A vehicle weighing method and system based on behavior monitoring and error compensation
By collecting triaxial acceleration data from the weighbridge and performing biaxial tilt angle calculation and error compensation, the problem of disconnect between weighing accuracy and violation identification in existing technologies has been solved, achieving accurate weighing and violation identification.
Patent Information
- Application Number
- CN202611053848.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-15
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies are disconnected in identifying weighing violations and eliminating weighing errors, and cannot simultaneously improve weighing accuracy and the precision of violation identification.
By collecting triaxial acceleration data from both sides of the weighbridge, dual-axis tilt angle calculation is performed, a vehicle-specific tilt angle-error calibration database is constructed, vehicle behavior types are identified and weighing error compensation is performed, and weighing results and violation identification results are output.
It achieves accurate identification of violations such as skipping or dragging the scale, and linear quantification and compensation of weighing errors, thereby improving the accuracy of weighing results and the accuracy of identifying violations.
Smart Images

Figure CN123111439A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of vehicle weighing and measurement technology, and in particular to a vehicle weighing method and system based on behavior monitoring and error compensation. Background Technology
[0002] Currently, dynamic weighing systems at toll stations generally rely on pressure sensors on the weighbridge to collect weight data. However, behaviors such as trucks jumping on the weighbridge (inertial offset caused by sudden braking or acceleration), dragging the weighbridge (low-speed dragging with uneven load), and road slope disturbances can cause slight pitch and roll angle deformations on the weighbridge. These not only interfere with the accuracy of weight collection and introduce measurement errors, but also require specialized technology to identify violations.
[0003] Existing technologies have publicly disclosed technical solutions for identifying weighing violations and eliminating weighing errors. For example:
[0004] Chinese patent CN113295253B discloses a method and system for monitoring abnormal vehicle weighing data on a weighbridge without requiring the user to perceive the weighing process. This method involves using multiple cameras to monitor the weighing site, sending the captured videos and images to a server, and analyzing the images to identify and display the weighing environment and data. During weighing data identification, it combines digit recognition with decimal point recognition. Decimal point recognition involves vertically projecting the image, analyzing the resulting array, and identifying the peak position to determine the decimal point location. Finally, the results of multiple recognitions are saved to a weighing label, which is used to identify cheating and abnormal data. By using vehicle recognition, quantity recognition, license plate recognition, and human body recognition, the method performs anomaly detection during the weighing process, enabling the detection of weighing cheating and the identification of abnormal weighing data.
[0005] Chinese Patent Publication No. CN119245793B discloses a piezoelectric-based vehicle overload alarm method and related device. The method involves measuring the target vehicle using k metering modules to obtain k first metering data; compensating and adjusting the k first metering data based on environmental parameters and performing anomaly correction to obtain k third metering data; determining the first center position of the target vehicle; acquiring k second center positions of k target tires; calculating the distance between the first center position and each second center position to obtain k distances and determining k weight values; determining the metering result based on the k third metering data and the k weight values; generating a first alarm message if the metering result is greater than a preset first alarm threshold and less than a preset second alarm threshold; and generating a second alarm message if the metering result is greater than or equal to the preset second alarm threshold. This method can improve the accuracy of vehicle overload alarms.
[0006] However, existing technical solutions like the aforementioned patents either only focus on identifying, warning, and collecting evidence of violations such as skipping or dragging the scale, failing to quantify the weighing error caused by posture deviation; or they simply optimize weighing accuracy without determining the source of the error (whether it is caused by violations), failing to balance law enforcement needs and measurement accuracy, resulting in a disconnect between monitoring and compensation. Summary of the Invention
[0007] The purpose of this invention is to overcome the shortcomings of the prior art where weighing behavior monitoring and weighing error compensation are disconnected, and to provide a vehicle weighing method and system based on behavior monitoring and error compensation.
[0008] In a first aspect, the present invention provides a vehicle weighing method based on behavior monitoring and error compensation. The vehicle weighing method includes: Collect the initial weighing value of the vehicle; Real-time acquisition of three-axis acceleration data on both sides of the weighing platform; Build a vehicle-specific camber-error calibration database; Based on the triaxial acceleration data, the biaxial tilt angle is calculated to identify the vehicle behavior type; and based on the calculated tilt angle value, the weighing error is obtained from the calibration database to compensate the initial weighing value, and the weighing result and violation behavior identification result are output.
[0009] According to a preferred embodiment, the step of calculating the biaxial tilt angle based on the triaxial acceleration data includes: Using the vector relationship between the triaxial acceleration data and gravitational acceleration, the pitch angle α and roll angle β of the weighing platform are calculated; The tilt angle data is smoothed by moving average filtering to obtain stable tilt angle values α0 and β0; Vehicle behavior type identification is performed based on tilt angle threshold and variation characteristics.
[0010] According to a preferred embodiment, the construction of a vehicle-specific tilt-error calibration database includes: Different vehicle models and loads were selected to simulate illegal and normal behavior scenarios. Initial weighing values W0 and standard weights W_std corresponding to different tilt angles were collected, errors were calculated, and calibration data were obtained. Based on the calibration data, a linear model is constructed to obtain a tilt-error calibration database specific to the vehicle model.
[0011] According to a preferred embodiment, the parameters of the linear model are iteratively updated based on measured data.
[0012] According to a preferred embodiment, the step of obtaining the weighing error from the calibration database based on the calculated tilt angle value and compensating for the initial weighing value includes: Substitute the stable tilt angles α0 and β0 into the linear model to calculate the current weighing error; The compensation is to obtain the actual weight of the vehicle; Error verification is performed. If the deviation between the compensated weight and the vehicle's load range exceeds the deviation threshold, a second calibration is performed. If the deviation between the compensated weight and the vehicle's load range does not exceed the deviation threshold, the weighing result and the violation identification result are output.
[0013] According to a preferred embodiment, the vehicle weighing method further includes: Calculate the rate of change of inclination angle and determine whether the rate of change of inclination angle exceeds the threshold. If the weight limit is not exceeded, vehicle behavior type identification and weighing error compensation will be performed. If the threshold is exceeded, a sudden interference is determined, and vehicle behavior type identification and weighing error compensation are not performed.
[0014] According to a preferred embodiment, the step of calculating the biaxial tilt angle based on the triaxial acceleration data further includes preprocessing the triaxial acceleration data, including: Filtering and normalization processing; Using the moment when the vehicle's front wheels drive onto the weighbridge as the time origin, stable data within a preset time period starting from the time origin is extracted.
[0015] In a second aspect, the present invention also provides a vehicle weighing system based on behavior monitoring and error compensation, comprising: a weighing pressure sensing module, a triaxial acceleration sensing module symmetrically installed on both sides of the weighbridge, a cloud calibration platform, and an edge computing module. The weighing pressure sensing module is used to collect the initial weighing value of the vehicle. The triaxial acceleration sensing module symmetrically installed on both sides of the weighbridge is used to acquire triaxial acceleration data in real time. The cloud calibration platform is used to construct a vehicle-specific tilt angle-error calibration database. The edge computing module performs dual-axis tilt angle calculation based on the sensing data from the triaxial acceleration sensing module, identifies the vehicle behavior type, and obtains the weighing error from the calibration database based on the calculated tilt angle value, compensates for the initial weighing value, and outputs the weighing result and violation behavior identification result.
[0016] According to a preferred embodiment, the vehicle weighing system further includes a data synchronization unit; the data synchronization unit is used to synchronously transmit the sensing data of the weighing pressure sensing module and the triaxial acceleration sensing module to the edge computing module.
[0017] According to a preferred embodiment, the vehicle weighing system further includes a video unit and a cloud storage unit. The video unit is used to capture images during the vehicle weighing process. The cloud storage unit is used to store the weighing results and violation identification results output by the edge computing module, as well as the vehicle weighing images captured by the video unit.
[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention provides a vehicle weighing method based on behavior monitoring and error compensation. It collects triaxial acceleration data from the weighbridge, calculates the biaxial tilt angle, and identifies vehicle behavior types. Based on the calculated tilt angle value and a vehicle-specific tilt angle-error calibration database, it obtains the weighing error. Then, it compensates for the initial weighing value collected when the vehicle passes through the weighbridge, outputting the weighing result and the result of identifying violations. This invention simultaneously identifies violations such as skipping or dragging the weighbridge using triaxial acceleration data from both sides of the same weighbridge, and linearly quantifies and compensates for weighing errors based on biaxial tilt angles, thus improving both the accuracy of the weighing results and the accuracy of violation identification. Attached Figure Description
[0019] Figure 1 This is a flowchart illustrating a vehicle weighing method based on behavior monitoring and error compensation.
[0020] Figure 2 This is a schematic diagram of a vehicle weighing system based on behavior monitoring and error compensation. Detailed Implementation
[0021] The present invention will now be described in further detail with reference to specific embodiments. However, this should not be construed as limiting the scope of the present invention to the following embodiments; all technologies implemented based on the content of the present invention fall within the scope of the present invention.
[0022] Unless otherwise specified, the terms "upper," "lower," "left," "right," "center," "inner," and "outer," etc., used in the description of specific embodiments of the present invention to indicate orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings, or the orientation or positional relationship in which the product / equipment / device is usually placed during use. These terms are merely for the purpose of facilitating the description of the present invention or simplifying the description in specific embodiments, and for enabling those skilled in the art to quickly understand the solution, and do not indicate or imply that a particular device / component / element must have a specific orientation, or be constructed and operated in a specific positional relationship. Therefore, they should not be construed as limitations on the present invention.
[0023] Furthermore, the use of terms such as "horizontal," "vertical," "suspended," "parallel," and "coaxial" does not imply that the corresponding device / component / element must be absolutely horizontal, vertical, suspended, parallel, or coaxial. Slight tilt or deviation is permissible, as long as it does not affect the normal function of the relevant component. For example, "horizontal" simply means that its direction is more horizontal relative to "vertical," not that the structure must be perfectly horizontal; a slight tilt is acceptable. "Coaxial" means that two components are arranged as coaxially as possible, allowing them to move coaxially or approximately coaxially when their relative positions change. Alternatively, it can be simplified to mean that the corresponding device / component / element, when arranged in "horizontal," "vertical," "suspended," "parallel," or "coaxial" directions, can have an error / deviation of ±10% relative to the corresponding direction, more preferably within ±8%, more preferably within ±6%, more preferably within ±5%, and more preferably within ±4%. For example, the deviation in the "coaxial" direction is controlled within 0.2-1mm, preferably within 0.2-0.5mm. As long as the corresponding device / component / element is within the error / deviation range, it can still achieve its function in the solution of the present invention.
[0024] Furthermore, the use of terms such as "first," "second," and "third" in terminology is merely for distinguishing descriptions of identical or similar components and should not be interpreted as emphasizing or implying the relative importance of a particular component.
[0025] Furthermore, in the description of the embodiments of the present invention, "several", "more than", and "a number of" represent at least two. The number can be any number, such as two, three, four, five, six, seven, eight, or nine, and can even exceed nine.
[0026] Furthermore, in the description of the technical solution of this invention, unless otherwise explicitly specified / limited / restricted, the terms "set up," "install," "connect," "link," "provided with," "laid out," and "arranged" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to connection methods commonly used in the art, such as welding, riveting, bolting, and threaded connections. Such connections can be mechanical, electrical, or communication connections; they can be direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components.
[0027] Example 1 This embodiment provides a vehicle weighing method based on behavior monitoring and error compensation. The vehicle weighing method includes: Collect the initial weighing value of the vehicle; Real-time acquisition of three-axis acceleration data on both sides of the weighing platform; Build a vehicle-specific camber-error calibration database; Based on the triaxial acceleration data, the vehicle behavior type is identified by calculating the biaxial tilt angle. The weighing error is then obtained from the calibration database based on the calculated tilt angle value to compensate for the initial weighing value. Finally, the weighing result and the violation identification result are output.
[0028] This embodiment provides a vehicle weighing method based on behavior monitoring and error compensation. It collects triaxial acceleration data from the weighbridge, calculates the biaxial tilt angle, and identifies vehicle behavior types. Based on the calculated tilt angle value and a vehicle-specific tilt angle-error calibration database, it obtains the weighing error. Then, it compensates for the initial weighing value of the vehicle, outputting the weighing result and the result of identifying violations. This invention uses triaxial acceleration data from both sides of the same weighbridge to simultaneously identify violations such as skipping or dragging the scale, and linearly quantifies and compensates for weighing errors based on biaxial tilt angles. This improves both the accuracy of the weighing results and the accuracy of violation identification.
[0029] Example 2 This embodiment is a further improvement on embodiment 1, and the repeated content will not be described again.
[0030] See Figure 1 The vehicle weighing method based on behavior monitoring and error compensation includes the following steps: Step 1: Data Collection; Step 2: Data preprocessing; Step 3: Calculation of double tilt angles; Step 4: Vehicle Behavior Type Identification; Step 5: Weighing error compensation; Step 6: Output the results.
[0031] Preferably, step one, data acquisition, includes: Vehicle type data collection can be achieved by acquiring vehicle images through cameras to determine the type of vehicle to be weighed, such as light truck, medium truck, or heavy truck.
[0032] Collect the initial weighing value of the vehicle when it is weighed.
[0033] Real-time acquisition of triaxial acceleration data on both sides of the weighing platform.
[0034] Preferably, during step one, data acquisition, technologies such as timestamp alignment are used to synchronize various types of acquired data, thereby reducing or eliminating transmission delay differences.
[0035] Preferably, step two, data preprocessing, mainly includes: preprocessing the triaxial acceleration data.
[0036] Preferably, the preprocessing of the triaxial acceleration data specifically includes: A Butterworth low-pass filter (cutoff frequency 10Hz) is used to filter out high-frequency interference caused by road bumps and sensor noise. Z-score normalization is used to normalize the triaxial acceleration data to eliminate baseline shift caused by temperature drift. Taking the moment when the vehicle's front wheels drive onto the weighbridge as the time origin t0, stable data (covering the core time period of attitude deviation) is extracted from t0 to t0+300ms.
[0037] Preferably, step three, dual tilt angle calculation, includes: using the vector relationship between the preprocessed triaxial acceleration data and gravitational acceleration, calculating the pitch angle α (tilt angle in the vehicle's direction of travel, corresponding to the weighing behavior) and roll angle β (tilt angle in the width direction of the weighing platform, corresponding to the towing and off-center loading behaviors).
[0038] Preferably, the formula for calculating the double tilt angle is as follows: Pitch angle α: α = arctan(ax / (ay² + az²)) 1 / 2 ) Roll angle β: β = arctan(ay / (ax² + az²)) 1 / 2 ) Where ax is the X-axis acceleration component (direction of travel), ay is the Y-axis component, and az is the Z-axis component (gravity-dominated).
[0039] After calculation, the dip angle data is smoothed by moving average filtering (the window size can be set according to the actual situation, such as 2, 3, 4, 5, 6... time steps, preferably 5 time steps) to obtain stable dip angle values α0 and β0 (unit: °).
[0040] Preferably, step four, vehicle behavior type identification, includes: identifying vehicle behavior type based on tilt angle threshold and change characteristics.
[0041] Preferably, step four, vehicle behavior type identification, specifically includes: simultaneously identifying tipping and towing behaviors based on the tilt angle threshold and change characteristics. When the absolute value of α0 is greater than 0.1° and the rate of change is greater than 0.05° / ms, it is determined to be tipping behavior; when the absolute value of β0 is greater than 0.08° and the fluctuation period is less than or equal to 100ms, it is determined to be towing behavior.
[0042] Preferably, the identification results are simultaneously labeled and associated with subsequent error data.
[0043] Preferably, step five, weighing error compensation, includes: obtaining the weighing error from the calibration database based on the calculated tilt angle value, and compensating for the initial weighing value.
[0044] Preferably, the calibration database is a tilt-error calibration database specifically designed for the vehicle model.
[0045] Preferably, the calibration database is constructed by modeling the linear relationship between tilt angle and vehicle weighing error through calibration experiments, building a multivariate linear model of the two-axis tilt angle and weighing error, and achieving error quantification. Specifically, this includes: Different vehicle types (light trucks, medium trucks, heavy trucks, etc.) and different loads (empty, half-loaded, fully loaded) were selected to simulate scenarios such as scale skipping, scale towing, and off-center loading. The initial weighing value W0 and the standard weight W_std (calibrated by static scale) corresponding to different tilt angle values (α∈[-0.5°,0.5°], β∈[-0.3°,0.3°], covering the actual weighing posture offset range) were collected, and the weighing error ΔW=W0-W_std was calculated. A linear model was constructed based on calibration data to correlate tilt angle with vehicle weighing error. Based on the calibration data, the least squares method is used to fit the multivariate linear equation: ΔW = k1·α + k2·β + b, where: k1 is the pitch angle error coefficient (unit: t / °), which represents the weighing error caused by a 1° change in pitch angle. It is obtained by calibration of the jumping weighing behavior. The value of k1 for different vehicle models ranges from 0.8 to 1.5 t / °. k2 is the roll angle error coefficient (unit: t / °), which represents the weighing error caused by a change of 1° of roll angle. It is obtained by calibrating the towing scale and off-center loading behavior. The value of k2 ranges from 0.5 to 1.2 t / ° for different vehicle models. b is the reference error constant (unit: t), which is obtained by calibrating the sensor's inherent error and the road slope reference value. The default value of b is [-0.1, 0.1]t.
[0046] Preferably, after constructing the linear model using calibration data, a vehicle-specific tilt-error calibration database can be obtained. Inputting the pitch angle α and roll angle β of the corresponding vehicle model into the multivariate linear equation yields the corresponding weighing error ΔW.
[0047] Preferably, the linear model parameters are iteratively updated based on measured data. More preferably, the parameters k1, k2, and b are iteratively updated based on daily measured data (normal weighing + illegal weighing), establishing a vehicle-specific calibration library to improve the fitting accuracy of weighing errors under different scenarios.
[0048] Preferably, step five, weighing error compensation, includes: Substitute the stable tilt angles α0 and β0 into the linear model to calculate the current weighing error; The compensation is to obtain the actual weight of the vehicle; Error verification is performed. If the deviation between the compensated weight and the vehicle's load range exceeds the deviation threshold, a second calibration is performed. If the deviation between the compensated weight and the vehicle's load range does not exceed the deviation threshold, the current compensated weighing result is output.
[0049] Preferably, step five, weighing error compensation, specifically includes completing weight correction based on the identified behavior type, the calculated tilt angle value, and the linear model: Substitute the stable tilt angles α0 and β0 into the linear model to calculate the weighing error ΔW0 = k1·α0 + k2·β0 + b in the current scenario; The actual weight of the vehicle is obtained through reverse compensation: W = W0 - ΔW0 (when ΔW0 is positive, the initial weighing value is too large and the error needs to be subtracted; when ΔW0 is negative, the initial weighing value is too small and the absolute value of the error needs to be added). Error verification: If the weight W after compensation deviates from the vehicle's load range by more than ±5%, a secondary calibration is triggered (tilt angle data is re-acquired, and the latest linear model parameters from the cloud are called for correction). Output weighing result: Output the actual weight W after compensation.
[0050] Preferably, step six, outputting results, may include: outputting weighing results and violation identification results. Preferably, step six, outputting results, may further include: outputting weighing results and violation identification results.
[0051] Preferably, when performing step six, based on the identified behavior type, the calculated tilt angle value, and the linear model, weight correction and data linkage output are completed to achieve the integration of "identification-compensation-evidence collection".
[0052] The system synchronously outputs the compensated actual weight W, initial weighing value W0, tilt angle data (α0, β0), error value ΔW0, and violation identification results (normal / skipping / dragging). It also links the corresponding moments in the video images to generate a four-element evidence chain of "behavior identification results + posture data + weight curve + video footage". The data is encrypted and uploaded to the cloud for storage. All data is associated with the behavior identification results, allowing for reverse tracing of the source of weight error (violation / environmental disturbance / sensor error), thus improving data credibility and law enforcement persuasiveness.
[0053] Preferably, this embodiment also incorporates an anti-interference design for the vehicle weighing method based on behavior monitoring and error compensation, in order to eliminate interference.
[0054] Preferably, the anti-interference design can specifically involve, during step four, statistically analyzing the tilt angle change rate and determining whether it exceeds a threshold. If it does not exceed the threshold, vehicle behavior type identification and weighing error compensation are performed. If it does exceed the threshold, a sudden interference is identified, and vehicle behavior type identification and weighing error compensation are not performed. For example, when the tilt angle change rate exceeds the threshold (|dα / dt|>0.1° / ms, |dβ / dt|>0.08° / ms), it is determined to be a sudden interference (such as a vehicle hitting the weighbridge), and compensation and violation determination are temporarily not triggered.
[0055] Preferably, when a sudden disturbance occurs, the stable tilt angle data from the three time steps prior to the sudden change in tilt angle data is used for calculation; at the same time, the disturbance event is marked to avoid misjudgment and miscompensation.
[0056] Preferably, during the weighing process, this embodiment also sets up an extreme scenario linkage processing operation: when the absolute value of the tilt angle exceeds the set threshold (α>0.5°, β>0.3°), it is judged as a serious violation. While completing the weight compensation, an audible and visual warning is immediately triggered, the toll station law enforcement terminal is linked to push violation information, and the evidence chain is strengthened by encrypted storage, taking into account both measurement accuracy and law enforcement control.
[0057] The vehicle weighing method based on behavior monitoring and error compensation provided in this embodiment reuses dual-sided triaxial acceleration data to simultaneously complete "jumping / towing behavior monitoring and identification" and "dual-axis tilt angle-weighing error linear modeling compensation," forming a closed-loop processing flow. This breaks through the bottleneck of "disconnection between monitoring and compensation," achieving an integrated closed loop of "violation identification-error quantification-reverse compensation-evidence retention." It avoids the redundant investment of deploying separate monitoring equipment and compensation systems, solves the problems of functional fragmentation and data incompatibility in existing solutions, and significantly improves the operational efficiency of toll stations.
[0058] The vehicle weighing method based on behavior monitoring and error compensation provided in this embodiment achieves a dual improvement in weighing accuracy and monitoring capability: through linear modeling of dual-axis tilt angle and error, the weight measurement accuracy after compensation is ≤±0.3%, which is better than the ±1% error standard of existing dynamic weighing systems; at the same time, it achieves accurate identification of skipping and dragging of scales based on the same sensor, with an identification accuracy rate of ≥99%, taking into account both metrological accuracy and law enforcement needs.
[0059] Example 3 This embodiment provides a vehicle weighing system based on behavior monitoring and error compensation. See also... Figure 2 The vehicle weighing system includes: a weighing pressure sensing module, a three-axis acceleration sensing module symmetrically installed on both sides of the weighbridge, a cloud calibration platform, an edge computing module, a data synchronization unit, a video unit, and a cloud storage unit.
[0060] The weighing pressure sensing module is used to collect the initial weighing value of the vehicle when it passes through the weighbridge. Preferably, the weighing pressure sensing module can reuse the existing built-in pressure sensors of the weighbridge (no less than 4 sets, evenly distributed at the bottom of the weighbridge) to collect the initial weighing value W0 of the vehicle when it passes through the weighbridge, and the sampling frequency is synchronized with the accelerometer (100Hz).
[0061] Two triaxial accelerometer modules are symmetrically installed on both sides of the weighbridge to acquire triaxial acceleration data in real time. Preferably, the two modules are symmetrically installed at the front-middle section of both sides of the weighbridge. Taking a typical 3m×18m / 3m×22m weighbridge as an example, the two modules are symmetrically installed at the front-middle section of both sides, 4.5m / 5.5m from the ends of the weighbridge. Preferably, the triaxial accelerometer modules are rigidly fixed to the weighbridge surface, with the X-axis parallel to the vehicle's direction of travel, the Y-axis parallel to the width of the weighbridge, and the Z-axis perpendicular to the weighbridge surface (in the direction of gravity). Preferably, the triaxial accelerometer modules can be triaxial accelerometers with an accuracy ≥ ±0.01°, a sampling frequency of 100Hz, an IP67 protection rating, and support real-time output of raw three-dimensional acceleration data. Preferably, the triaxial accelerometers can be fixed using an L-shaped stainless steel bracket (5mm thick) and silicone shock-absorbing pads (3mm thick), requiring no modification to the existing weighbridge structure and resulting in a short installation period.
[0062] The video unit is used to capture images during the vehicle weighing process.
[0063] A cloud-based calibration platform is used to build a vehicle-specific tilt angle-error calibration database. Preferably, the cloud-based calibration platform optimizes linear model parameters based on massive amounts of measured data to build a vehicle-specific tilt angle-error calibration database, which can adapt to the weighing error characteristics of trucks with different wheelbases and load capacities.
[0064] The edge computing module performs dual-axis tilt angle calculation based on the sensor data from the triaxial acceleration sensor module to identify vehicle behavior types; it also obtains the weighing error from the calibration database based on the calculated tilt angle value, compensates for the initial weighing value, and outputs the weighing result and violation behavior identification result.
[0065] The data synchronization unit is used to synchronously transmit the sensor data from the weighing pressure sensing module and the triaxial acceleration sensing module, as well as the images acquired by the video unit, to the edge computing module. The data synchronization unit uses timestamp alignment technology to achieve millisecond-level synchronization of acceleration data, pressure data, and image data, eliminating transmission delay differences (synchronization error ≤ 1ms).
[0066] The cloud storage unit is used to store the weighing results and violation identification results output by the edge computing module, as well as the vehicle weighing images captured by the video unit.
[0067] Preferably, the vehicle weighing system based on behavior monitoring and error compensation provided in this embodiment can be used to implement the vehicle weighing method based on behavior monitoring and error compensation involved in Embodiment 1 or Embodiment 2.
[0068] The vehicle weighing system based on behavior monitoring and error compensation provided in this embodiment has strong compatibility and low modification cost: it can reuse existing weighbridge pressure sensors, only adding dual-sided triaxial accelerometers and edge computing modules, without modifying the weighbridge structure, and is compatible with both new and old toll station equipment; the same sensor performs dual functions, further reducing hardware investment and installation and maintenance costs.
[0069] The vehicle weighing system based on behavior monitoring and error compensation provided in this embodiment has wide scenario adaptability and clear traceability: through cloud-based dynamic calibration and vehicle-specific models, it adapts to trucks with different loads and wheelbases as well as various scene interferences; all data are associated with behavior recognition results, which can trace the source of weight error (violation behavior / environmental disturbance / sensor error) in reverse, thereby improving data credibility and law enforcement persuasiveness.
[0070] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A vehicle weighing method based on behavior monitoring and error compensation, characterized in that, include: Collect the initial weighing value of the vehicle; Real-time acquisition of three-axis acceleration data on both sides of the weighing platform; Build a vehicle-specific camber-error calibration database; Based on the triaxial acceleration data, the biaxial tilt angle is calculated to identify the vehicle behavior type; and based on the calculated tilt angle value, the weighing error is obtained from the calibration database to compensate the initial weighing value, and the weighing result and violation behavior identification result are output.
2. The vehicle weighing method based on behavior monitoring and error compensation according to claim 1, characterized in that, The step of calculating the biaxial tilt angle based on the triaxial acceleration data includes: Using the vector relationship between the triaxial acceleration data and gravitational acceleration, the pitch angle α and roll angle β of the weighing platform are calculated; The tilt angle data is smoothed by moving average filtering to obtain stable tilt angle values α0 and β0; Vehicle behavior type identification is performed based on tilt angle threshold and variation characteristics.
3. The vehicle weighing method based on behavior monitoring and error compensation according to claim 2, characterized in that, The construction of a vehicle-specific tilt-error calibration database includes: Different vehicle models and loads were selected to simulate illegal and normal behavior scenarios. Initial weighing values W0 and standard weights W_std corresponding to different tilt angles were collected, errors were calculated, and calibration data were obtained. Based on the calibration data, a linear model is constructed to obtain a tilt-error calibration database specific to the vehicle model.
4. The vehicle weighing method based on behavior monitoring and error compensation according to claim 3, characterized in that, The parameters of the linear model are iteratively updated based on the measured data.
5. The vehicle weighing method based on behavior monitoring and error compensation according to claim 4, characterized in that, The step of obtaining the weighing error from the calibration database based on the calculated tilt angle value and compensating for the initial weighing value includes: Substitute the stable tilt angles α0 and β0 into the linear model to calculate the current weighing error; The compensation is to obtain the actual weight of the vehicle; Error verification is performed. If the deviation between the compensated weight and the vehicle's load range exceeds the deviation threshold, a second calibration is performed. If the deviation between the compensated weight and the vehicle's load range does not exceed the deviation threshold, the weighing result and the violation identification result are output.
6. The vehicle weighing method based on behavior monitoring and error compensation according to claim 5, characterized in that, Also includes: Calculate the rate of change of inclination angle and determine whether the rate of change of inclination angle exceeds the threshold. If the weight limit is not exceeded, vehicle behavior type identification and weighing error compensation will be performed. If the threshold is exceeded, a sudden interference is determined, and vehicle behavior type identification and weighing error compensation are not performed.
7. The vehicle weighing method based on behavior monitoring and error compensation according to claim 5, characterized in that, The step of calculating the biaxial tilt angle based on the triaxial acceleration data also includes preprocessing the triaxial acceleration data, including: Filtering and normalization processing; Using the moment when the vehicle's front wheels drive onto the weighbridge as the time origin, stable data within a preset time period starting from the time origin is extracted.
8. A vehicle weighing system based on behavior monitoring and error compensation, characterized in that, include: The weighing pressure sensor module is used to collect the initial weighing value of the vehicle. The three-axis accelerometer modules are symmetrically installed on both sides of the weighing platform to acquire three-axis acceleration data in real time. A cloud-based calibration platform is used to build a vehicle-specific tilt-error calibration database; The edge computing module performs dual-axis tilt angle calculation based on the sensing data from the triaxial acceleration sensing module to identify vehicle behavior types; The weighing error is obtained from the calibration database based on the calculated tilt angle value, the initial weighing value is compensated, and the weighing result and violation identification result are output.
9. A vehicle weighing system based on behavior monitoring and error compensation according to claim 8, characterized in that, It also includes a data synchronization unit; the data synchronization unit is used to synchronously transmit the sensing data of the weighing pressure sensing module and the triaxial acceleration sensing module to the edge computing module.
10. A vehicle weighing system based on behavior monitoring and error compensation according to claim 9, characterized in that, It also includes video units and cloud storage units; The video unit is used to capture images during the vehicle weighing process; The cloud storage unit is used to store the weighing results and violation identification results output by the edge computing module, as well as the vehicle weighing images captured by the video unit.
Citation Information
Patent Citations
A method and system for monitoring abnormal vehicle weighing data without the user's awareness
CN113295253B
Vehicle overload alarm method and related device based on piezoelectric
CN119245793B
