Visual management method and system for bus travel in combination with positioning technology

Abstract

The embodiment of the invention discloses a bus travel visual management method and system combined with a positioning technology, and the method comprises the steps: obtaining the vehicle real-time positioning data of a target vehicle, extracting the task execution information associated with the vehicle real-time positioning data, carrying out the dynamic track fitting of the vehicle real-time positioning data, and obtaining the task execution information of the target vehicle; generating a driving track map matched with the task execution information, performing multi-dimensional analysis on the driving track map based on a preset task rule, and determining a driving state feature and a task completion index of the target vehicle; performing space-time correlation mapping on the driving state characteristics, the task completion index and the driving track map to generate a visual dynamic management interface; and in a running state of the visual dynamic management interface, in response to a tracing instruction triggered by the management terminal, calling the historical track database to carry out backtracking optimization processing on the driving track map, and outputting management feedback data containing an abnormal event label and a path comparison result.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of data visualization technology, and specifically to a bus travel visualization management method and system combined with positioning technology. Background Art

[0002] In the current official vehicle dispatch management system, vehicle monitoring is usually achieved by relying on the basic GPS positioning module and the preset route planning algorithm. For example, the existing technology mostly adopts the discrete positioning data collection method, records the vehicle position coordinates at fixed time intervals, and combines the electronic fence technology to determine whether it deviates from the established route. In terms of task execution monitoring, it mainly adopts a simple matching mechanism between manually inputting task plans and vehicles arriving at key stations, and evaluates the task progress through time comparison.

[0003] However, the existing static path planning algorithm is difficult to adapt to the trajectory deviation caused by real-time road conditions, resulting in insufficient trajectory fitting accuracy; secondly, the correlation between traditional positioning data and task execution parameters is weak, and it cannot accurately reflect the dynamic relationship between driving behavior and task requirements; thirdly, vehicle status evaluation is mostly limited to single-dimensional indicators such as speed and position, and lacks a multi-dimensional evaluation system for comprehensive task elements; in addition, the visualization interface is usually presented as a static route coverage map, lacking dynamic correlation display in the time and space dimensions, resulting in low efficiency in tracing abnormal events; historical data analysis still relies on manual retrieval of discrete positioning records for comparison, making it difficult to achieve automated abnormal pattern recognition. Therefore, how to ensure controllability and explainability when realizing visual management of official vehicles is a long-standing technical problem in existing technologies. Summary of the invention

[0004] The embodiments of the present invention provide a method and system for visual management of public bus travel combined with positioning technology, which can significantly improve the controllability and explainability of official vehicles during task execution when realizing visual management of official vehicles, and provide dynamic optimization decision support for scenarios such as intelligent scheduling, abnormal diagnosis and path planning.

[0005] In a first aspect, an embodiment of the present invention provides a method for visual management of bus travel in combination with positioning technology, which is applied to a visual management system, the method comprising: obtaining real-time vehicle positioning data of a target vehicle, and extracting task execution information associated with the real-time vehicle positioning data, the real-time vehicle positioning data comprising positioning coordinates and motion parameters of a continuous time series; performing dynamic trajectory fitting processing on the real-time vehicle positioning data to generate a driving trajectory map matching the task execution information, the driving trajectory map comprising timestamps of path nodes and key event markers; performing multi-dimensional analysis on the driving trajectory map based on preset task rules to determine driving state characteristics and task completion indicators of the target vehicle; performing spatiotemporal correlation mapping of the driving state characteristics, the task completion indicators and the driving trajectory map to generate a visual dynamic management interface; in the running state of the visual dynamic management interface, in response to a tracing instruction triggered by a management terminal, calling a historical trajectory database to perform backtracking optimization processing on the driving trajectory map, and outputting management feedback data including abnormal event annotations and path comparison results.

[0006] In a second aspect, an embodiment of the present invention provides a visual management system, including: processor; a storage device having a computer program stored thereon, When the computer program is executed by the processor, the processor implements any of the bus travel visualization management methods combined with positioning technology.

[0007] An embodiment of the present invention provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the bus travel visualization management method combined with positioning technology are implemented.

[0008] It can be seen that the embodiments of the present invention have the following beneficial effects: by constructing a closed-loop analysis system for bus vehicle operation data, a multi-dimensional performance breakthrough in the field of intelligent bus traffic management is achieved.

[0009] First, the dynamic trajectory fitting is coupled with the task execution logic in time and space, and a high-fidelity map with event marking characteristics is generated through an adaptive trajectory modeling algorithm, which can break through the bottleneck of the fragmented analysis of traditional positioning data and business scenarios.

[0010] Secondly, the deep feature extraction mechanism based on the multi-dimensional analysis engine can simultaneously capture the dynamic correlation characteristics of driving behavior patterns and task execution performance, forming a composite evaluation indicator with decision-making value.

[0011] Then, through the interactive visualization interface built with spatiotemporal correlation mapping technology, the intelligent integration of massive spatiotemporal data and business rules is realized, which greatly improves the efficiency of tracing and responding to abnormal events.

[0012] Furthermore, the backtracking optimization algorithm provided by the embodiment of the present invention can autonomously identify trajectory deviation patterns and generate optimization suggestions through the comparative learning mechanism of the historical trajectory database, thereby forming a closed-loop management system with self-evolution capability.

[0013] Therefore, when the embodiments of the present invention realize the visual management of official vehicles, it can significantly improve the controllability and explainability of official vehicles during task execution, and provide dynamic optimization decision support for scenarios such as intelligent scheduling, abnormal diagnosis and path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A flowchart of a bus travel visualization management method combined with positioning technology provided in an embodiment of the present invention.

[0015] Figure 2 A schematic diagram of the basic structure of a visual management system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0017] See also Figure 1 As shown in FIG. 1 , this figure is a flow chart of a bus travel visualization management method combined with positioning technology provided by an embodiment of the present invention, which can be applied to a visualization management system. Figure 1 As shown, the method may include S101-S105.

[0018] S101: Acquire vehicle real-time positioning data of a target vehicle, and extract task execution information associated with the vehicle real-time positioning data, wherein the vehicle real-time positioning data includes positioning coordinates and motion parameters of a continuous time series.

[0019] In this embodiment, the visualization management system continuously collects the real-time positioning data and motion parameters of the vehicle by deploying a multi-modal sensor device on the target vehicle that performs the task of transporting information for the infrastructure project.

[0020] For example, the target vehicle is responsible for transporting construction drawings for a highway expansion project. It departs from the engineering design institute in the north of the city and plans to transport them southward along the G60 Expressway to the construction site of Section 3. The vehicle's real-time positioning data is composed of a redundant positioning system consisting of a dual-frequency high-precision global positioning receiver, an inertial measurement unit, and a wheel speed sensor. The dual-frequency high-precision global positioning receiver can provide centimeter-level positioning accuracy on open roads, and the three-axis gyroscope and accelerometer built into the inertial measurement unit maintain positioning continuity in the tunnel area through a dead reckoning mechanism.

[0021] For example, when the target vehicle enters a 2.1-kilometer-long tunnel in a mountain area, the global positioning signal is completely lost. At this time, the inertial measurement unit continuously outputs estimated position data based on the last known coordinates of the tunnel exit (longitude 118.764532°, latitude 31.992457°) and the vehicle kinematic model, and performs Kalman filtering fusion with the number of tire rotation pulses recorded by the wheel speed sensor to ensure that the positioning error is controlled within 0.5 meters.

[0022] Furthermore, the motion parameter acquisition module records the vehicle's longitudinal acceleration, lateral offset angle and pitch attitude data at a period of 50 milliseconds. When the target vehicle encounters a temporarily paved steel plate road surface on the construction access road, the system identifies abnormal road surface unevenness by analyzing the acceleration spectrum characteristics and triggers the on-board camera device to capture real-time images of the road surface.

[0023] In addition, the task execution information extraction module is deeply integrated with the engineering material management equipment. When the transportation task is started, it automatically associates the version number of the construction drawing (such as V3.2.1), the electronic label number of the packaging box (ET-20231127-003) and the information of the person responsible for signing. It also performs real-time verification when the target vehicle passes through the electronic fence boundary. For example, when the target vehicle deviates from the scheduled route and approaches an unauthorized area, the system immediately compares the geographic fence coordinates in the electronic fence database and generates an abnormal event log containing the deviation distance (such as 1.2 kilometers) and duration.

[0024] S102: Perform dynamic trajectory fitting processing on the real-time positioning data of the vehicle to generate a driving trajectory map that matches the task execution information, wherein the driving trajectory map includes timestamps of path nodes and key event marks.

[0025] In this embodiment, the visualization management system uses an improved Bayesian trajectory estimation algorithm to perform dynamic trajectory fitting processing on the discrete positioning data stream of the target vehicle to generate a driving trajectory map that matches the transportation task of the highway reconstruction and expansion project.

[0026] When the target vehicle needs to detour via County Road X205 due to the closure of G60 Expressway K12+300, the system automatically identifies the spatial correlation between the actual driving path and the original planned path by introducing road network topology constraints. During the trajectory fitting process, the system uses the cubic B-spline curve interpolation algorithm to encrypt the sparse positioning points to eliminate the trajectory breakage caused by the interruption of the positioning signal in the tunnel.

[0027] For example, in a mountain tunnel, the system spatially matches the trajectory points reconstructed based on inertial navigation data with the three-dimensional model of the tunnel to verify the consistency between the estimated path and the actual direction of the tunnel. The key event marking module of the driving trajectory map uses multi-source data fusion technology to correlate and analyze vehicle state parameters with environmental perception data. When the target vehicle reverses due to difficulties in meeting other vehicles while detouring the county road, the system combines the turn signal, reversing radar data and the on-board camera image to mark the "reversing on complex road" event at the corresponding position of the trajectory map (longitude 118.801234°, latitude 31.978901°), and records the duration (2 minutes and 15 seconds) and the number of operations (3 times).

[0028] For example, the timestamp generation mechanism adopts a layered time synchronization protocol to calibrate the vehicle terminal clock with the Beidou satellite timing signal at the millisecond level to ensure that the time reference of each path node in the trajectory map is unified. For example, when crossing the coverage areas of communication base stations of different operators, the system automatically corrects the time deviation caused by transmission delay through the network time protocol.

[0029] S103: Perform multi-dimensional analysis on the driving trajectory map based on preset task rules to determine the driving state characteristics and task completion index of the target vehicle.

[0030] In this embodiment, the visual management system performs a multi-dimensional analytical evaluation of the target vehicle's driving trajectory map based on a preset engineering transportation task rule library, which includes transportation timeliness constraints (such as a 4-hour full-trip time limit), path compliance requirements (such as prohibiting entry into unreinforced temporary access roads), and cargo safety indicators (such as a maximum allowable vibration acceleration of 0.3g).

[0031] When the target vehicle's cargo box vibration monitoring value reached 0.28g due to bumpy road conditions while detouring County Road X205, the system extracted the characteristic frequency of the vibration signal through a wavelet packet decomposition algorithm, compared it with the resonance frequency threshold of the packaging box in the construction drawing, and calculated the risk factor of cargo damage.

[0032] For example, the driving status feature extraction module uses a sliding window analysis method to calculate the vehicle's average speed, acceleration variation coefficient and heading stability index in 5-minute time units. On county roads with continuous curves, the system detects that the standard deviation of lateral acceleration exceeds the safety threshold and automatically generates an "abnormal driving behavior" warning.

[0033] For example, the task completion index evaluation model integrates parameters such as remaining mileage, time margin and energy consumption coefficient. When the actual mileage of the target vehicle increases by 8.7 kilometers due to detour, the system recalculates the arrival time estimate and compares it with the original planned construction drawing handover time window, and updates the task delay risk level to yellow warning status.

[0034] S104: Performing spatiotemporal correlation mapping on the driving state characteristics, the task completion index and the driving trajectory map to generate a visual dynamic management interface.

[0035] In this embodiment, the visual management system uses the spatiotemporal coding module to integrate the driving status characteristics of the target vehicle and the task completion index into the three-dimensional geographic information platform, and constructs a visual dynamic management interface with multi-layer semantics. The visual dynamic management interface integrates the BIM model data of the highway reconstruction and expansion project, and synchronously displays the actual driving trajectory, planned path and electronic fence of the construction area of ​​the target vehicle in the three-dimensional scene.

[0036] It can be understood that when the risk of mission delay increases due to detours, the system uses a dynamic coloring algorithm to gradually change the color of the trajectory segment from standard blue to warning orange, and the color transition rate is positively correlated with the probability of delay. The abnormal state visualization module superimposes multi-dimensional information through augmented reality technology. For example, when the vibration monitoring value of the cargo box approaches the safety threshold, the vehicle's three-dimensional model automatically displays the vibration frequency spectrum and compares and annotates it with the allowed safety interval. The spatiotemporal correlation mapping process uses a WebGL-based rendering optimization algorithm to ensure that smooth interaction at 60 frames per second can be maintained in complex scenes containing tens of thousands of spatial elements. For example, when showing the target vehicle passing through the construction area, the system renders the working radius of the construction machinery in real time and calculates the minimum safe distance between the vehicle and the excavator's working surface.

[0037] S105: In the running state of the visual dynamic management interface, in response to the tracing instruction triggered by the management terminal, the historical trajectory database is called to perform backtracking optimization processing on the driving trajectory map, and management feedback data including abnormal event annotations and path comparison results is output.

[0038] In this embodiment, the visualization management system responds to the traceability instruction issued by the engineering command center and performs a retrospective optimization analysis on the abnormal driving trajectory of the target vehicle on the County Road X205 section. The system calls the historical weather database to obtain the rainfall data of the day (12 mm / hour) and re-evaluates the safety of the detour route in combination with the road friction coefficient model. The path comparison module uses a dynamic time warping algorithm to eliminate the time axis offset caused by temporary parking and accurately calculates the spatial similarity between the actual path and the emergency backup path.

[0039] For example, when analyzing a reversing event, the system associates the traffic monitoring video and vehicle sensor data of the period, constructs an abnormal event chain containing evidence from six dimensions, and ultimately determines the necessity of the reversing operation. The optimization processing module generates path planning suggestions based on the reinforcement learning model. When it detects that County Road X205 has a continuous traffic risk, it automatically updates the detour solution library to give priority to County Road X207. The management feedback data output module uses the causal reasoning module to generate a report containing the root cause analysis, clearly pointing out that the delay in updating the construction closure information is the main factor leading to path deviation, and proposes optimization suggestions for increasing the dynamic update frequency of the electronic fence.

[0040] In the exemplary application scenario of the full-process visualization management of buses involved in the embodiment of the present invention, in a certain highway expansion project in the west, the target vehicle undertakes the core task of transporting construction drawings from the engineering design institute to the construction site. The vehicle is equipped with a redundant positioning system consisting of dual-frequency GPS, IMU inertial navigation and wheel speed sensors, which collect centimeter-level coordinates, acceleration and attitude data in a 50 millisecond period. When the vehicle travels south along the G60 Expressway to K12+300, it detours around County Road X205 due to construction closure. The multimodal sensing system immediately starts collaborative positioning: GPS provides precise coordinates of 118.764532°E / 31.992457°N on open roads; after entering the 2.1-kilometer mountain tunnel, the IMU calculates the trajectory based on the last valid positioning of the tunnel exit through the vehicle kinematic model, and fuses it with the wheel speed pulse data through Kalman filtering to control the positioning error within 0.5 meters. At the same time, the task execution information module associates the packaging box electronic label ET-20231127-003, the drawing version V3.2.1 and the information of the person responsible for signing in real time. When the vehicle deviates from the electronic fence by 1.2 kilometers, the system automatically triggers the abnormal log and starts the on-board camera to capture the actual road conditions.

[0041] Furthermore, the dynamic trajectory fitting module uses the Bayesian algorithm and cubic B-spline interpolation to encrypt discrete positioning points into continuous trajectory lines. In the tunnel section, the system matches the navigation data calculated by the IMU with the three-dimensional model of the tunnel to correct the trajectory break problem; when the vehicle performs three reversing operations on County Road X205 due to difficulties in meeting other vehicles, the system integrates the turn signal and the reversing radar data, marks the "reversing on complex roads" event at 118.801234°E / 31.978901°N on the trajectory map, and simultaneously records the operation time of 2 minutes and 15 seconds. The Beidou timing system ensures that all node timestamps are synchronized at the millisecond level. Even if it passes through the blind area of ​​the communication base station coverage, the timing deviation can still be calibrated through the network time protocol to form a driving trajectory map that is consistent in time and space.

[0042] In this application scenario, based on the preset transportation rule library, the system conducts in-depth analysis of the trajectory map: on the bumpy section of County Road X205, the vibration monitoring value of the cargo box reached 0.28g, close to the 0.3g safety threshold. The wavelet packet decomposition algorithm identified the 12Hz characteristic component overlapping with the resonance frequency of the packaging box, triggering an orange cargo risk warning; at the same time, the sliding window analysis showed that the standard deviation of the lateral acceleration of this section exceeded the limit by 40%. The system combined the temporary steel plate pavement image captured by the camera and determined it as "abnormal pavement conditions" and generated a maintenance work order. The task completion evaluation model dynamically calculated the 8.7-kilometer mileage increment caused by the detour, re-estimated the arrival time based on the energy consumption coefficient, raised the delay risk level to yellow, and synchronized the handover time window to the construction site through the 5G network.

[0043] At the same time, the 3D visualization platform integrates the above data into the engineering BIM model and presents a dynamic management interface: the vehicle trajectory line gradually changes from blue to orange according to the probability of delay, the frequency spectrum comparison chart is superimposed on the vibration limit position, and the safety radius of the construction machinery is rendered in a red semi-transparent layer. When the vehicle approaches the excavator working surface, the system calculates the 3.2-meter distance in real time and triggers an audible and visual alarm. With the help of WebGL rendering optimization, the interface still maintains a smooth interaction of 60 frames per second in the highway reconstruction and expansion scene containing tens of thousands of spatial elements, supporting the command center to inspect vehicle posture, path compliance and cargo status from multiple perspectives.

[0044] It is understandable that after the engineering command center initiated the tracing command, the system called the historical database to conduct a multi-dimensional backtracking of the abnormal events on County Road X205: integrating the 12 mm / hour rainfall data of the day, the road friction coefficient model and the on-board video to verify the rationality of the reversing operation; the dynamic time warping algorithm was used to eliminate the timing offset caused by temporary parking, and the spatial similarity between the actual path and the backup route X207 was calculated to be 92%, thereby triggering the optimization of the detour plan. The reinforcement learning model recommended X207 as the preferred alternative route based on historical traffic data. The causal reasoning module traced the source and found that the delay in updating the construction closure information was the main cause, and generated an optimization suggestion to increase the dynamic update frequency of the electronic fence to 15 minutes / time. Finally, the management feedback data package was stored through the blockchain, including event chain analysis, path comparison heat map and equipment maintenance work order, forming a complete closed loop from data perception to decision optimization, significantly improving the execution reliability and emergency response efficiency of subsequent tasks.

[0045] In one implementation, the performing of dynamic trajectory fitting processing on the real-time positioning data of the vehicle in S102 to generate a driving trajectory map matching the task execution information includes: S1021: Perform timestamp alignment filtering on the real-time positioning data of the vehicle according to the set of planned path points in the task execution information, and select candidate positioning points that meet a preset time window.

[0046] In this embodiment, the visualization management system performs timestamp alignment and filtering on the positioning data collected in real time by the target vehicle according to the set of construction drawing transportation plan path points predefined in the task execution information. Specifically, the system loads the planned path nodes such as the starting point of the engineering design institute, the construction closure point K12+300 of the G60 Expressway, the detour section of the county road X205, and the end point of the construction site of the third bid section corresponding to the transportation task, and sets the allowed time window threshold for each node. For example, when the target vehicle is planned to pass the K12+300 construction point between 10:15 and 10:30, the system performs timestamp matching on the positioning data uploaded by the vehicle during this period, and selects candidate positioning points with a time deviation within ±2 minutes. In this process, the system calibrates the timestamps of each positioning point at the millisecond level through the Beidou satellite timing signal, excludes the time sequence disorder data caused by network delay, and ensures that the time-space correlation between the candidate positioning point and the planned path meets the requirements of the engineering transportation task.

[0047] S1022: Performing a path continuity check on the candidate positioning points, using an interpolation algorithm to complete the missing intermediate positioning points, and generating a corrected continuous driving path.

[0048] In this embodiment, the system performs a path continuity check on the screened candidate positioning points, and uses a cubic B-spline interpolation algorithm to complete the missing intermediate positioning points. When the target vehicle's wheel speed sensor signal is briefly interrupted due to continuous curves in the detour section of County Road X205, the system detects that there is an abnormal spacing of more than 50 meters between adjacent positioning points, and automatically starts the interpolation calculation module. This module is based on the vehicle kinematic model, combined with historical driving speed, heading angle change rate and other parameters, to generate three virtual positioning points in the longitude range of 118.798765° to 118.802341° to form a corrected continuous driving path. During the interpolation process, the system introduces road network topology constraints to ensure that the completion points conform to the actual road shape of County Road X205, and avoid unreasonable trajectories such as crossing buildings or crossing the boundary to unauthorized areas.

[0049] S1023: extracting abnormal positioning points in the continuous driving path that match preset key event triggering conditions, and associating and marking the abnormal positioning points with task nodes in the task execution information.

[0050] In this embodiment, the visualization management system extracts abnormal positioning points that match the preset key event trigger conditions in the continuous driving path. When the target vehicle performs three reversing operations at a narrow bend on County Road X205, the system monitors in real time the occurrence of multiple abnormal features in the positioning point sequence, such as sudden change in heading angle, speed zeroing, and excessive stay time. By inputting the instantaneous speed fluctuation value (from 25km / h to 0km / h), direction deviation angle (120° counterclockwise turn) and stay time parameters (135 seconds) into the multimodal event classification model, the system outputs the "complex road section reversing" abnormal event type and a 92.7% confidence score. Subsequently, the system associates the abnormal positioning point (longitude 118.801234°, latitude 31.978901°) with the construction drawing handover time node in the task execution information, and generates a key event label containing the event type, occurrence time and impact range.

[0051] S1024: Calling the trajectory prediction model to perform driving mode recognition on the continuous driving path, dividing it into an acceleration segment, a constant speed segment and a stagnant segment, and generating a driving trajectory map containing segmented feature descriptions.

[0052] In this embodiment, the system calls a trajectory prediction model based on a deep neural network to identify the driving mode of the continuous driving path. When analyzing the detour section of County Road X205, the model identified three characteristic sections: the acceleration section from the K12+300 construction point to the entrance of the county road (the speed increases from 0 to 40km / h), the constant speed cruising section in the middle section of the county road (maintaining 35±2km / h), and the stagnation section in the temporary parking area (lasting 8 minutes). The model generates a driving trajectory map containing segmented feature descriptions based on the acceleration change rate, speed stability coefficient and energy consumption characteristics, in which the acceleration section is marked with a maximum longitudinal acceleration of 0.3m / s², and the stagnation section is associated with an abnormal state in which the energy consumption of the vehicle air conditioner increases by 12%.

[0053] In one implementation, the extracting of abnormal positioning points in the continuous driving path that match preset key event trigger conditions in S1023 includes: S10231: Obtain the instantaneous speed fluctuation value, direction deviation angle and dwell time parameters in the real-time positioning data of the vehicle.

[0054] In this embodiment, for example, when the target vehicle passes through a temporary steel plate road surface, the inertial measurement unit records that the instantaneous fluctuation value of the longitudinal acceleration reaches ±0.5m / s², the cumulative change of the direction deviation angle exceeds 15°, and the length of stay in the steel plate area reaches 45 seconds. These parameters are uploaded to the visual management system in real time through the vehicle communication module, providing multi-dimensional input data for subsequent event classification.

[0055] S10232: Input the instantaneous speed fluctuation value, the direction deviation angle and the dwell time parameters into a target multimodal event classification model, and output the abnormal event type and confidence score.

[0056] In this embodiment, when the target vehicle encounters temporary traffic control on the construction access road, the system inputs the instantaneous speed fluctuation value (from 40km / h to 5km / h), direction deviation angle (continuous left turn 30°) and stop time (210 seconds) into the model, and after joint processing by the convolutional neural network and the long short-term memory network, the "traffic control waiting" abnormal event type and 88.3% confidence score are output. During the model training process, 5,000 groups of similar event data in historical tasks are integrated to ensure classification accuracy.

[0057] S10233: Extracting a set of abnormal positioning points that meet a threshold condition from the continuous driving path according to the abnormal event type and the confidence score.

[0058] In this embodiment, the system screens abnormal positioning points according to preset thresholds. When the speed fluctuation threshold is set to ±0.4m / s², the direction deviation threshold is set to 20°, and the stay time threshold is set to 120 seconds, when the target vehicle stays in the temporary rest area for 150 seconds and the direction deviation reaches 25°, the system automatically extracts the positioning point (longitude 118.812345°, latitude 31.965432°) and adds it to the abnormal positioning point set. At the same time, combined with the confidence score threshold of 85%, false alarm events caused by sensor noise are excluded.

[0059] S10234: Perform similarity matching on the abnormal location point set and similar events in the historical trajectory database to generate a key event tag including an event association identifier.

[0060] In this embodiment, the system matches abnormal events with the historical database. For the reversing event of the target vehicle on County Road X205, the system retrieves the data of 18 reversing events that occurred on similar roads in the past three months. By calculating the similarity of trajectory morphology (Euclidean distance <5 meters) and comparing operation features (number of reversing times, duration), the system generates key event markers with the same event identification code to provide data support for subsequent responsibility tracing.

[0061] In one implementation, the performing of multi-dimensional analysis on the driving trajectory map based on preset task rules in S103 to determine the driving state characteristics and task completion index of the target vehicle includes: S1031: Extracting the task priority weight, planned completion time and preset path constraint conditions from the task execution information.

[0062] In this embodiment, the visual management system extracts core parameters from the task execution information, among which the construction drawing transportation task is given the highest priority weight (level A), the planned completion time is set to 4 hours (including 1 hour of redundancy), and the preset path constraints include prohibition of driving on unreinforced side roads and speed limit of 80km / h. These parameters constitute the benchmark framework for task analysis and guide the subsequent deviation calculation and score generation.

[0063] S1032: Calculate the deviation index between the actual driving path and the planned path according to the path node density distribution in the driving trajectory map.

[0064] In this embodiment, the system calculates the path deviation based on the path node density distribution in the driving trajectory map. When the target vehicle detours around County Road X205, the system matches the actual path with the planned path at intervals of 200 meters, and finds that the node density of the county road section increases by 35%. Combined with the spatial position offset (maximum lateral deviation of 1.2 kilometers), the path deviation index is calculated to be 0.78 (threshold > 0.5 triggers an early warning).

[0065] S1033: Determine a dynamic score of the task completion index by combining the deviation index and the task priority weight.

[0066] In this embodiment, the system dynamically scores the task based on the deviation index and priority weight. Since this task has the highest priority, the system uses a weighted algorithm to convert the deviation index of 0.78 caused by the county road detour into a 22% decrease in the task completion score. The scoring model also considers factors such as time margin (1.5 hours remaining) and energy consumption increment (8% increase), generating a comprehensive score of 72 points (out of 100), corresponding to the yellow warning level.

[0067] S1034: Perform time series analysis on key event markers in the driving trajectory map to generate a status assessment report including event impact coefficients.

[0068] In this embodiment, for the reversing event (10:25-10:27) and the vibration exceeding limit event (10:35-10:37), the start timestamp, duration and spatial distribution density parameters are extracted. By constructing the event impact propagation model, the correlation strength coefficient between the two events is calculated to be 0.65 (threshold>0.6), indicating that the vibration exceeding limit may be a subsequent impact caused by the reversing operation.

[0069] S1035: Compare the dynamic score and the status assessment report with the acceptance conditions in the preset task rules, and output the compliance determination result of the driving status characteristics.

[0070] In this embodiment, the system compares the dynamic score with the acceptance conditions. According to the preset rules, the yellow warning status (score 70-80 points) triggers the secondary response mechanism, and the system automatically generates a status report containing three corrective measures: it is recommended to reduce the speed of county roads, notify the construction site in advance to delay the handover, and arrange for the maintenance unit to inspect the temporary steel plate pavement.

[0071] In one implementation, the step of performing time series analysis on key event markers in the driving trajectory map in S1034 to generate a status assessment report including event impact coefficients includes: S10341: Extract the start timestamp, duration and spatial distribution density corresponding to the key event mark.

[0072] In S10341, the system extracts the time-space features of key events. For example, the starting timestamp of the reversing event is 2023-11-27T10:25:13.456, the duration is 165 seconds, and there are 3 positioning point anomalies within a 50-meter section, with a spatial distribution density of 6 times / 100 meters. These parameters provide basic data for building an event impact model.

[0073] S10342: Generate an event impact propagation model based on the time axis at least based on the start timestamp and the duration, and calculate the correlation strength coefficient between adjacent key events.

[0074] In S10342, taking the reversing event as the starting point, the vibration exceeding limit, speed fluctuation and other events occurring within the subsequent 30 minutes are analyzed, and the propagation coefficient is calculated to be 0.78 through time axis correlation analysis, indicating that the reversing operation has a continuous impact on the subsequent driving status.

[0075] S10343: Generate an event chain reaction probability map according to the correlation strength coefficient and the spatial distribution density.

[0076] In S10343, when the target vehicle reverses on County Road X205, the model predicts that the probability of road anomaly detection within the next 5 kilometers will increase to 65%, and the risk of damage to the construction drawing package will increase by 40%. The probability map is presented in the form of a heat map on the 3D geographic information platform.

[0077] S10344: Input the event chain reaction probability graph into the risk prediction model, and output the quantitative evaluation value of the event impact coefficient.

[0078] In S10344, combined with the consequence analysis of 20 similar events in historical data, the system calculated that the impact coefficient of the current reversing event is 0.58 (range 0-1), corresponding to a medium risk level, triggering the on-board camera device to continuously monitor the status of the cargo box.

[0079] S10345: Prioritize abnormal events in the status assessment report based on the quantitative assessment value.

[0080] In S10345, the system prioritizes abnormal events. For example, according to the quantitative evaluation value, the reversing event (0.58) and the vibration exceeding limit event (0.62) are marked as P1 and P2 priorities respectively. The generated status evaluation report gives priority to the treatment suggestions for vibration exceeding limit, including immediate deceleration to 30km / h, arranging on-site personnel to pre-inspect the packaging boxes, and other instructions.

[0081] With this design, the visual management system realizes accurate monitoring of the entire construction drawing transportation task, significantly improving the task execution reliability and management efficiency of official vehicles.

[0082] In one implementation, the step of performing spatiotemporal correlation mapping of the driving state characteristics, the task completion index and the driving trajectory map in S104 to generate a visual dynamic management interface includes: S1041: Extracting a geographic coordinate set and a timestamp sequence of path nodes from the driving trajectory map.

[0083] In this embodiment, the visualization management system extracts the geographic coordinate set and timestamp sequence of the path nodes from the target vehicle's driving trajectory map. Specifically, the system parses the trajectory data stream generated by the improved Bayesian trajectory estimation algorithm, separates the geographic coordinate set containing longitude and latitude information, and the Beidou satellite timing timestamp sequence corresponding to each coordinate point. For example, during the target vehicle's task of transporting data for the highway reconstruction and expansion project, the system extracts more than one thousand path nodes from the Engineering Design Institute (starting point: longitude 118.750123°, latitude 32.001234°) to the construction site of Section 3 (end point: longitude 118.845678°, latitude 31.956789°), and each node accurately records the coordinated universal time timestamp when the vehicle arrives at the location, with a time resolution of milliseconds. For the dead reckoning coordinate points generated by the interruption of the global positioning signal in the tunnel area, the system automatically adds the inertial navigation data source identifier to ensure the integrity and traceability of the geographic coordinate set.

[0084] S1042: Map the geographic coordinate set and the timestamp sequence to layer data of an electronic map to generate a basic track overlay layer.

[0085] In this embodiment, the system maps the geographic coordinate set and timestamp sequence to the vector layer of the high-precision electronic map to generate a basic track overlay. This process uses a spatial reference system conversion algorithm to convert the WGS84 coordinate system data collected by the global positioning receiver into the local projection coordinate system used by the engineering electronic map. For example, when the target vehicle detoured along County Road X205, the system matched the track point coordinates (longitude 118.798765°, latitude 31.978901°) with the center line of the road in the electronic map to generate a track overlay with adjustable width. For sections of mountain tunnels that lack satellite positioning signals, the system calls the tunnel three-dimensional model data and spatially calibrates the coordinate points (longitude 118.769876°, latitude 31.987654°) calculated by inertial navigation with the lane lines inside the tunnel to ensure the continuous display of the basic track overlay in signal-free areas.

[0086] S1043: Overlaying the speed heat map of the driving state characteristics and the progress indicator of the task completion index on the basic trajectory overlay layer.

[0087] In this embodiment, the visualization management system overlays a speed heat map of driving state characteristics and a progress indicator of task completion indicators on the basic trajectory overlay. The speed heat map uses a color gradient mapping algorithm to visualize the target vehicle's speed change characteristics on each section of County Road X205 using a color spectrum from dark blue (low speed) to bright red (high speed). For example, in a section where the vehicle reversed three times due to difficulties in meeting other vehicles (longitude 118.801234°, latitude 31.978901°), the system detected that the speed dropped to 0km / h and lasted for 135 seconds. The area is displayed as a dark blue patch in the heat map. At the same time, the progress indicator displays the percentage of task completion in real time in the form of a floating window. When the vehicle detours and causes the actual progress to lag behind the planned time by 15%, the logo automatically switches to an orange warning state and marks the estimated delay time.

[0088] S1044: Embed an interactive event annotation window in the electronic map according to the spatial position of the key event mark.

[0089] In this embodiment, the system embeds an interactive event annotation window in the electronic map according to the spatial position of the key event marker. When the target vehicle triggers a vibration over-limit warning on the construction access road, the system generates an annotation window with an event type icon at the corresponding coordinate point (longitude 118.812345°, latitude 31.965432°). This window supports multi-level interactive operations: the primary click displays the event summary (such as "vibration acceleration 0.28g, close to the safety threshold"), and the secondary expansion provides a detailed parameter list (including vibration spectrum characteristics, road surface image capture records). The spatial positioning accuracy of the annotation window reaches 0.1 meters. Even if the vehicle triggers an event while moving, it can still be accurately associated with the corresponding position on the electronic map.

[0090] S1045: Integrate the basic track overlay layer, the speed heat map, the progress indicator, and the interactive event annotation window to generate a multi-layered visual dynamic management interface.

[0091] In this embodiment, the system integrates the basic trajectory overlay, speed heat map, progress indicator and interactive event annotation window to generate a multi-layered visual dynamic management interface. The interface uses layered rendering technology to ensure the independent control and coordinated display of each visualization element. For example, when showing the target vehicle passing through the detour section of County Road X205, the basic trajectory layer presents a blue solid line path, the speed heat map covers the speed distribution with translucent blocks, the progress indicator is dynamically updated in the upper right corner of the interface, and the interactive annotation window of the reversing event flashes at the corresponding position. The system accelerates the rendering pipeline through the graphics processor to achieve smooth interaction at 60 frames per second in complex scenes containing tens of thousands of spatial elements.

[0092] In one implementation, the step of generating a multi-layer overlay visual dynamic management interface in S1045 includes: S10451: In response to a user's touch operation on the interactive event annotation window, retrieving sensor raw data associated with the key event marker.

[0093] In this embodiment, the visual management system responds to the touch operation of the interactive event annotation window by the operator of the engineering command center to retrieve the raw sensor data associated with the key event mark. When the user clicks the County Road X205 reversing event annotation icon in the interface, the system immediately retrieves the raw acceleration data of the inertial measurement unit, the wheel speed sensor pulse signal and the 1280×720 resolution video stream captured by the on-board camera device during the period (2023-11-27T10:25:13 to 10:27:28) from the distributed storage cluster. The data retrieval process uses a timestamp alignment mechanism to ensure strict synchronization of multimodal sensor data.

[0094] S10452: Analyze the raw data of the sensor in real time to generate an event details panel including vehicle fault codes, environmental parameters and operation logs.

[0095] In this embodiment, the system performs real-time analysis of the sensor raw data and generates an event details panel containing vehicle fault codes, environmental parameters, and operation logs. For reversing events, the analysis module extracts the turn signal control signal (left turn activated 3 times), the reversing radar detection data (the nearest obstacle distance is 0.8 meters), and the gear switching record recorded by the engine control unit (forward gear / reverse gear switching 3 times). The environmental parameter analysis module also analyzes the on-board atmospheric pressure sensor data and identifies that the rainfall intensity reached 12 mm / hour when the event occurred, and the road friction coefficient dropped to 0.35. All analysis results are presented in the event details panel in the form of a structured table, with a time-aligned multimedia evidence chain attached.

[0096] S10453: Linking the event details panel with the visual dynamic management interface for rendering, and updating the extended information layer that displays the abnormal event.

[0097] In this embodiment, the system renders the event details panel in conjunction with the visual dynamic management interface, and updates the extended information layer that displays the abnormal event. When the reversing event details panel is expanded, the interface automatically adjusts the layer overlay order: the transparency of the basic trajectory layer is increased to 50%, the resolution of the satellite map layer of the event-related road section is enhanced to 0.5 meters, and the extended information panel slides out on the right side of the interface. The panel synchronously plays the in-vehicle video recording of the reversing process, and highlights the activated reversing radar detection area on the three-dimensional vehicle model, realizing the spatial-temporal correlation display of multi-dimensional data.

[0098] S10454: According to the comparison mode selected by the user, the historical driving trajectory map of the same task is loaded from the historical trajectory database.

[0099] In this embodiment, the system loads the historical driving trajectory map of the same task from the historical trajectory database according to the comparison mode selected by the user. When the engineering manager selects the "historical optimal path comparison" mode, the system retrieves the data of 18 transportation tasks with the same starting and ending points completed in the past three months, and selects the trajectory record with the shortest time and no abnormal events (task number T20231015-003). The loading process adopts the differential data transmission protocol, which only incrementally synchronizes the spatial deviation data of the current trajectory and the historical trajectory to reduce network bandwidth usage.

[0100] S10455: Displaying the path difference analysis results of the current driving trajectory map and the historical driving trajectory map in parallel in the extended information layer.

[0101] In this embodiment, the system displays the path difference analysis results of the current driving trajectory map and the historical driving trajectory map in parallel in the extended information layer. The comparison view is rendered with two-color trajectory lines: the current path around County Road X205 is displayed in orange, and the historical optimal path (via County Road X207) is displayed in green. The spatial analysis module calculates the length difference (8.7 kilometers), the cumulative value of elevation change (+35 meters), and the total turning angle (the current path increases by 120°) of the two paths, and embeds the statistical results in the form of a comparison bar chart in the extended information layer. At the same time, the system calls the path similarity algorithm to generate a spatial overlapping heat map of the two trajectories, focusing on marking the peak points of spatial deviations between the sharp bend area of ​​County Road X205 and the historical path.

[0102] In one implementation, the calling of the historical trajectory database in S105 to perform backtracking optimization processing on the driving trajectory map includes: S1051: extracting historical trajectory segments of a corresponding time period from the historical trajectory database according to the time range parameter in the tracing instruction.

[0103] In this embodiment, the visualization management system extracts the historical trajectory fragments of the corresponding period from the historical trajectory database according to the time range parameters set in the received tracing instruction. Specifically, when the target vehicle performs the transportation task of the engineering materials of the third bid section on November 27, 2023, the system receives a tracing instruction to retrieve the historical transportation records of all the same starting and ending points from October 1 to November 26, 2023. The system quickly locates 17 complete trajectory data that meet the time range constraints in the storage cluster through the spatiotemporal index mechanism, and each trajectory contains no less than one thousand path nodes and their associated sensor data sets. For example, when the historical trajectory of the task number T20231115-007 on November 15, 2023 is retrieved, the system extracts the special path fragment (starting point longitude 118.750123° to end point longitude 118.845678°) generated by the detour of the county road X205 during this transportation, and the fragment contains 328 consecutive trajectory node data that deviate from the standard route.

[0104] S1052: Calculate trajectory smoothness and match path overlap on the historical trajectory segments to identify repetitive driving patterns.

[0105] In this embodiment, the system performs trajectory smoothness calculation and path overlap matching on the extracted historical trajectory fragments to identify repetitive driving patterns. The trajectory smoothness calculation uses a moving average filtering algorithm to eliminate noise points caused by the jitter of the satellite positioning signal. For example, the 5 abnormal jump points (maximum offset 2.3 meters) that appeared on the County Road X205 section on November 15, 2023 were automatically corrected. The path overlap matching uses a spatial topology analysis algorithm to calculate the spatial overlap rate of the current trajectory and the historical trajectory in the detour area. When it is detected that the target vehicle has chosen the County Road X205 detour route in the last three transportation tasks and the path overlap reaches 92%, the system marks it as a high-frequency repetitive pattern and associates the pattern with the validity period data of the construction control notice (October 20, 2023 to January 10, 2024).

[0106] S1053: Constructing a trajectory prediction model based on the repetitive driving pattern to generate task path optimization suggestions for the target period.

[0107] In this embodiment, the system constructs a trajectory prediction model based on the identified repetitive driving patterns and generates optimization suggestions for the task path during the target period. The model uses the time distribution characteristics of the periodic path node set and combines the real-time traffic status data stream for dynamic path planning. For example, in response to the congestion pattern of engineering vehicles on County Road X205 from 10:00 to 11:00 am every Monday, Wednesday, and Friday, the system generates an optimization suggestion for detouring County Road X207, which contains a detailed path node sequence (12 key turning points from longitude 118.802345° to longitude 118.823456°) and estimated time saving parameters (an average reduction of 18 minutes). The optimization suggestion is synchronously linked to the temporary control information of the road maintenance department to ensure the real-time effectiveness of the recommended path.

[0108] S1054: Simulate and superimpose the task path optimization suggestion for the target period on the current driving trajectory map, and output management feedback data including the path comparison result.

[0109] In this embodiment, the system simulates and superimposes the task path optimization suggestion with the current driving trajectory map, and outputs management feedback data including the path comparison results. The simulation process uses three-dimensional space projection technology to display the spatial difference between the actual driving trajectory (County Road X205 detour route) and the recommended path (County Road X207 alternative route) in parallel on the electronic map. The system automatically calculates the length difference between the two paths (County Road X205 route increases by 2.7 kilometers), the cumulative value of elevation change (increases by 45 meters), and the estimated fuel consumption difference (increases by 3.8 liters), and encodes the comparison results into a structured data packet. The management feedback data is transmitted to the engineering dispatch center through a dedicated communication protocol, triggering the path optimization approval process.

[0110] S1055: According to the distribution density of the abnormal event annotations, adjust the weight parameters of the trajectory prediction model to update the historical trajectory database.

[0111] In this embodiment, the system adjusts the weight parameters of the trajectory prediction model according to the distribution density of abnormal event annotations to update the historical trajectory database. When it is detected that the detour section of County Road X205 has triggered 27 vibration over-limit warnings in the past 30 days, the system automatically increases the risk weight parameter of the section (from 0.35 to 0.78) and reduces its ranking in the path recommendation priority accordingly. The update process uses the gradient descent algorithm to optimize the model parameters, and re-associate and store the adjusted weight parameters with the historical trajectory data to ensure that high-frequency abnormal event areas are avoided first during subsequent path planning.

[0112] In one implementation, the step of constructing a trajectory prediction model based on the repetitive driving pattern in S1053 includes: S10531: Extract the set of periodic path nodes and driving time intervals from the historical trajectory segments.

[0113] In this embodiment, the system extracts the set of periodic path nodes and driving time intervals from the historical trajectory segments. For the task of detouring on County Road X205 that the target vehicle executes three times a week regularly, the system extracts 42 characteristic path nodes that repeatedly appear during the period from 9:30 to 10:15 in the morning on Mondays, Wednesdays, and Fridays, forming a set of periodic path nodes. Each node is associated with an accurate arrival timestamp. For example, it will surely arrive at the intersection of County Road X205 and the construction access road (longitude 118.812345°, latitude 31.965432°) at 9:47:23 on Wednesday morning, and the time deviation is controlled within ±2 minutes. The driving time interval calculation module synchronously analyzes the passing time consumption between adjacent nodes and establishes a reference time matrix for typical periods.

[0114] S10532: Use a time series clustering algorithm to classify the patterns of the set of periodic path nodes and generate a typical driving pattern template.

[0115] In this embodiment, the system uses a time series clustering algorithm to classify the patterns of the set of periodic path nodes and generate a typical driving pattern template. The algorithm performs similarity matching on the time series of path nodes in different tasks through dynamic time warping technology. When it is detected that 8 transportation tasks show highly consistent node distribution and time interval characteristics on the County Road X205 section, the system clusters them into the "weekday morning peak detour mode" and generates a pattern template containing 76 core path nodes. This template accurately records the standard arrival time window for each node (such as the standard arrival time of node 15 is 38 minutes and 12 seconds ± 15 seconds after the start of the task), forming a reusable path planning reference.

[0116] S10533: Fuse the driving time intervals with the real-time traffic state data and calculate the time cost coefficients of each typical driving pattern.

[0117] In this embodiment, the system fuses the driving time intervals with the real-time traffic state data and calculates the time cost coefficients of each typical driving pattern. The real-time traffic state data includes the released construction control information, the traffic flow density data identified by in-vehicle cameras, and the visibility parameters provided by the meteorological department. For example, when the visibility on the alternative route of County Road X207 is less than 50 meters in rainy and foggy weather, the system dynamically increases the time cost coefficient of this route (from 1.2 to 1.8), and at the same time reduces the time cost coefficient of the County Road X205 detour route (from 1.5 to 1.3). The fusion processing uses a multi-source data weighting algorithm to ensure that the time cost coefficients can reflect the actual road traffic conditions in real time.

[0118] S10534: Dynamically adjust the path recommendation priority of the trajectory prediction model according to the time cost coefficient and the task completion index.

[0119] In this embodiment, the system dynamically adjusts the path recommendation priority of the trajectory prediction model according to the time cost coefficient and the task completion index. When the remaining time margin of the target vehicle for this transportation task is less than 15%, the system prioritizes the County Road X207 route (coefficient 1.2) with the lowest time cost coefficient, replacing the original County Road X205 route (coefficient 1.5). The recommendation priority adjustment algorithm comprehensively considers the remaining mileage, load status (the load of this transport vehicle is 8.5 tons) and the driver's operating habits parameters to generate the optimal path sequence that takes into account efficiency and safety. The dynamic adjustment results are fed back to the vehicle navigation terminal in real time, triggering the route replanning instruction.

[0120] S10535: Embed the dynamically adjusted path recommendation priority into the path comparison result in the management feedback data.

[0121] In this embodiment, the system embeds the dynamically adjusted path recommendation priority into the path comparison result in the management feedback data. When generating the recommended alternative route for County Road X207, the system simultaneously marks the basis for priority adjustment in the management feedback data packet, including real-time traffic status data snapshot (the current queue length of County Road X205 is 380 meters), weather warning information (the probability of rainfall in the next 2 hours is 85%), and historical task completion rate comparison data (the on-time rate of County Road X207 route is 97%). The embedding process uses data layering encapsulation technology to ensure that engineering management personnel can trace the technical basis of each priority decision through feedback data, and support manual review and solution optimization.

[0122] In one implementation, the step of dynamically adjusting the path recommendation priority of the trajectory prediction model according to the time cost coefficient and the task completion index in S10534 includes: S1053401: Obtain a set of time consumption parameters of each typical driving mode corresponding to the time cost coefficient, and a set of path deviation scores in the task completion index.

[0123] In this embodiment, the visualization management system obtains the time consumption parameter set of each typical driving mode corresponding to the time cost coefficient and the path deviation score set in the task completion index. Specifically, when the target vehicle performs the engineering data transportation task of Section 3, the system extracts the time consumption parameter set of the detour mode of County Road X205 from the trajectory prediction model, which includes parameters such as the average time consumption of 48 minutes, the maximum time consumption of 62 minutes, and the fuel consumption coefficient of 1.35. At the same time, the system obtains the path deviation score set of the last three transportation tasks on the same road section, among which the path deviation score of the task on November 15, 2023 due to temporary traffic control is 0.87 (a full score of 1.0 means that it is completely in line with the planned path), the deviation score on November 18 due to the driver's operating habits is 0.68, and the deviation score on November 22 due to vehicle failure is 0.42, forming a dynamic scoring data set containing three time points.

[0124] S1053402: Perform weight allocation processing on the time consumption parameter set according to the path deviation score set to generate a time cost combination weight based on task completion.

[0125] In this embodiment, the system performs weight distribution processing on the time consumption parameter set according to the path deviation score set, and generates a time cost combination weight based on the task completion degree. During the processing, the system uses the path deviation score as a credibility factor to dynamically weight the time consumption parameters. For example, when the deviation scores of the detour mode of County Road X205 in three tasks are 0.87, 0.68, and 0.42 respectively, the system calculates its average deviation score as 0.656, and uses this value as the attenuation coefficient of the time consumption parameter. The average time consumption of 48 minutes in the original time consumption parameter set is adjusted to an equivalent value of 48×(1-0.656)=16.9 minutes, and the maximum time consumption of 62 minutes is adjusted to an equivalent value of 62×(1-0.656)=21.7 minutes, forming a new time cost combination weight set.

[0126] S1053403: Input the time cost combination weight into a priority sorting algorithm, perform initial priority sorting on the candidate path set pre-stored in the trajectory prediction model, and generate a path priority queue with a weight identifier.

[0127] In this embodiment, the system inputs the time cost combination weight into the priority sorting algorithm, performs initial priority sorting on the candidate path set pre-stored in the trajectory prediction model, and generates a path priority queue with weight identification. The candidate path set includes three optional options: the detour route of County Road X205, the alternative route of County Road X207, and the emergency route of the spare township road Y305. The priority sorting algorithm sorts according to the adjusted time cost combination weight, among which the alternative route of County Road X207 obtains the highest priority weight value of 0.92 due to the equivalent average time consumption of 16.2 minutes, the detour route of County Road X205 obtains the second best weight of 0.85 due to the equivalent time consumption of 16.9 minutes, and the emergency route of the spare township road Y305 obtains the weight of 0.73 due to the equivalent time consumption of 23.4 minutes, forming a path priority queue with three-color weight identification (red high, yellow medium, green low).

[0128] S1053404: Based on the weight identifier of each candidate path in the path priority queue, extract a subset of candidate paths that match the preset path constraint condition in the current task execution information.

[0129] In this embodiment, the system extracts a subset of candidate paths that match the preset path constraints in the current task execution information based on the weight identifier of each candidate path in the path priority queue. The current task execution information includes constraints such as vehicle load of 8.5 tons, transport material shockproof grade B, and driver continuous driving time limit of 4 hours. The system automatically filters out routes that do not meet the load requirements in the candidate paths. For example, the emergency route of Township Road Y305 is eliminated due to the bridge weight limit of 8 tons. The remaining detour route of County Road X205 and the alternative route of County Road X207 constitute the candidate path subset. This subset is synchronously associated with the driver's operating habit database to exclude steep slope sections that require frequent gear shifting, ensuring that the path selection meets ergonomic requirements.

[0130] S1053405: ​​Perform a real-time traffic status data adaptability check on each candidate path in the candidate path subset, and eliminate invalid paths with traffic congestion marks or temporary control events.

[0131] In this embodiment, the system performs real-time traffic status data adaptability check on each candidate path in the candidate path subset, and eliminates invalid paths with traffic congestion marks or temporary control events. For example, the system accesses the real-time data interface of the municipal traffic management platform to obtain the latest control information of the County Road X205 section: due to sudden road collapse, the section will be closed in both directions from 09:00 on the same day. The system immediately marks the County Road X205 detour route as an invalid path in the candidate path subset, and displays a red no-entry sign on the electronic map overlay. At the same time, the system verifies the real-time traffic status of the County Road X207 alternative route, confirms that it has not triggered any traffic abnormality event marks, and retains it as a valid candidate path.

[0132] S1053406: Determine a path priority adjustment interval that meets the dynamic adjustment conditions based on the time cost combination weights of the remaining candidate paths and the task completion degree deviation threshold.

[0133] In this embodiment, the system determines the path priority adjustment interval that meets the dynamic adjustment conditions based on the time cost combination weights and task completion deviation thresholds of the remaining candidate paths. The remaining candidate paths only include the County Road X207 alternative route, whose time cost combination weight is 0.92 and the task completion deviation threshold is 0.7 (the preset minimum allowable value). The system calculates the dynamic deviation score of the path as 0.89, which is higher than the deviation threshold of 0.7, confirming that it meets the dynamic adjustment conditions. The system automatically generates a priority adjustment interval of [0.85, 1.0], which indicates that the path priority can be downgraded by up to 15% or upgraded infinitely while ensuring the task completion.

[0134] S1053407: Within the path priority adjustment interval, remap the priorities of the remaining candidate paths in descending order of time-cost combination weights to generate an updated path recommendation priority sequence.

[0135] In this embodiment, the system remaps the priorities of the remaining candidate paths in descending order of the time-cost combination weights within the path priority adjustment interval to generate an updated path recommendation priority sequence. Since the County Road X207 alternative route is the only valid candidate path, the system increases its priority weight from 0.92 to the upper limit of the interval 1.0, and marks it as a mandatory recommended path in the priority sequence. At the same time, the system retrieves similar cases in the historical path database, and inserts the historical record of the emergency route of the rural road Y305 on October 12, 2023 as an alternative plan at the end of the sequence, but its priority weight remains unchanged at 0.73, forming an updated sequence containing the main recommended path and the emergency alternative path.

[0136] S1053408: Compare the updated path recommendation priority sequence with the priority records of similar tasks in the historical priority database to identify abnormal paths whose priority jump amplitude exceeds a preset threshold.

[0137] In this embodiment, the system compares the updated path recommendation priority sequence with the priority records of similar tasks in the historical priority database to identify abnormal paths whose priority jump exceeds the preset threshold. The historical database shows that the average priority weight of the County Road X207 alternative route in the past 30 days was 0.68. After this adjustment, it jumped to 1.0, with a jump of 47%, exceeding the system preset threshold of 30%. The system automatically triggers the abnormal path detection mechanism, marks the path as a yellow warning state, and associates its jump reason data packet (including real-time traffic control information, task urgency parameters and other verification basis).

[0138] S1053409: Perform task node coverage integrity verification on the abnormal path, and if the verification passes, insert the abnormal path into a specified position of the path recommendation priority sequence.

[0139] In this embodiment, the system verifies the task node coverage integrity of the abnormal path. If the verification is successful, the abnormal path is inserted into the specified position of the path recommendation priority sequence. For the alternative route of County Road X207, the system verifies whether it completely covers the six key nodes required for the transportation task: including two data handover points (longitude 118.765432° and 118.823456°), three curve speed monitoring points and the terminal toll station. The verification results show that the route completely contains all the necessary nodes, and two temporary checkpoints are added (longitude 118.801234° and 118.815678°). After the system confirms that the verification is successful, the alternative route of County Road X207 is inserted at the first position of the path recommendation priority sequence, and the traffic precautions of the newly added checkpoints are marked on the electronic map.

[0140] S1053410: Synchronizing parameters of the path output interface of the trajectory prediction model according to the finally determined path recommendation priority sequence, and locking the target recommended path with the highest priority.

[0141] In this embodiment, the system synchronizes the parameters of the path output interface of the trajectory prediction model according to the final determined path recommendation priority sequence, and locks the target recommended path with the highest priority. The parameter synchronization process includes writing the path node sequence (a total of 58 navigation points), priority weight 1.0 and associated constraints (load capacity 8.5 tons, speed limit 60km / h) of the alternative route of County Road X207 into the output buffer of the trajectory prediction model. The system sends the target recommended path data packet to the driver terminal through the on-board communication module, triggering the navigation interface to automatically switch to the County Road X207 route, and highlighting the first turning prompt point (longitude 118.756789°, 2.3 kilometers from the current location) in the HUD head-up display. During the synchronization process, the system implements a write protection mechanism for the path output interface to prevent unauthorized path modifications before the transportation task is completed.

[0142] As an optional implementation, the method further includes: S201: receiving a fault alarm signal of the target vehicle in real time, and extracting a fault type code and a severity level parameter from the fault alarm signal.

[0143] In this embodiment, the visual management system receives the fault alarm signal of the target vehicle in real time, and extracts the fault type code and severity level parameters in the fault alarm signal. When the target vehicle was performing the transportation task of the engineering materials of Section 3 on November 28, 2023, the vehicle chassis sensor detected that the engine coolant temperature continued to exceed the safety threshold, triggering the fault alarm signal E-2023-OC-015. The system receives a structured data packet containing a fault type code (E-2023-OC-015 indicates a level 3 engine overheating alarm) and a severity level parameter (level L3, requiring immediate disposal) through the on-board communication module, and associates the vehicle operating status data when the fault occurs, including the engine speed of 1820rpm, the coolant temperature of 112℃, and the current position coordinates (longitude 118.756789°, latitude 31.987654°).

[0144] S202: Retrieving a pre-stored emergency response plan according to the fault type code, and generating an emergency response instruction including a backup vehicle dispatch plan and a maintenance station navigation path.

[0145] In this embodiment, the system retrieves the pre-stored emergency handling plan according to the fault type code and generates an emergency response instruction including a spare vehicle dispatch plan and a maintenance station navigation path. After the system parses the fault type code E-2023-OC-015, it activates the emergency plan retrieval module and matches the engine overheating special plan numbered EP-015 from the plan database. The plan includes the emergency handling process of the cooling system, the conditions for enabling the spare vehicle, and the contact information of the maintenance station. The system automatically generates two parallel instructions: dispatching the spare vehicle V-023 (current location longitude 118.723456°) closest to the fault point to take over the transportation task, and planning the emergency path for the target vehicle to the designated maintenance station (longitude 118.765432°) 5 kilometers away.

[0146] S203: Synchronize the emergency response instruction with the visual dynamic management interface in real time, and highlight the affected task nodes and path adjustment areas.

[0147] In this embodiment, the system synchronizes the emergency response instructions with the visual dynamic management interface in real time, and highlights the affected task nodes and path adjustment areas. On the electronic map interface of the engineering dispatch center, the No. 3 task node (longitude 118.801234°) of the original transportation route (County Road X207 route) is marked as a flashing red state, and the dispatch path (blue dotted line) of the spare vehicle V-023 and the target vehicle maintenance path (yellow dotted line) are displayed at the same time. The path adjustment area covers a polygonal area with a radius of 3 kilometers centered on the fault point. The two data handover points in the area (longitude 118.789012° and 118.803456°) are superimposed with orange warning boxes to remind task executors to pay attention to changes in the handover process.

[0148] S204: Activate a remote control protocol according to the severity level parameter to send a speed limit or stop instruction to an actuator of the target vehicle.

[0149] In this embodiment, for the L3 severity level parameters, the system sends a mandatory speed limit command to the target vehicle through the security authentication channel, reducing the vehicle's maximum speed from 60km / h to 30km / h, and triggering the slow-moving mode to stabilize the engine speed below 1500rpm. After detecting that the coolant temperature has dropped to 105°C, the system sends a parking command to control the vehicle to automatically drive into the nearest safe parking area (the emergency lane at longitude 118.758765°), and activates the double flash warning lights and the rear fault sign.

[0150] S205: Recording the execution status of the emergency response instruction and the vehicle response log in the management feedback data.

[0151] In this embodiment, the system creates an emergency response record with event number INC-20231128-015, and continuously writes the real-time position of the standby vehicle V-023 (updated every 30 seconds), the navigation deviation of the target vehicle maintenance path (accumulated 0.2%), and the execution response delay of the remote control command (average 380 milliseconds). The log data includes the engine temperature change curve, the speed limit command trigger timestamp (14:25:37) and the vehicle parking action completion flag (TRUE), forming a structured event chain that can be traced back later.

[0152] As an optional implementation, the step of retrieving a pre-stored emergency response plan according to the fault type code in S202 to generate an emergency response instruction including a backup vehicle dispatch plan and a maintenance station navigation path includes: S2021: Based on the fault type code, a pre-stored emergency response plan database is matched to extract a plan execution condition set and a plan operation instruction sequence in the emergency response plan corresponding to the fault type code.

[0153] In this embodiment, for the fault type code E-2023-OC-015, the system retrieves matching entries from the engine fault classification directory of the plan database, and extracts the plan execution condition set including the ambient temperature threshold (enhanced cooling process is enabled when >35°C), vehicle load limit (autonomous slow driving is allowed when <10 tons), and spare vehicle activation radius (<8 kilometers). The plan operation instruction sequence contains 12 standard operation steps, for example, step 5 is "cut off non-essential electrical loads" and step 8 is "send maintenance station appointment request".

[0154] S2022: Prioritize the plan execution condition set according to the severity level parameter to determine a target emergency response plan that meets the current fault severity level.

[0155] In this embodiment, for the severity level parameter L3, the system filters the plan execution condition set to retain only the condition items related to emergency handling, including immediately activating the standby vehicle (without manual confirmation) and forcibly turning on the engine protection mode (ignoring the driver's operating instructions). The system excludes condition items applicable to level L1 (minor faults), such as "recommend maintenance in the next mission cycle" and "maintain current driving status for observation", and finally generates the target emergency handling plan EP-015-L3 adapted to the severity level L3.

[0156] S2023: Extracting the vehicle identification set, dispatch path planning parameters and task handover node information of the standby vehicle dispatch plan from the target emergency response plan, and generating a standby vehicle dispatch instruction.

[0157] In this embodiment, the system searches the spare vehicle database, selects the vehicle V-023 (10-ton load capacity, equipped with shockproof cargo box) that meets the task requirements, and plans its dispatch path from the current position (longitude 118.723456°) to the faulty vehicle stop (longitude 118.758765°), which contains 7 navigation nodes (longitude 118.730123° to 118.752345°). The task handover node is set at the emergency lane 500 meters downstream of the fault point (longitude 118.756789°), and the system generates a dispatch instruction data packet containing the vehicle identification V-023, the path node sequence, and the handover time window (14:40-14:45).

[0158] S2024: synchronously acquiring the real-time positioning data and the maintenance station location information of the target vehicle, and combining the maintenance station location information with the real-time positioning data to generate a path node sequence and an estimated arrival time parameter of a maintenance station navigation path.

[0159] In this embodiment, the system takes the target vehicle stop point (longitude 118.758765°) as the starting point, calculates the shortest path to the designated maintenance station (longitude 118.765432°), and generates a path node sequence containing 3 key turning points (longitude 118.760123° left turn, 118.762345° straight, 118.764567° right turn). Combined with real-time traffic data, the system predicts the arrival time to be 14:55±3 minutes, and associates the maintenance station workstation reservation status (Lift No. 2 is available until 15:30).

[0160] S2025: Encapsulate the standby vehicle dispatch instruction and the maintenance station navigation path in an instruction format to generate an emergency response instruction including a dispatch instruction trigger timestamp and a path navigation verification code.

[0161] In this embodiment, the system uses TLV (Tag-Length-Value) format to encapsulate instruction data, where the dispatch instruction part contains the trigger timestamp 20231128142537 (year-month-day hour-minute-second) and the MAC address 00:1A:3B:4C:5D:6E of the vehicle identification V-023. The maintenance station navigation path part is embedded with the SHA-256 checksum 7D2A45...B9C1 to ensure the data integrity during the transmission process. The size of the encapsulated emergency response instruction is controlled within 512 bytes, which meets the transmission protocol requirements of the vehicle communication module.

[0162] S2026: Verify the task link integrity of the emergency response instruction according to the task handover node information, and identify the conflicting nodes between the backup vehicle scheduling plan and the current task execution information.

[0163] In this embodiment, the system compares the list of materials to be handed over (12 boxes of construction drawings of Section 3, 5 sets of testing instruments) with the cargo box volume parameters of the spare vehicle V-023 (18 cubic meters), and verifies that the loading space is sufficient (15.6 cubic meters are required). At the same time, it is detected that there is a 0.05-meter safety margin between the height-restricted facilities in the original task path (3.5-meter height-restricted frame on County Road X207) and the height of the spare vehicle V-023 (3.45 meters), marking it as a potential conflict node but maintaining the validity of the solution.

[0164] S2027: Dynamically replan the dispatch path planning parameters in the emergency response instruction based on the conflicting nodes, and update the path node sequence of the maintenance station navigation path.

[0165] In this embodiment, when it is detected that there is temporary road construction (pipeline maintenance) at the node of longitude 118.762345° in the original planned path, the system starts the dynamic path replanning module and adjusts the path node sequence to turn right at longitude 118.761234° to bypass the auxiliary road. The updated path increases the driving distance by 0.8 kilometers, but avoids the construction area. The estimated arrival time is corrected to 14:58±2 minutes, and the voice prompt parameter "Pay attention to the auxiliary road entrance sign" is added to the navigation instruction.

[0166] S2028: Re-encapsulate the updated path node sequence and the estimated arrival time parameter into the emergency response instruction to generate an emergency response instruction to be executed.

[0167] In this embodiment, the system updates the path node number field (increased from 3 to 4) in the TLV format data packet, adds the latitude and longitude coordinates of the auxiliary road turning point (118.761234°, 31.976543°), and recalculates the SHA-256 checksum to A3D5F1...8E9C2. The estimated arrival time parameter is synchronously updated to 14:58±2 minutes, and the secondary time synchronization is performed with the maintenance station reservation system to confirm that the reservation period for lift No. 2 is extended to 15:35.

[0168] S2029: associate the to-be-executed emergency response instruction with the fault type code and embed it into the execution queue of the plan operation instruction sequence.

[0169] In this embodiment, the globally unique event association code generated by the system can be INC-20231128-015-EOC015, which is written into the metadata area of ​​the emergency response instruction header (Header Section) and the preplan operation instruction sequence at the same time. The preplan operation instruction sequence is sorted by execution priority, with priority given to "activating the spare vehicle" (instruction priority 1) and "sending a maintenance request" (instruction priority 2), followed by secondary operations such as "monitoring engine temperature" (instruction priority 3), forming an instruction execution queue with a weight identifier.

[0170] As an optional but non-limiting embodiment, the step of rendering the event details panel in conjunction with the visual dynamic management interface in S10453 to update and display the extended information layer of the abnormal event includes: (1) In response to a user's touch operation in the interactive event annotation window, a sensor raw data set associated with the key event marker is retrieved, where the sensor raw data set includes vehicle acceleration sensor data, engine status parameters, and ambient temperature and humidity records.

[0171] In this step, when the target vehicle triggers the engine overheating abnormal event mark at 14:25 on November 28, 2023, the system detects the user's operation instruction to double-click the mark on the electronic map interface, and immediately extracts the sensor raw data set 3 minutes before and after the event from the on-board data storage module. The set includes the XYZ axial vibration value recorded by the three-axis acceleration sensor (peak value up to 2.3g), the coolant temperature curve transmitted by the engine control unit (from 89℃ to 112℃), and the temperature and humidity data collected by the cockpit environment sensor (temperature 32℃ / humidity 65%). All data are marked with millisecond timestamps, forming a raw data set containing 487 sets of time series data.

[0172] (2) performing timestamp alignment and noise filtering on the sensor raw data set to generate a standardized event data stream corresponding to the abnormal event marker; dividing the data parsing channels according to the parameter types of the standardized event data stream to extract the vehicle operation fault code sequence, the environmental abnormal parameter fluctuation range and the driver operation log fragment respectively.

[0173] In this step, the system uses a sliding window average algorithm to smooth the acceleration sensor data and eliminate the instantaneous noise caused by road bumps (such as the 2.3g peak recorded at 14:25:37 was corrected to 1.8g). The timestamp alignment module interpolates and synchronizes the engine status parameters (sampling frequency 10Hz) with the ambient temperature and humidity records (sampling frequency 1Hz) to generate a standardized event data stream with a unified time base. The data analysis channel division module creates three parallel processing channels according to the parameter type: the vehicle operation fault code analysis channel extracts the DTC P0217 fault code sequence stored in the engine control unit, the environmental abnormal parameter analysis channel calculates the abnormal range of temperature fluctuation standard deviation up to 4.3℃, and the driver operation analysis channel intercepts the operation log fragment with abnormal sudden change of clutch pedal travel.

[0174] (3) Pattern matching is performed on the vehicle operation fault code sequence with a pre-stored fault knowledge base to generate a fault type label and a maintenance suggestion summary.

[0175] In this step, after parsing the DTC P0217 code, the system matches the three-level associated fault mode under the engine cooling system classification in the fault knowledge base: reduced efficiency of the coolant circulation pump (matching degree 87%), radiator fin blockage (matching degree 72%), and thermostat valve body stagnation (matching degree 65%). Based on the multi-mode weighted evaluation, the system generates the main fault label "Insufficient efficiency of the coolant circulation system (code F-OC-015)" and outputs a maintenance suggestion summary: immediately replenish the coolant to the standard liquid level, check the operating current of the circulation pump, and clean the radiator surface attachments. The suggestion summary is associated with the standard maintenance man-hour data (estimated to take 45 minutes) and the spare parts replacement list (OEM part number WP-2023-015).

[0176] (4) Compare the fluctuation range of the abnormal environmental parameters with the historical environmental data samples for similarity, and generate a probability distribution map of environmental influencing factors.

[0177] In this step, the system retrieved environmental data samples of the same section of road (County Road X207 K3+200 to K5+700) in the past 30 days, and found that the ambient temperature during the period of 14:20-14:30 on that day was 5.2°C higher than the historical average, and the humidity dropped by 23%. The Kolmogorov-Smirnov test calculated that the probability of temperature anomaly was 92%, and the system generated a normal distribution curve showing that the probability of engine overheating risk at an ambient temperature of 32°C was 3.8 times higher than the normal temperature (27°C). The probability distribution diagram marked the key threshold points, and when the ambient temperature exceeded 30°C, the cooling system load entered the yellow warning interval.

[0178] (5) Inputting the driver operation log fragment into the operation behavior analysis model, and outputting the operation compliance score and the timestamp location result of the abnormal operation action.

[0179] In this step, when the model analyzed the clutch operation data, it detected three consecutive semi-clutch operations (the pedal travel was maintained in the range of 40%-45% for more than 3 seconds) from 14:25:21 to 14:25:24, which violated Article 5.2 of the standard operating specification. After comparing the operation behavior analysis model with the standard operation template, it output an operation compliance score of 76.3 points (out of 100) and marked three abnormal operation timestamps on the timeline (14:25:21.345, 14:25:22.112, 14:25:23.887). The model also recognized that the throttle opening was maintained at 82% during the abnormal operation, which exceeded the 70% upper limit recommended for this slope section.

[0180] (6) Mapping the fault type label, the maintenance suggestion summary, the probability distribution diagram, the operation compliance score, and the timestamp location result to multiple independent display areas of the event details panel according to event attribute classification.

[0181] In this step, the event details panel is divided into five functional areas: the upper left area displays the color-coded fault type label (red F-OC-015), the upper right area embeds a probability distribution chart in the form of a three-dimensional bar chart, the middle left area scrolls to display the maintenance suggestion items, the middle right area displays the operation compliance score progress bar, and the bottom timeline synchronously highlights the three abnormal operation timestamps. The visual elements of each display area are bound to the database fields. For example, the third item of the maintenance suggestion summary, "Clean the surface attachment of the radiator", is associated with the operation and maintenance knowledge base entry KB-2023-0876.

[0182] (7) Determining the dynamic floating position and hierarchical stacking order of the event details panel in the visual dynamic management interface according to the spatial coordinates of the key event mark in the electronic map.

[0183] In this step, based on the coordinates of the target vehicle abnormal event trigger point (longitude 118.758765°, latitude 31.976543°), the system calculates the screen pixel coordinates of the point in the current map view (X=435, Y=287). The event details panel uses a dynamic anchor algorithm to position its upper left corner at (X+20, Y-50) pixels, avoiding the original map annotation information. The layer overlay order is set to Z-index 999 to ensure that the panel always floats above the base track layer (Z-index 500) and the real-time location marker layer (Z-index 800).

[0184] (8) Performing transparency fusion processing on the display area content of the event details panel and the layer data of the electronic map to generate a semi-transparent extended information layer superimposed on the basic track overlay layer.

[0185] In this step, the system uses an alpha blending algorithm to set the transparency of the event details panel background to 70%, while retaining the opacity of text and chart elements. The map road name annotation layer (Z-index 600) and the extended information layer perform depth buffer detection to ensure that important geographic information such as the County Road X207 label is not obscured. During the fusion process, the system detects the map zoom level in real time, and automatically reduces the panel transparency to 50% when the zoom ratio exceeds 1:5000 to improve the visibility of the underlying map.

[0186] (9) Embed a slidable timeline control in the extended information layer according to the timestamp sequence of the abnormal event mark, and establish a time synchronization association relationship between the timeline control and the content displayed on the event details panel.

[0187] In this step, the timeline control covers the period from 14:24:30 to 14:26:30, with a total length of 480 pixels corresponding to 120 seconds (4 pixels / second). The initial position of the slider is located at the abnormal event trigger point 14:25:37. When the user drags the slider to 14:25:21, the middle area of ​​the event details panel synchronously displays the engine speed (1820rpm) and coolant temperature (98℃) at that moment. Time synchronization association is achieved through the message bus mechanism, and the slider position change event triggers a dynamic refresh request for sensor data.

[0188] (10) In response to a user sliding operation on the timeline control, the time series segments of the sensor raw data set and the corresponding fault type labels displayed in the extended information layer are updated in real time.

[0189] In this step, when the user moves the slider to the 14:25:15 position, the system extracts the data window of 500 milliseconds before and after this moment from the standardized event data stream: the acceleration Z-axis value increases from 1.2g to 1.7g, the coolant temperature increases from 93℃ to 96℃, and the ambient temperature is recorded as 31.5℃. The fault type label is dynamically updated according to the temperature change rate, and the secondary alarm information of "Coolant temperature rise abnormal (code T-RA-002)" is displayed. When the time series segment is updated, the system generates a data comparison prompt at the bottom of the event details panel, marking the percentage difference between the current segment parameters and the event peak parameters.

[0190] (11) Dynamically adjust the background color warning level of the driver operation log segment in the extended information layer according to the operation compliance score, and render it in color synchronously with the mark icon of the corresponding spatial coordinates in the electronic map.

[0191] In this step, the operation compliance score of 76.3 points triggers the yellow warning level (threshold range 70-85 points), and the background color of the driver operation log area changes to #FFFF99. The abnormal event mark icon on the electronic map is synchronously switched to a yellow triangle, and the warning description is updated in the map legend column. The color rendering module uses the HSV color space conversion algorithm to ensure that the color difference between different devices is controlled within the industrial standard range of ΔE<3.0.

[0192] (12) Loading a maintenance site location mark associated with the maintenance suggestion summary in an edge area of ​​the extended information layer, and generating an optimal path navigation line from the key event mark location to the maintenance site location.

[0193] In this step, the system analyzes the "Recommended Maintenance Site" field in the maintenance suggestion summary and locates the authorized maintenance center 5.2 kilometers away from the event point (longitude 118.765432°). The optimal path planning module calculates two candidate paths: the direct route of County Road X207 (5.2 kilometers / estimated 11 minutes) and the detour route of Township Road Y305 (6.1 kilometers / estimated 13 minutes). The system selects the shortest path and draws a blue dotted navigation line on the electronic map. The path nodes include 3 turning points (longitude 118.760123° left turn, 118.762345° straight, 118.764567° right turn).

[0194] (13) Conflict detection is performed on the optimal path navigation line and the historical driving trajectory in the basic trajectory overlay layer. If it is detected that there is an unfinished task node in the path overlap area, a path detour suggestion prompt box is inserted in the extended information layer.

[0195] In this step, the system performs conflict detection between the optimal path navigation line and the historical driving trajectory in the basic trajectory overlay layer. If it is detected that there is an unfinished task node in the path overlap area, a path detour suggestion prompt box is inserted in the extended information layer. The conflict detection module found that there were two delay records due to bridge construction at K4+500 (longitude 118.762345°) of County Road X207 in the historical trajectory. The system pops up a prompt box in the lower right corner of the extended information layer, suggesting to detour to Township Road Y305 to avoid this section. The prompt box is embedded with comparison data: the detour plan increases 0.9 kilometers but saves 7 minutes. After the user confirms the prompt, the optimal path navigation line is automatically updated to the Township Road Y305 route, and a right turn instruction at longitude 118.756789° is added to the path node sequence.

[0196] (14) Rendering the abnormal thermal coverage area of ​​the environment in the extended information layer according to the probability distribution map, and performing spatial overlay analysis on the thermal coverage area and the congested road sections in the real-time traffic data to generate an environmental risk warning level mark.

[0197] In this step, the Gaussian kernel density estimation algorithm is used to generate a temperature anomaly distribution map with a radius of 2 kilometers with the event point as the center. The red highlight area (temperature>31℃) covers the K3+200 to K5+700 section of County Road X207. Spatial overlay analysis shows that the overlap rate between this area and the real-time traffic congestion section (K4+100 to K4+800) is 73%. The system generates a three-level warning sign: high temperature environment aggravates the risk of congestion (code ER-015), and it is recommended that subsequent tasks avoid this section during the 12:00-15:00 period.

[0198] (15) In the layer update event of the visual dynamic management interface, the floating position coordinates of the extended information layer and the transparency parameters of the semi-transparent overlay effect are synchronously refreshed to ensure visual consistency with the underlying map.

[0199] In this step, when the map view switches from 1:5000 to 1:2000 due to user zoom operation, the system recalculates the screen coordinates of the event details panel (X=702, Y=415) and adjusts the transparency parameter to 65% to meet the needs of higher-precision map display. During the layer refresh process, the system maintains the spatial correspondence between the extended information layer and the real-time vehicle position mark. When the target vehicle moves to longitude 118.759876°, the panel automatically translates to maintain a 20-pixel offset from the event point.

[0200] See also Figure 2 As shown, this figure is a schematic diagram of the basic structure of a visual management system 200 provided by an embodiment of the present invention. The visual management system 200 includes: Processor 201; a storage device 202 on which a computer program 2020 is stored; When the computer program 2020 is executed by the processor 201, the processor 201 implements any of the bus travel visualization management methods combined with positioning technology.

[0201] Based on the above, a readable storage medium is provided, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the above method are implemented.

[0202] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the system or device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

Claims

1. A bus travel visualization management method combined with positioning technology, characterized in that: include: Acquire vehicle real-time positioning data of the target vehicle, and extract task execution information associated with the vehicle real-time positioning data, wherein the vehicle real-time positioning data includes positioning coordinates and motion parameters of a continuous time series; Performing dynamic trajectory fitting processing on the real-time positioning data of the vehicle to generate a driving trajectory map matching the task execution information, wherein the driving trajectory map includes timestamps of path nodes and key event markers; Perform multi-dimensional analysis on the driving trajectory map based on preset task rules to determine the driving state characteristics and task completion index of the target vehicle; Performing spatiotemporal correlation mapping of the driving state characteristics, the task completion index and the driving trajectory map to generate a visual dynamic management interface; When the visual dynamic management interface is in operation, in response to a tracing instruction triggered by a management terminal, the historical trajectory database is called to perform backtracking optimization processing on the driving trajectory map, and management feedback data including abnormal event annotations and path comparison results is output.

2. The method according to claim 1, characterized in that The performing dynamic trajectory fitting processing on the real-time positioning data of the vehicle to generate a driving trajectory map matching the task execution information includes: According to the set of planned path points in the task execution information, the real-time positioning data of the vehicle is filtered by time stamp alignment to select candidate positioning points that meet the preset time window; Performing a path continuity check on the candidate positioning points, using an interpolation algorithm to complete the missing intermediate positioning points, and generating a corrected continuous driving path; Extracting abnormal positioning points in the continuous driving path that match preset key event trigger conditions, and associating and marking the abnormal positioning points with task nodes in the task execution information; The trajectory prediction model is called to perform driving mode recognition on the continuous driving path, divide the path into acceleration segments, uniform speed segments and stagnant segments, and generate a driving trajectory map containing segmented feature descriptions.

3. The method according to claim 2, characterized in that The extracting of abnormal positioning points in the continuous driving path that match preset key event triggering conditions includes: Obtaining instantaneous speed fluctuation value, direction deviation angle and dwell time parameters in the real-time positioning data of the vehicle; Input the instantaneous speed fluctuation value, the direction deviation angle and the dwell time parameter into a target multimodal event classification model, and output the abnormal event type and confidence score; Extracting a set of abnormal positioning points that meet a threshold condition from the continuous driving path according to the abnormal event type and the confidence score; The abnormal positioning point set is matched with similar events in the historical trajectory database to generate a key event marker containing an event association identifier.

4. The method according to claim 1, characterized in that The multi-dimensional analysis of the driving trajectory map based on the preset task rules to determine the driving state characteristics and task completion index of the target vehicle includes: Extracting task priority weights, planned completion times, and preset path constraints from the task execution information; Calculating the deviation index between the actual driving path and the planned path according to the path node density distribution in the driving trajectory map; Determine a dynamic score of the task completion index by combining the deviation index and the task priority weight; Performing time series analysis on key event markers in the driving trajectory map to generate a status assessment report including event impact coefficients; The dynamic score and the status assessment report are compared with the acceptance conditions in the preset task rules, and the compliance determination result of the driving status characteristics is output.

5. The method according to claim 4, characterized in that The time series analysis of the key event markers in the driving trajectory map is performed to generate a status assessment report containing event impact coefficients, including: Extracting the starting timestamp, duration and spatial distribution density corresponding to the key event marker; At least based on the start timestamp and the duration, generate an event impact propagation model based on the time axis, and calculate the correlation strength coefficient between adjacent key events; generating an event chain reaction probability map according to the association intensity coefficient and the spatial distribution density; Inputting the event chain reaction probability graph into the risk prediction model, and outputting a quantitative evaluation value of the event impact coefficient; The abnormal events in the status assessment report are prioritized based on the quantitative assessment values.

6. The method according to claim 1, characterized in that The step of performing spatiotemporal correlation mapping of the driving state characteristics, the task completion index and the driving trajectory map to generate a visual dynamic management interface includes: Extracting a set of geographic coordinates and a time stamp sequence of path nodes from the driving trajectory map; Mapping the geographic coordinate set and the timestamp sequence to layer data of an electronic map to generate a basic track overlay layer; Overlaying a speed heat map of the driving state characteristics and a progress indicator of the task completion index on the basic trajectory overlay; embedding an interactive event annotation window in the electronic map according to the spatial position of the key event mark; The basic track overlay layer, the speed heat map, the progress indicator and the interactive event annotation window are integrated to generate a multi-layered visual dynamic management interface.

7. The method according to claim 6, characterized in that The generation of a multi-layer superimposed visual dynamic management interface includes: In response to a user's touch operation on the interactive event annotation window, retrieving sensor raw data associated with the key event mark; Parsing the raw data of the sensor in real time to generate an event detail panel including vehicle fault codes, environmental parameters and operation logs; Linking the event details panel with the visual dynamic management interface for rendering, and updating the extended information layer showing the abnormal event; According to the comparison mode selected by the user, the historical driving trajectory map of the same task is loaded from the historical trajectory database; The path difference analysis results of the current driving trajectory map and the historical driving trajectory map are displayed in parallel in the extended information layer.

8. The method according to claim 1, characterized in that The calling of the historical trajectory database to perform backtracking optimization processing on the driving trajectory map includes: According to the time range parameter in the tracing instruction, extract the historical trajectory fragments of the corresponding time period from the historical trajectory database; Calculating trajectory smoothness and matching path overlap on the historical trajectory segments to identify repetitive driving patterns; Building a trajectory prediction model based on the repetitive driving pattern to generate a task path optimization suggestion for a target period; Simulate and superimpose the task path optimization suggestion for the target period with the current driving trajectory map, and output management feedback data including path comparison results; According to the distribution density of the abnormal event annotations, adjusting the weight parameters of the trajectory prediction model to update the historical trajectory database; The constructing of a trajectory prediction model based on the repetitive driving pattern comprises: Extracting periodic path node sets and travel time intervals in historical trajectory segments; Using a time series clustering algorithm to classify the periodic path node set into patterns, and generating a typical driving pattern template; The driving time interval is integrated with the real-time traffic status data to calculate the time cost coefficient of each typical driving mode; Dynamically adjusting the path recommendation priority of the trajectory prediction model according to the time cost coefficient and the task completion index; embedding the dynamically adjusted path recommendation priority into the path comparison result in the management feedback data; The dynamically adjusting the path recommendation priority of the trajectory prediction model according to the time cost coefficient and the task completion index includes: Acquire a set of time consumption parameters of each typical driving mode corresponding to the time cost coefficient, and a set of path deviation scores in the task completion index; Performing weight allocation processing on the time consumption parameter set according to the path deviation score set to generate a time cost combination weight based on task completion; Inputting the time cost combination weight into a priority sorting algorithm, performing initial priority sorting on the candidate path set pre-stored in the trajectory prediction model, and generating a path priority queue with a weight identifier; Based on the weight identifier of each candidate path in the path priority queue, extracting a subset of candidate paths that match the preset path constraint condition in the current task execution information; Performing real-time traffic status data adaptability check on each candidate path in the candidate path subset, and eliminating invalid paths with traffic congestion marks or temporary control events; According to the time cost combination weights of the remaining candidate paths and the task completion degree deviation threshold, the path priority adjustment interval that meets the dynamic adjustment conditions is determined; Within the path priority adjustment interval, priority remapping is performed on the remaining candidate paths in descending order of time-cost combination weights to generate an updated path recommendation priority sequence; Compare the updated path recommendation priority sequence with the priority records of similar tasks in the historical priority database to identify abnormal paths whose priority jump amplitude exceeds a preset threshold; Performing task node coverage integrity verification on the abnormal path, and inserting the abnormal path into a specified position of the path recommendation priority sequence if the verification passes; The path output interface of the trajectory prediction model is synchronized with parameters according to the finally determined path recommendation priority sequence, and the target recommended path with the highest priority is locked.

9. The method according to claim 1, characterized in that The method further comprises: Receiving the fault alarm signal of the target vehicle in real time, and extracting the fault type code and severity level parameter in the fault alarm signal; Retrieving a pre-stored emergency response plan according to the fault type code, and generating an emergency response instruction including a backup vehicle dispatch plan and a maintenance station navigation path; The emergency response instruction is synchronized with the visual dynamic management interface in real time, and the affected task nodes and path adjustment areas are highlighted; activating a remote control protocol according to the severity level parameter to send a speed limit or parking instruction to an actuator of the target vehicle; recording the execution status of the emergency response instruction and the vehicle response log in the management feedback data; The method of retrieving a pre-stored emergency response plan according to the fault type code and generating an emergency response instruction including a backup vehicle dispatch plan and a maintenance station navigation path includes: Based on the fault type code, a pre-stored emergency response plan database is matched, and a plan execution condition set and a plan operation instruction sequence in the emergency response plan corresponding to the fault type code are extracted; Prioritize the plan execution condition set according to the severity level parameter to determine a target emergency response plan that meets the current fault severity level; Extract the vehicle identification set, dispatch path planning parameters and task handover node information of the standby vehicle dispatch plan from the target emergency response plan, and generate a standby vehicle dispatch instruction; Synchronously acquiring the real-time positioning data and the location information of the maintenance station of the target vehicle, and combining the location information of the maintenance station with the real-time positioning data to generate a path node sequence and an estimated arrival time parameter of a navigation path of the maintenance station; Encapsulating the standby vehicle dispatch instruction and the maintenance station navigation path in an instruction format, and generating an emergency response instruction including a dispatch instruction trigger timestamp and a path navigation verification code; Perform task link integrity verification on the emergency response instruction according to the task handover node information, and identify conflicting nodes between the backup vehicle scheduling plan and the current task execution information; Dynamically replanning the dispatch path planning parameters in the emergency response instruction based on the conflicting nodes, and updating the path node sequence of the maintenance station navigation path; Re-encapsulating the updated path node sequence and the estimated arrival time parameter into the emergency response instruction to generate an emergency response instruction to be executed; The emergency response instruction to be executed is associated with the fault type code and identified as an instruction, and is embedded in the execution queue of the plan operation instruction sequence.

10. A visual management system, characterized in that: include: processor; A storage device having a computer program stored thereon, wherein when the computer program is executed by the processor, the processor implements the bus travel visualization management method combined with positioning technology as described in any one of claims 1-9.

Images