.1 Spatial Mapping Database Construction Steps: A mapping database linking the spatial locations of communication network infrastructure with the physical locations of roads is pre-constructed. Specifically, based on geographic surveying and historical data analysis, a correspondence table is established between signal handover boundaries between cellular network base station sectors and the location of highway / highway chainages. For example, the mapping relationship is established as follows: the typical handover boundary between base station sectors S_1 and S_2 corresponds to the geographical area near highway chainage K120+500.

[0061] 3.2 Real-time Mobile Network Event Capture Steps: Track the signaling data stream of mobile terminals already associated with the target vehicle (associated through the aforementioned step 1. Dynamic Accompaniment Anchoring). From the signaling data stream, detect and capture specific network events characterizing significant location changes of the mobile terminal in real time. These specific network events include at least: Handover events: events where the mobile terminal switches service connections between different cellular base station sectors. Location area update events: registration update events initiated by the mobile terminal when moving between different tracking areas.

[0062] 3.3 Virtual Trajectory Record Generation Step: In response to the capture of any network event described in step 3.2, the virtual trajectory generation algorithm is executed. This algorithm, based on the mapping database established in step 3.1, converts the captured network event (including event type Handover_Event or TAU, and event occurrence time T_signal) into a virtual vehicle passage record corresponding to the target vehicle.

[0063] The generated virtual vehicle passage record data structure should contain at least the following fields: Plate: The associated target vehicle identifier (such as license plate number).

[0064] Loc: The road station location (e.g., K120+500) that best matches the network event, as determined by the mapping database.

[0065] Time: The timestamp of the network event, T_signal.

[0066] Type: Record type identifier, marked as Virtual to distinguish it from the real records generated by visual recognition (HD checkpoints / electronic police).

[0067] In one embodiment, combined with Figure 8 As shown, step 4 above includes: 4.1. Heterogeneous trajectory node fusion step: The virtual trajectory records generated in step 3 above are inserted into the original trajectory sequence formed by visual perception data (such as high-definition checkpoint capture records) in sequence according to their timestamps, to form a multi-source heterogeneous trajectory node sequence containing real nodes and virtual nodes.

[0068] 4.2 Spatiotemporal Trajectory Smoothing Optimization Steps: A Kalman filter is used to construct a trajectory smoothing model with the vehicle's motion state (including position and velocity) as the state vector. Based on this model, iterative prediction and correction processing is performed on the multi-source heterogeneous trajectory node sequence: Prediction phase: Based on the vehicle's optimal estimated state at the previous moment, combined with a preset motion model (such as a uniform speed or uniform acceleration model), predict its state at the current moment.

[0069] Correction Phase: When a new trajectory node (whether a real visual node or a virtual node) is received, it is input into the model as an observation. The model performs weighted correction on the predicted state based on the reliability of the observation itself (assigning different observation noise covariances to visual and virtual nodes, for example, virtual nodes have higher initial uncertainty), to obtain the optimal state estimate for the current moment. Signaling Drift Elimination: The core function of this process is to utilize the physical continuity constraint of vehicle motion itself to smooth and filter the inherent time lag and spatial drift (such as the "sawtooth" positioning jumps caused by base station handover boundary effects) in virtual nodes from signaling, thereby suppressing multi-source noise.

[0070] 4.3 Continuous Trajectory Output Steps: After iterative processing by the trajectory smoothing model, a complete vehicle trajectory that remains continuous and smooth in both time and space is finally output. In sections where visual data is available, the trajectory is dominated by visual data; in sections where visual data is missing, smoothed virtual nodes effectively complete and bridge the gaps.

[0071] Compared with the prior art, this application also discloses a method and system for highway vehicle trajectory continuity repair based on non-line-of-sight perception signal assistance. First, the method uses high-confidence visual perception events as anchor points to drive real-time correlation analysis of multi-source heterogeneous data, realizing dynamic and accurate binding of vehicle and mobile terminal identities in the spatiotemporal dimension, providing a reliable data association foundation for subsequent trajectory fusion and restoration.

[0072] Secondly, by using both logical rules and device status checks, automated and highly reliable detection of visual data failures is achieved. By introducing a "shadow mode" as a safety buffer, the system ensures that even when the dominant visual data source is interrupted or its quality degrades, it can still maintain continuous and secure tracking of the target, avoiding trajectory loss or misjudgment due to data deficiency, thus improving the overall robustness and reliability of the system.

[0073] Furthermore, this method creatively utilizes inherent signaling events (handover, TAU) in cellular networks that are strongly correlated with user movement as "virtual capture points" to generate vehicle passage records with clear spatiotemporal markers in visual blind spots or failure zones. Through pre-mapping with road network station numbers, discrete communication events are transformed into accurate location points on continuous roads, providing crucial data supplementation for maintaining trajectory continuity and reliability when visually dominant data is interrupted. This method effectively overcomes the excessive reliance on a single data source in traditional schemes and significantly improves the resilience of the vehicle trajectory reconstruction system in complex environments.

[0074] The key innovation of this application lies in its approach: instead of simply splicing virtual nodes with visual nodes, it intelligently fuses and optimizes heterogeneous data through a filtering framework based on a physical motion model. This method achieves data complementarity: visual data provides high-precision absolute spatial anchor points, while virtual nodes provide continuous relative movement cues. It overcomes the limitations of heterogeneous data: through the inherent smoothness of the model and configurable confidence weights, it effectively suppresses the "sawtooth effect" and random drift error in signaling data. It results in a unified output: ultimately generating a single, smooth trajectory with clear physical meaning that conforms to the laws of vehicle motion, greatly improving the reliability and usability of trajectory reconstruction results and laying a high-quality data foundation for subsequent applications such as traffic analysis and event diagnosis.

[0075] According to another aspect of the embodiments of this application, such as Figure 9 As shown, a vehicle trajectory determination device is also provided, characterized in that the device comprises: The acquisition unit is used to acquire visual recognition data of vehicles passing through the monitoring point, as well as associated mobile terminals that meet the spatiotemporal association conditions with the vehicles, and to establish an association relationship between the vehicles and the associated mobile terminals; the spatiotemporal association conditions include: communication signaling for location updates exists within a preset time window corresponding to the vehicle entering the monitoring point and within a preset range, the preset range including the monitoring point. The generation unit is used to determine that the continuous trajectory data stream of the vehicle meets the triggering conditions for visual failure detection, and to generate virtual trajectory data corresponding to the vehicle based on the communication signaling of the associated mobile terminal. The fusion unit is used to fuse the virtual trajectory data with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle.

[0076] In this embodiment, visual recognition data of vehicles passing through monitoring points and associated mobile terminals that meet spatiotemporal association conditions with the vehicles are acquired, and an association relationship between the vehicles and the associated mobile terminals is established. The spatiotemporal association conditions include: communication signaling for location updates within a preset time window when the vehicle enters the monitoring point and within a preset range, where the preset range includes the monitoring point; determining that the continuous trajectory data stream of the vehicle meets the triggering conditions for visual failure detection, generating virtual trajectory data corresponding to the vehicle based on the communication signaling of the mobile terminals; and fusing the virtual trajectory data with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle. This application can proactively identify visual tracking data interruptions caused by adverse environments such as heavy rain, dense fog, and equipment power outages. It generates virtual trajectory points using continuous mobile terminal signaling data and connects them with previous and subsequent valid visual trajectory nodes to fit a temporally continuous and spatially complete vehicle driving trajectory. This avoids inaccurate vehicle trajectory analysis caused by trajectory interruptions and achieves reliable and continuous vehicle trajectory reconstruction in complex real-world environments. It greatly improves the reliability and usability of vehicle trajectory reconstruction results and provides a high-quality data foundation for subsequent applications such as traffic analysis and event diagnosis.

[0077] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described vehicle trajectory determination method is also provided. This electronic device may be... Figure 1 The terminal device or server shown. This embodiment uses the electronic device as a server as an example for illustration. Figure 10 As shown, the electronic device includes a memory 1002 and a processor 1004. The memory 1002 stores a computer program, and the processor 1004 is configured to execute the steps of any of the above method embodiments via the computer program.

[0078] like Figure 10 As shown, the electronic device includes a memory 1002 and a processor 1004. The memory 1002 stores a computer program, and the processor 1004 is configured to execute the steps of any of the above method embodiments via the computer program.

[0079] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program: S1, acquire visual recognition data of vehicles passing through monitoring points, and associated mobile terminals that meet the spatiotemporal association conditions with the vehicles, and establish an association relationship between the vehicles and the associated mobile terminals; the spatiotemporal association conditions include: communication signaling for location updates exists within a preset time window when the vehicle enters the monitoring point and within a preset range, the preset range including the monitoring point. S2, determine that the continuous trajectory data stream of the vehicle meets the triggering conditions for visual failure detection, and generate virtual trajectory data corresponding to the vehicle based on the communication signaling of the associated mobile terminal; S3, the virtual trajectory data and the visual trajectory data are fused to obtain the target vehicle trajectory corresponding to the vehicle.

[0080] Alternatively, as those skilled in the art will understand, Figure 10 The structure shown is for illustrative purposes only. Figure 10 This does not limit the structure of the aforementioned electronic devices. For example, the electronic device may also include components that are more... Figure 10 The more or fewer components shown (such as network interfaces, etc.), or having the same Figure 10 The different configurations shown.

[0081] The memory 1002 can be used to store software programs and modules, such as the program instructions / modules corresponding to the vehicle trajectory determination method and device in this embodiment. The processor 1004 executes various functional applications and data processing by running the software programs and modules stored in the memory 1002, thereby realizing the aforementioned vehicle trajectory determination method. The memory 1002 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1002 may further include memory remotely located relative to the processor 1004, and these remote memories can be connected to terminal devices via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Specifically, the memory 1002 may be used, but is not limited to, for storing vehicle trajectory data, etc., which will not be elaborated further in this example.

[0082] Optionally, the transmission device 1006 described above is used to receive or send data via a network. Specific examples of the network described above may include wired networks and wireless networks. In one example, the transmission device 1006 includes a Network Interface Controller (NIC), which can be connected to other network devices and a router via a network cable to communicate with the Internet or a local area network. In another example, the transmission device 1006 is a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0083] In addition, the aforementioned electronic device also includes: a display 1008 for displaying vehicle trajectory; and a connection bus 1010 for connecting the various module components in the aforementioned electronic device.

[0084] In one or more embodiments, this application also provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the vehicle trajectory determination method described above. The computer program is configured to execute the steps of any of the method embodiments described above when running.

[0085] Optionally, in this embodiment, the computer-readable storage medium described above may be configured to store a computer program for performing the following steps: S1, acquire visual recognition data of vehicles passing through monitoring points, and associated mobile terminals that meet the spatiotemporal association conditions with the vehicles, and establish an association relationship between the vehicles and the associated mobile terminals; the spatiotemporal association conditions include: communication signaling for location updates exists within a preset time window when the vehicle enters the monitoring point and within a preset range, the preset range including the monitoring point. S2, determine that the continuous trajectory data stream of the vehicle meets the triggering conditions for visual failure detection, and generate virtual trajectory data corresponding to the vehicle based on the communication signaling of the associated mobile terminal; S3, the virtual trajectory data and the visual trajectory data are fused to obtain the target vehicle trajectory corresponding to the vehicle.

[0086] Optionally, in this embodiment, those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0087] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0088] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned 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. This computer software product is stored in a storage medium and includes several instructions to cause one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention.

[0089] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0090] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0091] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for determining vehicle trajectory, characterized in that, The method includes: The system acquires visual recognition data of vehicles passing through monitoring points, as well as associated mobile terminals that meet spatiotemporal association conditions with the vehicles, and establishes an association relationship between the vehicles and the associated mobile terminals. The spatiotemporal association conditions include: communication signaling for location updates within a preset time window when the vehicle enters the monitoring point and within a preset range, where the preset range includes the monitoring point. The continuous trajectory data stream of the vehicle is determined to meet the triggering conditions for visual failure detection, and virtual trajectory data corresponding to the vehicle is generated based on the communication signaling of the associated mobile terminal. The virtual trajectory data is fused with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle.

2. The method according to claim 1, characterized in that, The visual recognition data includes vehicle attribute information. The acquisition of visual recognition data corresponding to vehicles passing through monitoring points, and the associated mobile terminal that meets the spatiotemporal correlation conditions with the vehicle, includes: When the vehicle passes the monitoring point, the visual recognition data corresponding to the vehicle is acquired; If the visual recognition confidence level of the visual recognition data exceeds the first threshold, then the vehicle is determined to meet the dynamic accompaniment anchoring condition. The vehicle's passage time is matched with the update time of the communication signaling of each mobile terminal within the preset range, and the vehicle's location information is matched with the location information of the mobile terminal to filter out at least one associated mobile terminal that meets the spatiotemporal association condition.

3. The method according to claim 2, characterized in that, Establishing the association between the vehicle and the associated mobile terminal includes: Obtain the vehicle identifier and the terminal identifier of the associated mobile terminal from the visual recognition data; Based on the frequency of trajectory matching and spatiotemporal consistency between the vehicle and the associated mobile terminal, a confidence score is obtained for the credibility of the association between the associated mobile terminal and the vehicle. Based on the vehicle identifier, terminal identifier, and confidence score, an association mapping table is created between the vehicle and the associated mobile terminal.

4. The method according to claim 1 or 2, characterized in that, The triggering conditions for visual failure detection include at least one of the following: visual trajectory breakage condition and visual sensing device abnormality condition; The visual trajectory breakage condition includes: after the vehicle enters the target road segment, it is not detected by the downstream visual perception device of the target road segment within a preset time period; Abnormal conditions for visual perception devices include: the visual perception device that identifies the vehicle is offline, or the image quality score acquired by the visual perception device is lower than the confidence threshold.

5. The method according to claim 1, characterized in that, The generation of virtual trajectory data corresponding to the vehicle based on the communication signaling of the associated mobile terminal includes: Obtain a pre-built database of mapping relationships between the spatial locations of communication network base stations and the physical locations of roads; The communication signaling data stream corresponding to the associated mobile terminal is tracked, and network events representing location changes of the associated mobile terminal are captured from the communication signaling data stream; the network events include handover events or location area update events; Based on the network events and the mapping database, virtual trajectory data corresponding to the vehicle is generated.

6. The method according to claim 5, characterized in that, The step of generating virtual trajectory data corresponding to the vehicle based on the network events and the mapping relationship database includes: Obtain the handover boundary or tracking area boundary of the communication network base station corresponding to the network event, as well as the time of occurrence of the network event; Based on the mapping relationship database, determine the road information corresponding to the switching boundary or tracking area boundary; Based on the road information and the time of occurrence, virtual trajectory data corresponding to the vehicle is generated.

7. The method according to claim 1, characterized in that, The step of fusing the virtual trajectory data with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle includes: According to the timestamp order corresponding to the virtual trajectory data, the virtual trajectory data is inserted into the original trajectory sequence formed by the visual trajectory data to construct a multi-source heterogeneous trajectory node sequence; A trajectory smoothing model with the vehicle's motion state as the state vector is constructed. Based on the trajectory smoothing model, an iterative operation is performed on the multi-source heterogeneous trajectory node sequence for prediction and correction to obtain the target vehicle trajectory that satisfies the spatiotemporal continuity condition.

8. A vehicle trajectory determination device, characterized in that, The device includes: The acquisition unit is used to acquire visual recognition data of vehicles passing through the monitoring point, as well as associated mobile terminals that meet the spatiotemporal association conditions with the vehicles, and to establish an association relationship between the vehicles and the associated mobile terminals; the spatiotemporal association conditions include: communication signaling for location updates exists within a preset time window corresponding to the vehicle entering the monitoring point and within a preset range, the preset range including the monitoring point. The generation unit is used to determine that the continuous trajectory data stream of the vehicle meets the triggering conditions for visual failure detection, and to generate virtual trajectory data corresponding to the vehicle based on the communication signaling of the associated mobile terminal. The fusion unit is used to fuse the virtual trajectory data with the visual trajectory data to obtain the target vehicle trajectory corresponding to the vehicle.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by a processor to implement the method as described in any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, The computer program is executed by a processor to implement the method according to any one of claims 1 to 7.

Images