Road vehicle overrun dynamic monitoring management method and system

Abstract

The invention discloses a road vehicle overrun dynamic monitoring management method and system, and relates to the technical field of vehicle monitoring. The method comprises the following steps: firstly, on the basis of collected road vehicle data, synthesizing vehicle passing frequency of each specified road section area to obtain a whole road network flow thermodynamic diagram, and performing road traffic management; then, in the road traffic management process, according to the vehicle weighing data corresponding to the dispatching dynamic weighing section, in combination with the vehicle image identification data in the preset monitoring time period, carrying out overrun state monitoring, and finally, based on the overrun state monitoring result, in combination with the vehicle image monitored by the road network, carrying out road traffic management. Track monitoring is performed on the driving track of the specified vehicle, load and risk road sections are determined through traffic management, dynamic resource scheduling is realized, and meanwhile, the accuracy and reliability of overrun state monitoring are improved by combining cross validation of dynamic weighing data and vehicle identification data. The problem that the accuracy of vehicle feature recognition monitoring management is low is effectively solved.

Description

Technical Field

[0001] This invention relates to the field of vehicle monitoring technology, and in particular to a method and system for dynamic monitoring and management of overloaded highway vehicles. Background Technology

[0002] Highways, characterized by high speeds and large traffic volumes, demand real-time, accurate, and collaborative monitoring of overloaded vehicles. To address this, the highway overload dynamic monitoring and management system has upgraded from basic identification to precise supervision through the integrated application of multiple levels and technologies. In terms of hardware deployment, monitoring stations integrating dynamic identification and high-definition image acquisition are deployed at key nodes along highways, simultaneously collecting multi-dimensional information such as vehicle weight, dimensions, and license plates. In terms of technical architecture, the computing power of edge computing nodes is enhanced, and some data processing and transmission are moved to the front end to reduce cloud transmission pressure and improve response speed.

[0003] To adapt to the high-requirement monitoring scenarios of highways, Chinese invention patent CN117351439B discloses a smart highway overloaded vehicle dynamic monitoring and management system. This system includes: an image acquisition section, a vehicle size detection section, a license plate recognition section, and a monitoring and management section. The image acquisition section acquires images of vehicles traveling on the target highway; the vehicle size detection section identifies the vehicle body image to determine if the vehicle exceeds the weight limit; the license plate recognition section identifies the license plate based on the image when the vehicle size detection section determines the vehicle is overloaded; and the monitoring and management section obtains the vehicle's corresponding contact information based on the license plate number, automatically sends overload warning messages, and adds overload records.

[0004] For example, Chinese invention patent application CN114863398A discloses a vehicle overload detection method and terminal, and a vehicle overload detection system, including: acquiring a detection image of a target vehicle in the overload detection area of ​​a highway intersection, and determining whether the target vehicle exceeds the length, height, and width limits based on the detection image; if the target vehicle exceeds the length, height, and width limits, acquiring the weight information of the target vehicle, extracting the license plate information of the target vehicle from the detection image, determining the weight threshold of the target vehicle based on the license plate information, and determining whether it is overweight based on the weight information and the threshold.

[0005] With the widespread application of smart highway technology in modern traffic management, the ability to monitor overloaded vehicles through dynamic weighing, real-time path tracking, and cross-regional collaborative enforcement has become a core element in ensuring road network safety and operational efficiency. Overload monitoring on highways is a fundamental aspect of ensuring road network safety and traffic efficiency. Its dynamic monitoring system typically consists of front-end sensing devices, edge computing nodes, a monitoring center, and a cloud-based monitoring platform, forming a closed-loop management process integrated with road network monitoring. However, due to significant differences in the resource capabilities and deployment locations of various detection devices, as well as varying communication link stability, vehicle traffic data exhibits complex multi-source collection and multi-path transmission relationships between different levels of nodes.

[0006] However, in actual monitoring, ordinary roads face more complex problems than expressways. They not only experience constantly changing traffic flow but also dynamic situations such as vehicles traveling across regions, complex route selection, and sudden traffic peaks. In these complex operational scenarios, such as during peak periods on key freight routes, sudden changes in traffic flow lead to a backlog of weighing data. Pre-set monitoring strategies are limited by the pre-defined monitoring point layout during the deployment phase, making it impossible to adaptively adjust detection sensitivity according to real-time traffic load or optimize monitoring resource distribution for route changes. This results in fluctuations in monitoring coverage and data processing efficiency, leading to low accuracy in vehicle feature recognition and monitoring management. Summary of the Invention

[0007] To address the issue of low accuracy in existing vehicle feature recognition and monitoring management technologies, this invention provides a method and system for dynamic monitoring and management of overloaded vehicles on highways. The technical solution is as follows:

[0008] On the one hand, a dynamic monitoring and management method for overloaded highway vehicles is provided. The method includes: S1, based on the collected highway vehicle data, a comprehensive traffic flow heat map of the entire road network is obtained by integrating the vehicle passage frequency of each designated road section area, and highway traffic volume management is carried out; S2, in the process of highway traffic volume management, overload status monitoring is carried out according to the vehicle weighing data corresponding to the scheduling dynamic weighing section and combined with vehicle image recognition data within a preset monitoring time period; S3, based on the results of overload status monitoring and combined with the vehicle images monitored by the road network, trajectory monitoring is carried out on the driving trajectory of designated vehicles.

[0009] On the other hand, a dynamic monitoring and management system for overloaded highway vehicles is provided. This system applies a dynamic monitoring and management method for overloaded highway vehicles. The system includes: a highway traffic volume management module, an overload status monitoring module, and a trajectory monitoring module. The highway traffic volume management module is used to obtain a comprehensive traffic flow heat map of the entire road network based on the collected highway vehicle data and the vehicle passage frequency of each designated road section area, and to manage highway traffic volume. The overload status monitoring module is used to monitor the overload status based on the vehicle weighing data corresponding to the dispatch dynamic weighing section and combined with vehicle image recognition data within a preset monitoring time period. The trajectory status monitoring module is used to monitor the trajectory of designated vehicles based on the results of the overload status monitoring and combined with the vehicle images monitored by the road network.

[0010] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0011] 1. By integrating vehicle traffic data to construct a traffic flow heat map for the entire road network, quantitative and visual management of traffic load is achieved. Through vehicle traffic frequency statistics within a specified time window, road segment frequency analysis, and load index calculation, combined with normalization processing, a standard load rate is generated, thereby accurately dividing different load level intervals. This processing method no longer relies on data from a single monitoring point, but integrates traffic information from multiple road segments and time periods across the entire road network. Through multi-dimensional indicators such as load standard deviation, range, and coefficient of variation, it comprehensively quantifies the stability of traffic conditions and the cumulative load effect. Management strategies can be dynamically adjusted based on the color coding presented in the heat map, improving the smoothness of traffic operation across the entire road network and the efficiency of management resource utilization.

[0012] 2. The design of the mileage ratio analysis stage enables adaptive dynamic allocation of monitoring resources. The mileage ratio deviation rate is introduced as a core indicator for resource allocation. By comparing the total mileage of congested road sections with the total mileage of the entire road network, and combining this with a statistical baseline constructed from historical data, the scope and degree of road network congestion are accurately determined. Resources are flexibly allocated based on real-time congestion conditions. By tilting resources towards high-load areas, the monitoring accuracy and response speed of congested road sections are effectively improved, while resource redundancy in low-load areas is avoided. This achieves optimal allocation of regulatory resources, enhances the overall effectiveness of road network supervision, and ensures that limited management resources play their maximum role.

[0013] 3. In highway traffic flow correlation analysis, a processing method combining adjacent road segment data linkage and time series inference is adopted to achieve early prediction and intervention of traffic load. By obtaining the vehicle passage frequency of a specified road segment and adjacent road segments, the trend of road segment load change is accurately predicted. The monitoring data of a single road segment is correlated with the surrounding road network data to build a regional traffic flow analysis network. This can identify potential congestion risks in advance. When the standard load rate corresponding to the predicted vehicle passage frequency is about to enter the high load range, it can promptly prompt staff to take intervention measures. This effectively solves the problem of congestion spread caused by sudden traffic flow in existing technologies and achieves source control of traffic congestion.

[0014] 4. In the monitoring of overload conditions, we innovatively achieved deep integration of dynamic weighing data and vehicle image recognition data. Through cross-validation of multi-source data, we ensured the accuracy of overload detection. We refined the voltage-time series signal of the dynamic weighing section, and generated a pressure curve through discretization, interpolation smoothing, dimensional conversion and filtering. Combined with peak analysis, we extracted key parameters such as wheel axle pressure area and inter-peak time interval to construct a complete wheel axle data set. Through precise processing of weighing signals and cross-validation of image recognition, we improved the effectiveness of wheel axle data and the accuracy of overload judgment, ensuring the accurate identification and location of overloaded vehicles.

[0015] 5. This solution achieves comprehensive overload control through intelligent traffic volume management, precise monitoring of overloaded vehicles, dynamic resource allocation, and full-process trajectory tracking. Utilizing multi-module technology collaboration and multi-source data fusion, and supported by a full-network traffic heatmap, it integrates functions such as mileage ratio analysis, traffic correlation analysis, multi-source verification of overloaded vehicles, and dynamic trajectory tracking. This provides all-round coverage from macro-level perception of the road network to micro-level monitoring of vehicle behavior. Each stage possesses dynamic adaptive capabilities, allowing for flexible strategy adjustments based on real-time road network changes. This effectively solves problems such as insufficient monitoring coverage leading to low vehicle feature recognition accuracy in complex scenarios like road sections with sudden traffic flow changes. It enhances the intelligence, precision, and collaboration of dynamic monitoring and management of overloaded vehicles on highways, providing comprehensive technical support for the safe and efficient operation of the road network. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating a dynamic monitoring and management method for overloaded highway vehicles provided in an embodiment of the present invention;

[0018] Figure 2 A flowchart corresponding to highway traffic volume management provided in this embodiment of the invention;

[0019] Figure 3 The flowchart corresponding to the over-limit state monitoring provided in the embodiments of the present invention;

[0020] Figure 4 The flowchart corresponding to the trajectory status monitoring provided in the embodiments of the present invention;

[0021] Figure 5 This is a schematic diagram of the structure of a dynamic monitoring and management system for overloaded highway vehicles provided in an embodiment of the present invention. Detailed Implementation

[0022] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0023] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0024] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0025] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0026] This invention provides a method for dynamic monitoring and management of overloaded highway vehicles, such as... Figure 1 The flowchart shown is a method for dynamic monitoring and management of overloaded highway vehicles. The processing flow of this method may include the following steps: S1. Based on the collected highway vehicle data, a comprehensive traffic flow heat map of the entire road network is obtained by integrating the vehicle passage frequency of each designated road segment area, and highway traffic volume management is carried out based on the obtained comprehensive road network traffic flow heat map; S2. In the process of highway traffic volume management, overload status monitoring is carried out according to the vehicle weighing data corresponding to the scheduling dynamic weighing section and combined with the vehicle image recognition data within the preset monitoring time period; S3. Based on the results of overload status monitoring and combined with the vehicle images monitored by the road network, the trajectory monitoring of the designated vehicles is carried out.

[0027] In this embodiment, a full-network traffic heatmap generated by integrating highway vehicle data from various road sections clearly presents the traffic load distribution in different areas, providing an intuitive and comprehensive basis for traffic volume management decisions. This helps managers adjust management strategies in a targeted manner and optimize road network traffic efficiency. Highway vehicle data refers to the multi-source, heterogeneous raw data collection on vehicle traffic events, such as vehicle location and model, collected by various sensors and monitoring devices deployed along highways through the Internet of Things (IoT). During traffic volume management, dynamic weighing data and vehicle identification data are combined to monitor overload status. Dual data verification improves the accuracy of overload detection, avoids errors that may occur with single data detection, and ensures accurate identification of overloaded vehicles. Dynamic weighing data comes from dynamic weighing sensors buried under the road surface and is used to quantify the physical load of vehicles. Vehicle identification data comes from high-definition cameras deployed above or to the side of the weighing section and is used to identify the visual identity and characteristics of vehicles, such as license plate numbers. The collected information is uploaded to the IoT. Meanwhile, by linking vehicle images with the monitoring results of overload status, precise tracking of driving trajectories can be achieved, enabling a comprehensive understanding of the dynamic operation of designated vehicles. Combined with the road network load status, the allocation of monitoring resources can be optimized, making road network supervision more targeted and effective, and improving the intelligence level of highway traffic supervision and overall operational safety.

[0028] like Figure 2 The diagram shows a flowchart of highway traffic volume management provided in an embodiment of the present invention. The standard load rate is obtained based on the number of vehicles passing through, and the load level is divided. Based on the division results, mileage ratio analysis, highway flow correlation analysis, and no processing are performed for load smoothness. The mileage ratio analysis determines the adjustment of the monitoring point allocation plan based on the obtained mileage ratio, and maintains the current plan if the mileage is not exceeded. The highway flow correlation analysis obtains the predicted vehicle passage frequency, determines whether it is within the first interval, and if so, determines that it is in an over-limit state and provides feedback; otherwise, it maintains the current state and does not perform any processing.

[0029] Furthermore, the full road network traffic heat map is used to quantify the spatial distribution of traffic load. Its acquisition process is as follows: within a specified time window, the number of vehicles passing through a specified road segment area is counted, and at the end of the specified time window, a road segment flow record is generated to reflect the vehicle traffic situation in the specified road segment area; the total number of vehicle passages in the specified road segment area is summarized, and combined with the number of vehicle passages within a preset monitoring time period (combined with time cycles, for example, if the number of vehicle passages during working hours is greater than the number during other times, then the working hours are divided into more intervals, using historical monitoring time periods as experience for pre-division), the number of vehicle passages within the preset monitoring time period is summarized to obtain the road segment vehicle frequency, which reflects the frequency of vehicle passage in the specified road segment area. Simultaneously, statistical analysis is performed based on the road segment vehicle frequency. The preset monitoring time period includes at least one continuous time interval of a specified time window; that is, the specified time window is the smallest statistical unit divided within the preset monitoring time period (e.g., 15 minutes / 30 minutes / 1 hour). Statistical analysis is used to quantify the stability of vehicle traffic conditions.

[0030] Using the vehicle frequency of all road segments corresponding to designated time windows within a preset monitoring period as the basic data sample, the difference between the maximum and minimum vehicle frequency of each road segment in the sample is used to obtain the load range. First, the sum of squared deviations of each data point from the average is calculated to obtain the variance. Then, the square root of the variance is used to obtain the load standard deviation. This is then summed with the calculated load standard deviation divided by the sample arithmetic mean to obtain the load variation coefficient, resulting in a load index used to quantify the cumulative load effect of a specified road segment area. After normalization, the standard load rate is obtained. The standard load rate is input into a preset load-code mapping table for matching, generating numerical data records of color codes and standard load rates. Spatial continuity completion and linear transformation mapping are performed using Lagrange interpolation to generate a full-network traffic heat map with spatial continuity and numerical gradient characteristics for highway traffic volume management. Specifically:

[0031] Based on numerical data records of standard load rates, load levels are classified for designated road segments using preset standard load intervals: when the standard load rate is within the first load interval (typically (0.7, 1)), it indicates congestion in the current designated road segment, and mileage ratio analysis is performed to predict the overall road network congestion range and dynamically optimize monitoring resource allocation; when the standard load rate is within the second load interval (typically (0.5, 0.7)), it indicates slow traffic in the current designated road segment, and highway traffic correlation analysis is performed to quantify traffic flow evolution trends and proactively intervene in congestion risks; when the standard load rate is within the third load interval... A value of [0, 0.5] indicates that the current specified road segment area is under smooth load and no action is taken. This value is not fixed and can be adjusted by the administrator according to different application purposes. The load level classification results are summarized and visualized in the specified road segment area of ​​the full road network traffic heat map using color coding. The first load interval, the second load interval, and the third load interval are displayed in different colors in the full road network traffic heat map, such as red, yellow, and green. The load level of the first load interval, the second load interval, and the third load interval decreases sequentially, and the sum of the proportions of the three intervals and the total length of the corresponding interval is 1.

[0032] The mileage ratio analysis specifically involves comparing the total mileage of a designated road segment area corresponding to the first load interval with the total mileage of the entire road network to obtain the mileage ratio used to quantify the load and congestion range of the entire road network. The total mileage of the entire road network represents the sum of the lengths of the pre-divided highway segment areas corresponding to the designated road segment area, i.e., the total length of the highway segment areas in the divided areas under road network monitoring. For designated road segment areas with a mileage ratio greater than the preset mileage ratio, based on the mileage ratio deviation rate and the vehicle traffic frequency of the corresponding designated road segment area, a monitoring point allocation plan adjustment prompt is sent to adjust the number of resource monitoring points in the road network management resources, including those in the entire road network during the monitoring process. The network traffic heatmap's attention and computing resources are based on a preset mileage percentage calculated from the statistical baseline of historical mileage percentage data. The adjustment of the monitoring point allocation plan depends on the mileage percentage deviation rate. The current mileage percentage deviation rate is used as the adjustment ratio for the number of monitoring points, resulting in an adjusted number of monitoring points that is a multiple of (1 + mileage percentage deviation rate) to enhance the continuity of vehicle tracking. The mileage percentage deviation rate is obtained by using the mileage percentage as the numerator and the preset mileage percentage as the denominator through a ratio. For designated road sections where the mileage percentage is not greater than the preset mileage percentage, the monitoring status of the current monitoring point allocation plan remains unchanged.

[0033] Highway traffic flow correlation analysis specifically involves: obtaining vehicle traffic frequency data for adjacent road segments within a specified road segment area; using time series analysis to predict vehicle traffic frequency values ​​to quantify the load level of the specified road segment area; firstly, collecting historical vehicle traffic frequency data for the specified road segment and adjacent road segments within the same past period; and then constructing a time series model after preprocessing such as denoising and smoothing; inputting current and recent real-time traffic frequency data of adjacent road segments into the trained model; and outputting predicted vehicle traffic frequency values ​​for the specified road segment over a future period through model calculation; if the standard load rate corresponding to the predicted vehicle traffic frequency value is within the first load interval, it indicates that the corresponding specified road segment area is in an over-limit state, and prompting pre-defined personnel to intervene in the specified road segment area, such as extending the green light time at upstream intersections or adding traffic diversion phases at downstream intersections; otherwise, no action is taken.

[0034] In this embodiment, the construction and application of a full-network traffic heatmap optimizes and upgrades traffic load management from fragmented monitoring to precise global control, making the quantitative assessment and visualization of road network traffic status more scientific and practical. Through the integrated analysis and normalization of traffic data from multiple time periods and road segments, the generated standard load rate can objectively reflect the cumulative effect and stability of traffic load in different areas. Combined with color-coded visualization, the clarity of the load status of each road segment in the road network is improved.

[0035] Differentiated management based on load levels makes traffic management strategies more targeted, allowing for precise implementation of measures according to the actual load conditions of different road sections. Corresponding analytical methods are used for high-load congested road sections and medium-load slow-moving road sections, while resources are rationally allocated to low-load smooth road sections. This avoids ineffective investment of management resources, improves the overall efficiency of traffic management, and enables managers to quickly locate bottleneck areas in the road network, providing a reliable basis for traffic diversion and resource allocation. This promotes the transformation of traffic management from experience-driven to data-driven, improving the balance and smoothness of road network traffic operation.

[0036] The application of mileage ratio analysis and highway traffic flow correlation analysis has enabled an optimized transformation in traffic management from passive response to proactive prediction, improving the road network congestion control capabilities and the efficiency of management resource allocation. Mileage ratio analysis, by accurately quantifying the entire road network congestion area and combining it with a reasonable baseline constructed from historical data, dynamically adjusts the distribution of monitoring resources to ensure that congested areas receive sufficient monitoring support. This avoids the problem of resource allocation being out of sync with actual needs under the traditional fixed monitoring point layout, allowing limited management resources to be tilted towards high-demand areas. This improves the monitoring accuracy and response speed of congested road sections, while reducing resource redundancy consumption in low-load areas, maximizing resource utilization efficiency.

[0037] like Figure 3The diagram shows a flowchart of the over-limit status monitoring provided in this embodiment of the invention. The wheel axle pressure area is obtained by acquiring the pressure curve. The wheel axle pressure areas are combined into a wheel axle data set. It is determined whether the acquisition time interval outside the wheel axle data set is greater than a preset value. If so, the wheel axle data set is cut into a sequence. Otherwise, the current set is used as the wheel axle sequence, and vehicle image matching is performed thereafter. Over-limit calculation is performed based on the matching result, including over-limit detection of each axle for axle load calculation and detection of the total weight obtained by adding the weights of each axle.

[0038] Furthermore, monitoring of over-limit status is performed by combining vehicle image recognition data within a preset monitoring time period, including: converting the voltage-time series signal corresponding to the dynamic weighing section into a discrete digital signal point sequence, and obtaining a filtered data point sequence after noise suppression processing; linearly fitting the data point sequence based on the least squares method to obtain a standardized signal sequence, and smoothing it to obtain a pressure curve reflecting the change law of road pressure with time and the peak value of wheel axle pressure; the voltage-time analog signal output by the dynamic weighing section sensor is sampled by an analog-to-digital converter at a preset frequency and converted into a discrete digital signal point sequence containing timestamps and voltage values; through adaptive median filtering, outliers are replaced with the sliding window median to obtain a denoised data point sequence; a linear model is fitted using the least squares method to eliminate fluctuations and obtain a standardized signal sequence; cubic spline interpolation is then used to supplement the data between nodes, smoothing the discrete sequence into a continuous curve; finally, based on the sensor voltage-pressure calibration relationship, the voltage-time series signal is adjusted to the standard signal sequence. The voltage curve is converted into a pressure curve with time as the horizontal axis and pressure as the vertical axis, which can clearly show the pressure change pattern and the peak pressure of the wheel and axle. The pressure peak points in the pressure curve that are greater than the preset pressure amplitude are obtained, and all pressure peak points are counted. Each valid peak point represents a wheel (or a pair of wheels mounted in parallel) passing over the sensor. That is, the peak shape is complete and must have a clear rising edge and falling edge. The peak width must conform to the normal time range of the wheel (or the pair of wheels mounted in parallel) passing over the sensor to avoid misjudgment of a single sharp noise point. The peak amplitude, peak width and inter-peak time interval in the pressure curve are combined and normalized to obtain the wheel and axle pressure area to reflect the cumulative effect of pressure and time. The normalization process adopts the Min-Max normalization method to map the peak amplitude, peak width and inter-peak time interval to the dimensionless target interval. The preset pressure amplitude is determined by calculating the quantiles of the historical pressure amplitude in the historical over-limit state monitoring based on the real-time noise statistical characteristics.

[0039] The specific expression for the wheel axle pressure area S is as follows:

[0040]

[0041] In the formula, k represents the pressure-time calibration coefficient, which is obtained by fitting the sensor voltage-pressure calibration coefficient with actual road pressure verification data. It is used to correct systematic errors caused by different road materials and sensor installation depths, ensuring that the calculation results are consistent with the actual pressure cumulative effect; n represents the number of effective pressure peak points, A i W represents the peak amplitude of the i-th effective pressure peak point. i α represents the peak width of the i-th effective pressure peak point, and α represents the inter-peak time correction time, which is derived from historical axle traffic data. It is the deviation between the initial axle pressure area calculated from historical data that did not include this correction time and the actual verified value. Through the statistical method of linear regression, the correction coefficient that minimizes the deviation is fitted to balance the influence of the time interval between adjacent peaks on the pressure accumulation effect and avoid the calculation deviation caused by the difference in wheel spacing; T represents the inter-peak time interval of the effective peak points.

[0042] The combined effect of pressure and time is characterized by quantifying the amplitude-time cumulative effect of wheel and axle pressure. At the same time, the systematic errors caused by hardware and environmental factors such as sensors and road surface are corrected by the pressure-time calibration coefficient. The influence of wheel spacing difference on pressure accumulation calculation is compensated by the inter-peak time correction coefficient and the average inter-peak time interval. Finally, the combined effect of wheel and axle pressure over time is obtained.

[0043] The wheel axle pressure area, timestamp, and inter-peak time interval are combined into a wheel axle data set. For each effective pressure peak in the pressure curve, the corresponding sampling timestamp is extracted, and the wheel axle pressure area calculated from that peak is associated with it. Then, all basic data entries are sorted in chronological order according to their timestamps. Based on the timestamps of two adjacent wheel axle units, the time difference between the next unit and the previous unit is calculated as the inter-peak time interval corresponding to the current wheel axle unit. Finally, all the verified entries are integrated in chronological order to obtain the wheel axle data set.

[0044] When the axle set interval between adjacent data in the wheel axle dataset is greater than the preset axle set interval, the streaming segmentation algorithm is used to cut the wheel axle dataset into a wheel axle dataset sequence. The preset axle set interval is obtained by calculating the quantile statistics of the axle set interval driven by historical data. When the axle set interval between adjacent data is not greater than the preset axle set interval, the corresponding wheel axle dataset is used as the wheel axle dataset sequence.

[0045] The preset axle topology is matched with the vehicle identification data in the vehicle images monitored by the road network. The preset axle topology represents a multi-dimensional sequence matching rule set in advance based on the highway vehicle axle classification standard and the corresponding road network monitoring requirements. The vehicle identification data includes a preset wheel and axle data set sequence. If the preset wheel and axle data set sequence completely matches the acquired wheel and axle data set sequence, the axle signal of the vehicle is confirmed to be valid, and an over-limit calculation is performed.

[0046] If the pre-set wheel axle data set sequence does not completely match or does not match the obtained wheel axle data set sequence, the corresponding wheel axle data set sequence will be marked as verification failure and reported back to the preset personnel.

[0047] Specifically, the overload calculation process is as follows: Based on the acquired wheel and axle data set sequence, the wheel and axle data set sequence is merged into axle groups at preset intervals within the axle set. Simultaneously, based on the wheel and axle feature recognition results in the wheel and axle data set sequence, the wheel and axle pressure area is obtained. The wheel and axle pressure area is used to reflect the wheel and axle pressure distribution and the stress state of the road surface. If the wheel and axle pressure area corresponding to any axle group exceeds the preset wheel and axle pressure area, the corresponding axle group is marked as having an overload, and the corresponding overload value is recorded; otherwise, it is recorded as not exceeding the axle load limit. At the same time, the wheel and axle pressure areas of each axle group are summed to obtain the total wheel and axle pressure area. If the obtained total wheel and axle pressure area exceeds the preset total wheel and axle pressure area, the vehicle corresponding to the total wheel and axle pressure area is marked as having an overload, the overload value of the total wheel and axle pressure area is recorded, and feedback is sent to preset personnel for intervention.

[0048] In this embodiment, by optimizing data processing across multiple stages and cross-validating data from multiple sources, the accuracy and reliability of overload detection are improved, solving the problems of susceptibility to interference and high false positive rates in existing technologies using single-data detection. After digital conversion, smoothing, and filtering optimization, the weighing signal can more realistically reflect the pressure of the vehicle's wheel axles on the road surface, reducing the impact of environmental noise and equipment errors on data acquisition, and providing high-quality basic data for subsequent overload judgment.

[0049] The segmentation of the wheel and axle data set and the matching verification of the axle type topology further eliminate invalid data and interference information, ensuring the validity and authenticity of the data involved in the overload calculation, and avoiding misjudgments or omissions due to data errors. The dual overload judgment mode of axle group and total weight can accurately identify the overload situation of individual axle groups while comprehensively controlling the overall load status of the vehicle, making overload judgment more comprehensive and rigorous, providing accurate and reliable overload basis, helping to achieve efficient and standardized overload management, and ensuring the safe and stable operation of the highway network.

[0050] like Figure 4The flowchart shown is a process for trajectory status monitoring provided in an embodiment of the present invention. The traffic heat map of the entire road network is combined with the obtained traffic heat map to obtain the road network load pressure distribution map. The load pressure in the road network load pressure distribution map is dynamically allocated. It is determined whether the obtained load pressure is greater than the preset load pressure. If so, the video analysis parameters, including the video frame rate and image resolution, are improved based on the deviation. Otherwise, the focus of monitoring and sensing resources is reduced.

[0051] Further, the status monitoring process is as follows: Vehicle identification data from the vehicle images of the road network monitoring corresponding to the current specified road segment area is spatially overlaid and fused with the traffic flow heat map of the entire road network. Based on a unified data spatial coordinate system, the vehicle identification data is calibrated and matched with the road segment spatial boundaries and coordinate system of the traffic flow heat map of the entire road network. The vehicle identification data of each road segment is associated and bound with the load data of that road segment in the heat map. The vehicle traffic feature information is overlaid onto the corresponding heat map area through a spatial mapping algorithm to obtain a road network load pressure distribution map for visualizing the load superposition effect and spatial distribution characteristics of high-load road segments.

[0052] Based on the load pressure of a specified road segment area in the road network load pressure distribution map, monitoring and sensing resources are dynamically allocated. Monitoring and sensing resources refer to the general term for various hardware devices, computing and processing resources and data transmission channels used to collect traffic data during the road network monitoring process. These include dynamic weighing sensors, high-definition surveillance cameras, traffic flow detectors, edge computing nodes, data storage modules and communication links, etc., and are the core supporting resources for realizing real-time monitoring and analysis of road network load pressure.

[0053] The dynamic allocation process is as follows: For designated road sections where the load pressure exceeds the preset load pressure, the video analysis parameters are adjusted based on the load pressure deviation. These parameters include the video frame rate and image resolution, acquired through video equipment during road network monitoring. The specific adjustment process involves using the load pressure deviation as the numerator and the preset load pressure as the denominator, proportionally calculating the load pressure deviation rate. The load pressure deviation is represented by subtracting the preset load pressure from the acquired load pressure. When the load pressure deviation rate is in the range [0, 0.3), the adjustment is performed using the target adjustment value = video analysis parameter (1 + load pressure deviation rate). When the load pressure deviation rate is in the range [0.3, 0.6), the system directly switches to the preset high-smoothness mode (typically 30 FPS and 2K resolution). When the load pressure deviation rate is in the range [0.6, 1], a load deviation warning is issued.

[0054] For designated road sections where the load pressure does not exceed the preset load pressure, the focus of monitoring and sensing resources is reduced based on the load pressure deviation. An adaptive algorithm combining fuzzy control and reinforcement learning maps continuous load pressure deviations to discrete resource adjustment strategies, continuously optimizing the mapping relationship through a feedback mechanism. First, the load pressure deviation value is fuzzified, and the entire adjustment process is modeled as a Markov decision process to learn an optimal strategy function. Finally, a dynamic resource emphasis coefficient is output, reducing the current focus of sensing resources to optimize resource consumption. Monitoring strategy optimization instructions are then generated and fed back to designated personnel to achieve targeted and accurate road network monitoring. The load pressure deviation is obtained by subtracting the preset load pressure from the acquired load pressure, and the preset load pressure is obtained through cluster analysis of historically acquired load pressures.

[0055] In this embodiment, the trajectory monitoring step, through multi-dimensional data linkage and dynamic resource allocation, achieves an optimized upgrade of road network monitoring from comprehensive coverage to precise focus, improving monitoring efficiency and resource utilization efficiency. By presenting vehicle travel trajectories in conjunction with traffic records, managers can clearly grasp the entire operational dynamics of designated vehicles, providing coherent and complete data support for the tracing and supervision of overloaded vehicles.

[0056] Meanwhile, the road network load pressure distribution map, generated by deeply integrating driving trajectories with the traffic flow heat map of the entire road network, intuitively presents the correlation between vehicle operation and road segment load pressure, providing a precise basis for the scientific allocation of monitoring resources. Implementing differentiated resource allocation strategies for road segments with different load pressures ensures that high-load-pressure road segments receive sufficient monitoring resources. By improving video analysis accuracy and data processing priority, the monitoring capability of key areas is strengthened, avoiding monitoring omissions due to insufficient resources. Simultaneously, by reducing the resource emphasis on low-load-pressure road segments, unnecessary resource consumption is reduced, achieving rational allocation and efficient utilization of monitoring resources. This makes road network monitoring more targeted and flexible, improving the intelligence level and overall operational stability of the entire road network monitoring management.

[0057] like Figure 5The diagram shown is a structural schematic of a dynamic monitoring and management system for overloaded highway vehicles provided in an embodiment of the present invention. This system includes a highway traffic volume management module, an overload status monitoring module, and a trajectory status monitoring module. The highway traffic volume management module is used to obtain a comprehensive traffic flow heat map of the entire road network based on the collected highway vehicle data and the vehicle passage frequency of each designated road segment area, and to manage highway traffic volume. The overload status monitoring module is used to monitor the overload status based on the vehicle weighing data corresponding to the dynamic weighing section and combined with vehicle image recognition data within a preset monitoring period. The trajectory status monitoring module is used to monitor the trajectory of designated vehicles based on the results of the overload status monitoring and combined with vehicle images monitored by the road network.

[0058] In this embodiment, the highway traffic volume management module generates a full-network traffic heat map, providing accurate background traffic data support for the overload status monitoring module, ensuring that overload monitoring can be carried out efficiently in conjunction with road segment traffic load. The monitoring results from the overload status monitoring module provide a clear tracking target for the trajectory status monitoring module, enabling it to specifically lock onto designated vehicles for trajectory tracking. Simultaneously, vehicle operation data acquired by trajectory monitoring can be fed back to the traffic volume management module, optimizing road network traffic management strategies. The three modules work together to form a complete regulatory closed loop, achieving efficient collaboration between road network status perception, accurate overload monitoring, and vehicle trajectory tracking, improving the systematic nature and effectiveness of highway vehicle monitoring and management, and ensuring the safe and orderly operation of the road network.

[0059] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0060] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0061] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0062] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0063] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0064] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0065] If the functionality is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0066] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for dynamic monitoring and management of overloaded highway vehicles, characterized in that, Includes the following steps: S1. Based on the collected highway vehicle data, the frequency of vehicle passage in each designated road section area is used to obtain a traffic flow heat map of the entire road network, and highway traffic volume management is carried out. S2. In the process of highway traffic volume management, the vehicle weighing data corresponding to the dispatch dynamic weighing section is used to monitor the over-limit status in combination with the vehicle image recognition data within the preset monitoring time period. S3. Based on the results of over-limit status monitoring and combined with vehicle images from road network monitoring, the driving trajectory of the specified vehicle is monitored.

2. The method for dynamic monitoring and management of overloaded highway vehicles as described in claim 1, characterized in that, The traffic flow heatmap for the entire road network is used to quantify the spatial distribution of traffic load, and its acquisition process is as follows: Within a specified time window, the number of vehicles passing through a specified road segment area is counted, and at the end of the specified time window, a road segment streaming record is generated to reflect the vehicle traffic situation in the specified road segment area. The total number of vehicle passages in a specified road segment area is summarized and combined with a preset monitoring time period to obtain the road segment vehicle frequency, which reflects the frequency of vehicle passage in the specified road segment area. At the same time, statistical analysis is performed based on the road segment vehicle frequency. The preset monitoring time period includes at least one continuous time interval of a specified time window. Based on the results of statistical analysis, a load index is obtained to quantify the cumulative load effect of a specified road section area, and the standard load rate is obtained after normalization. The standard load rate is input into a preset load-code mapping table for matching, generating numerical data records of color codes and standard load rates. Spatial continuity completion and linear transformation mapping are then performed to generate a full road network traffic heat map with spatial continuity and numerical gradient characteristics for highway traffic volume management.

3. The method for dynamic monitoring and management of overloaded highway vehicles as described in claim 2, characterized in that, The specific process for managing highway traffic volume is as follows: Based on the color coding obtained from the traffic heat map of the entire road network, the load level of the specified road segment area is divided according to the preset standard load range: When the standard load rate is within the first load range, it indicates that the current specified road segment area is congested. This is used to perform mileage ratio analysis to predict the congestion range of the entire road network and dynamically optimize the allocation of monitoring resources. When the standard load rate is within the second load range, it indicates that the current designated road section area is experiencing slow traffic. Highway traffic correlation analysis is then conducted to quantify traffic flow evolution trends and intervene in congestion risks in advance. When the standard load rate is within the third load range, it indicates that the load in the current designated road section area is smooth, and no action is taken. The results of load level classification are summarized and visualized in designated road segment areas of the full road network traffic heat map; The load levels of the first load interval, the second load interval, and the third load interval decrease sequentially, and the sum of the proportions of the three intervals to the total length of the corresponding interval is 1.

4. The method for dynamic monitoring and management of overloaded highway vehicles as described in claim 3, characterized in that, The mileage ratio analysis is performed as follows: The total mileage of the designated road segment area corresponding to the first load interval is compared with the total mileage of the entire road network to obtain the mileage ratio used to quantify the load and congestion range of the entire road network. The total mileage of the entire road network represents the sum of the lengths of the pre-divided highway segment areas corresponding to the designated road segment area. For designated road sections where the mileage percentage is greater than the preset mileage percentage, a notification to adjust the monitoring point allocation plan is sent based on the mileage percentage deviation rate and the vehicle traffic frequency of the corresponding designated road section. For designated road sections where the mileage percentage is no greater than the preset mileage percentage, the monitoring status of the current monitoring point allocation plan will remain unchanged.

5. The method for dynamic monitoring and management of overloaded highway vehicles as described in claim 3, characterized in that, The highway traffic flow correlation analysis specifically includes: Obtain the vehicle traffic frequency of adjacent road segments in a specified road segment area to obtain a predicted vehicle traffic frequency value used to quantify the load level of the specified road segment area; If the standard load rate corresponding to the predicted vehicle traffic frequency is within the first load range, it indicates that the corresponding designated road segment is in an over-limit state, and the preset personnel will be prompted to intervene in the designated road segment; otherwise, no action will be taken.

6. The method for dynamic monitoring and management of overloaded highway vehicles as described in claim 1, characterized in that, The monitoring of over-limit status by combining vehicle image recognition data within a preset monitoring time period includes: The voltage-time series signal corresponding to the dynamic weighing section is converted into a discrete digital signal point sequence, and after noise suppression processing, a filtered data point sequence is obtained. Linear fitting of the data point sequence based on the least squares method yields a standardized signal sequence, which is then smoothed to obtain a pressure curve that reflects the change of road pressure over time and the peak value of wheel axle pressure. Obtain the pressure peak points in the pressure curve that are greater than the preset pressure amplitude, and count all the pressure peak points. Combine them with the peak amplitude, peak width and inter-peak time interval in the pressure curve to obtain the wheel and axle pressure area that reflects the cumulative effect of pressure and time. The wheel axle pressure area, timestamp, and inter-peak time interval are combined into a wheel axle data set. When the interval between adjacent data in the wheel axle data set is greater than the preset interval, the wheel axle data set is cut into a wheel axle data set sequence. When the interval between adjacent data is not greater than the preset interval, the corresponding wheel axle data set is used as the wheel axle data set sequence.

7. The method for dynamic monitoring and management of overloaded highway vehicles as described in claim 6, characterized in that, The method of monitoring over-limit status by combining vehicle image recognition data within a preset monitoring time period also includes: The preset axle topology is matched with vehicle identification data in vehicle images monitored by the road network. The preset axle topology represents a multi-dimensional sequence matching rule pre-set based on the highway vehicle axle classification standard and the corresponding road network monitoring requirements. The vehicle identification data includes a sequence of wheel and axle data sets. If the pre-set wheel and axle data set sequence completely matches the acquired wheel and axle data set sequence, the axle type signal of the vehicle is confirmed to be valid, and an over-limit calculation is performed. If the pre-set wheel axle data set sequence does not completely match or does not match the obtained wheel axle data set sequence, the corresponding wheel axle data set sequence will be marked as verification failure and reported back to the preset personnel.

8. The method for dynamic monitoring and management of overloaded highway vehicles as described in claim 7, characterized in that, The aforementioned over-limit calculation specifically refers to: Based on the acquired wheel and axle data set sequence, the wheel and axle data set sequence is merged into axle groups at a preset interval within the axle set. At the same time, based on the wheel and axle feature recognition results in the wheel and axle data set sequence, the wheel and axle pressure area is obtained. The wheel and axle pressure area is used to reflect the wheel and axle pressure distribution and the road surface stress state. If the wheel axle pressure area of ​​any axle group exceeds the preset wheel axle pressure area, the corresponding axle group will be marked as overloaded and the corresponding overload value will be recorded; otherwise, it will be recorded as not overloaded. Simultaneously, the wheel axle pressure areas of each axle group are summed to obtain the total wheel axle pressure area. If the obtained total wheel axle pressure area exceeds the preset total wheel axle pressure area, the vehicle corresponding to the total wheel axle pressure area is marked as exceeding the total weight limit, the excess value of the total wheel axle pressure area is recorded, and feedback is sent to the preset personnel for intervention.

9. The method for dynamic monitoring and management of overloaded highway vehicles as described in claim 1, characterized in that, The specific process of the status monitoring is as follows: Vehicle identification data from vehicle images monitored in the current designated road segment area are spatially overlaid and fused with the traffic flow heat map of the entire road network to obtain a road network load pressure distribution map for visualizing the load superposition effect and spatial distribution characteristics of high-load road segments. Based on the load pressure of a specified road segment area in the road network load pressure distribution map, the monitoring and sensing resources are dynamically allocated. The specific process is as follows: For designated road sections where the load pressure exceeds the preset load pressure, the frame rate and image resolution of the video analysis are increased based on the load pressure deviation. For designated road sections where the load pressure does not exceed the preset load pressure, the focus of monitoring and sensing resources is reduced based on the load pressure deviation in order to optimize resource energy consumption, and monitoring strategy optimization instructions are generated and fed back to the preset personnel.

10. A dynamic monitoring and management system for overloaded highway vehicles, using the dynamic monitoring and management method for overloaded highway vehicles as described in any one of claims 1-9, comprising a highway traffic volume management module, an overload status monitoring module, and a trajectory status monitoring module; The highway traffic volume management module is used to obtain a comprehensive traffic flow heat map of the entire road network based on the collected highway vehicle data and the vehicle passage frequency of each designated road section area, and to manage highway traffic volume. The over-limit status monitoring module is used to monitor over-limit status based on the vehicle weighing data corresponding to the scheduling dynamic weighing section and combined with vehicle image recognition data within a preset monitoring time period. The trajectory status monitoring module is used to monitor the driving trajectory of a specified vehicle based on the results of over-limit status monitoring and the vehicle images monitored by the road network.

Images