A vehicle weighing method, apparatus, equipment and storage medium

Abstract

This invention relates to the field of vehicle weighing technology, and discloses a vehicle weighing method, device, equipment, and storage medium. The method provided by this invention first aligns multimodal sensor data with vehicle images using a time synchronization method; then, it independently calculates the initial weight of each mode based on the physical response models of piezoelectric sensors, strain gauges, and fiber optic sensors, preserving the physical characteristics of each mode; subsequently, it performs error compensation on the initial weight of each mode by combining vehicle speed, temperature, and road surface conditions, reducing environmental and dynamic interference in the actual scenario; finally, it performs dynamic weighted fusion based on the reliability results of each mode and outputs the vehicle weight monitoring result, effectively suppressing errors from using a single sensor to monitor weight, and significantly improving the accuracy and robustness of vehicle weight detection.

Description

Technical Field

[0001] This invention relates to the field of vehicle weighing technology, specifically to a vehicle weighing method, apparatus, equipment, and storage medium. Background Technology

[0002] Vehicle weighing is a crucial means of ensuring safe transportation in scenarios such as highway overload detection, logistics vehicle management, and weighing at park entrances and exits. While fixed weighbridges monitor vehicle weight, they require foundation construction, installation, and external power supply. Furthermore, vehicles must stop or pass slowly before reaching the weighbridge, significantly impacting road traffic efficiency. Therefore, achieving dynamic vehicle weighing is a major challenge for safe transportation today.

[0003] A dynamic vehicle weighing method is disclosed in the related technology, which collects the pressure signal when the vehicle passes through a temporarily set pressure sensor, and then uses a data-driven model to directly estimate the vehicle weight based on the pressure signal.

[0004] However, the data collected by sensors in the vehicle dynamic weighing methods disclosed in related technologies are affected by the external environment in the actual scenario, resulting in a large error in the final vehicle weight estimation result, which is difficult to meet the accuracy requirements of current vehicle weight monitoring. Summary of the Invention

[0005] This invention provides a vehicle weighing method, apparatus, device, and storage medium to address the problem that the vehicle dynamic weighing methods disclosed in related technologies are insufficient to meet the accuracy requirements of current vehicle weight monitoring.

[0006] In a first aspect, the present invention provides a vehicle weighing method, the method comprising: Based on the collected multimodal weight monitoring data, combined with the vehicle images and acquisition time captured by the camera, time synchronization is performed using a time synchronization method to obtain the vehicle speed parameters and weight monitoring data of multiple modes after time synchronization; the multimodal weight monitoring data includes piezoelectric sensing monitoring data, strain gauge sensing monitoring data and fiber optic sensing monitoring data. Based on the weight monitoring data after time synchronization of each mode, the data is processed using their respective pre-built physical response models to obtain the initial weight for each mode. Based on the initial weight of each mode, combined with vehicle speed parameters, temperature parameters and road condition parameters, the error compensation method is used to correct the weight and reliability of the corresponding mode. By combining the corrected weight assessment results and reliability assessment results for each mode, a weighted fusion method is used to obtain the final vehicle weight monitoring result.

[0007] By adopting the above implementation method, firstly, the multimodal sensing data and vehicle images are aligned using a time synchronization method; then, the initial weight of each mode is independently calculated based on the physical response models of piezoelectric sensors, strain gauges, and fiber optic sensors, preserving the physical characteristics of each mode; subsequently, error compensation is performed on the initial weight of each mode in combination with vehicle speed, temperature, and road conditions to reduce environmental and dynamic interference in the actual scene; finally, dynamic weighted fusion is performed based on the reliability results of each mode, and the vehicle weight monitoring results are output, effectively suppressing the error of using a single sensor to monitor weight and significantly improving the accuracy and robustness of vehicle weight detection.

[0008] In one optional implementation, the step of correcting the initial weight of each mode using error compensation methods, combined with vehicle speed parameters, temperature parameters, and road condition parameters, to obtain the corrected weight assessment result and reliability assessment result for the corresponding mode, includes: Based on the initial weight of each mode, combined with the vehicle speed parameters, the speed correction coefficient is used to correct the weight of the corresponding mode after speed correction. Based on the velocity-corrected weight for each mode, combined with temperature parameters, the temperature drift coefficient is used for correction to obtain the temperature-corrected weight for the corresponding mode. Based on the temperature-corrected weight of each mode, combined with road surface condition parameters, the road surface correction term is used for correction to obtain the corrected weight assessment result for the corresponding mode. Based on the corrected weight assessment results for each mode, combined with the signal-to-noise ratio of the corresponding mode, the deviation between the weight assessment results of other modes, and the historical drift, the reliability assessment results of the corrected weight assessment results for the corresponding mode are obtained.

[0009] By adopting the above implementation method, firstly, the impact of vehicle speed changes on the vehicle weight assessment of each mode is eliminated by using a speed correction coefficient; then, a temperature drift coefficient is used to compensate for sensor zero drift caused by ambient temperature; furthermore, the additional load introduced by road surface unevenness is corrected by a road surface correction term, thereby obtaining the corrected weight of each mode; finally, the reliability assessment results of the corrected weight assessment results of the mode are obtained by comprehensively considering the signal-to-noise ratio, inter-mode deviation, and historical drift amount, and the interference caused by vehicle speed, temperature, and road surface on the vehicle weight estimation results of each mode are eliminated step by step, and a credibility index is constructed for each mode, laying a high-quality data foundation for subsequent dynamic fusion.

[0010] In one optional implementation, the reliability assessment result of the corrected weight assessment result for each mode, based on the corrected weight assessment result for each mode, combined with the signal-to-noise ratio of the corresponding mode, the deviation between the weight assessment results of other modes, and the historical drift, includes: Based on the corrected weight assessment results of each mode, and combined with the corrected weight assessment results of other modes, the inter-modal deviation of the corresponding mode is obtained by using the deviation evaluation method. Based on the intermodal deviation of each mode, combined with the signal-to-noise ratio of the corresponding mode and the historical drift, the original confidence level of the mode is obtained by using a weighted summation or nonlinear mapping method. Based on the original confidence level of each modality, a preset reliability level threshold is used for comparison to determine the reliability level or normalized confidence score of the modality, and the output is the reliability assessment result of the corrected weight assessment result of the corresponding modality.

[0011] By adopting the above implementation method, firstly, the deviation evaluation method is used to compare the corrected weight assessment results of different modes to ensure timely identification of abnormal data modes; then, the signal-to-noise ratio, historical drift, and inter-modal deviation of each mode are fused, and the original credibility is obtained by weighted summation or nonlinear mapping, realizing a comprehensive quantification of the quality of the weight assessment results; finally, the reliability level or normalized credibility score is output through threshold grading to provide a clear quality label for each mode, so that the subsequent dynamic weight fusion can automatically allocate weights according to credibility, significantly improving the robustness and accuracy of vehicle weighing under complex working conditions.

[0012] In one optional implementation, when the time-synchronized weight monitoring data includes piezoelectric sensing monitoring data, the step of processing the time-synchronized weight monitoring data for each mode using its respective pre-built physical response model to obtain the initial weight corresponding to each mode includes: Based on the time-synchronized piezoelectric sensing monitoring data, the time-domain integral area and maximum peak value are extracted using the piezoelectric physical response model and output as piezoelectric response characteristic quantities. Based on the piezoelectric response characteristics, and combined with the pre-calibrated piezoelectric physical response model coefficients, the initial weight corresponding to the piezoelectric branch is calculated.

[0013] By adopting the above implementation method, for piezoelectric sensing monitoring data, the time-domain integral area and maximum peak value are first extracted from the time-synchronized monitoring data as piezoelectric response feature quantities, which effectively characterize the intensity and impact characteristics of the piezoelectric signal generated when the vehicle axle load passes the sensor; then, the initial weight is calculated by combining the pre-calibrated physical model coefficients to ensure that a quantifiable physical mapping is established between the piezoelectric response feature quantities and the actual weight of the vehicle, making full use of the sensitivity advantage of piezoelectric sensors to dynamic loads, and providing accurate and reliable piezoelectric branch weight estimation for subsequent multimodal fusion.

[0014] In one optional implementation, when the time-synchronized weight monitoring data includes strain gauge sensing data, the step of processing the time-synchronized weight monitoring data for each mode using its respective pre-built physical response model to obtain the initial weight for each mode includes: Based on the strain gauge sensing and monitoring data after time synchronization, the strain response of the strain gauge is obtained by conversion using the first feature conversion method. Based on the strain response of the strain gauge, the time-domain integral area and maximum peak value are extracted using the strain physics response model and output as strain response characteristic quantities. Based on the strain response characteristics, the initial weight corresponding to the strain branch is calculated using the pre-calibrated strain physical response model coefficients.

[0015] By adopting the above implementation method, for strain gauge sensing monitoring data, the monitoring data is first converted into strain gauge strain response using the first feature conversion method to eliminate the influence of electrical signal drift and bridge nonlinearity on the original data; then, the time-domain integral area and maximum peak value are extracted from the strain gauge strain response to characterize the strain accumulation effect and peak load characteristics, respectively, providing two-dimensional feature input for subsequent weight estimation; finally, the initial weight is calculated by combining the pre-calibrated strain physical response model coefficients, so that a close physical mapping is established between the strain gauge strain response and the axle load of the vehicle under test, providing a stable and reliable strain branch weight estimate for subsequent multimodal fusion.

[0016] In one optional implementation, when the time-synchronized weight monitoring data includes fiber optic sensing monitoring data, the step of processing the time-synchronized weight monitoring data for each mode using its respective pre-built physical response model to obtain the initial weight corresponding to each mode includes: Based on the time-synchronized fiber optic sensing monitoring data, the second feature conversion method is used to convert the data, and the output is fiber optic mechanical strain. Based on the optical fiber mechanical strain, the time-domain integral area and maximum peak value are extracted using the optical fiber physical response model and output as optical fiber response characteristic quantities. Based on the fiber response characteristics, the initial weight of the fiber branch is calculated using the pre-calibrated fiber physical response model coefficients.

[0017] By adopting the above implementation method, for fiber optic sensing monitoring data, the original monitoring data is first converted into fiber optic mechanical strain using the second feature conversion method, effectively eliminating the influence of non-load factors such as light intensity fluctuations and coupling losses on measurement accuracy. Then, the time-domain integral area and maximum peak value are extracted from the fiber optic mechanical strain to characterize the strain energy accumulation and instantaneous deformation extreme value, respectively, providing two-dimensional feature input for subsequent weight estimation. Finally, the initial weight is calculated by combining the pre-calibrated fiber optic physical response model coefficients, realizing a precise physical mapping between the fiber optic strain response characteristic quantity and the axle load of the measured vehicle, providing a stable fiber optic branch weight estimate for subsequent multi-mode fusion.

[0018] In one optional implementation, the step of combining the collected multimodal weight monitoring data with vehicle images captured by a camera and the acquisition time, and then using a time synchronization method to perform time synchronization, yields time-synchronized vehicle speed parameters and multimodal weight monitoring data, including: Based on the peak trigger time of piezoelectric sensing monitoring data, the rising edge time of strain gauge sensing monitoring data, and the timestamp of camera image frames, the data are aligned using the same time coordinate to obtain the alignment result of each modal data on a unified time axis. Based on the trigger time difference of piezoelectric sensor monitoring data set adjacent to each other along the vehicle's direction of travel or the vehicle's displacement in consecutive frames of camera images, the vehicle speed parameters are obtained using a vehicle speed evaluation method. Based on the vehicle speed parameters, the duration of multiple modes is mapped to the same time window using a time scale normalization method to obtain time-synchronized multimodal weight monitoring data. The output is time-synchronized vehicle speed parameters and weight monitoring data of multiple modes.

[0019] By employing the above implementation method, the peak time of piezoelectric sensing monitoring data, the rising edge time of strain gauge sensing monitoring data, and the timestamp of camera image frames are used to perform unified timeline alignment, ensuring the consistency of multi-source data in event timing. Then, the vehicle speed is calculated based on the time difference between adjacent piezoelectric triggering or image displacement, providing parameter basis for subsequent speed correction. Finally, the duration of multimodal data is normalized and mapped using vehicle speed to eliminate the difference in vehicle passage time caused by changes in vehicle speed, making the monitoring data of each modality comparable within a unified time window. This lays a precise time reference for subsequent physical response modeling and dynamic fusion for each modal branch, significantly improving the accuracy of the final weight detection results.

[0020] In a second aspect, the present invention provides a vehicle weighing device, the device comprising: The time synchronization module is used to synchronize the collected multimodal weight monitoring data with the vehicle images and acquisition time captured by the camera using a time synchronization method, so as to obtain the time-synchronized vehicle speed parameters and weight monitoring data of multiple modes; the multimodal weight monitoring data includes piezoelectric sensing monitoring data, strain gauge sensing monitoring data and fiber optic sensing monitoring data. The initial evaluation module is used to process the weight monitoring data after time synchronization of each mode using their respective pre-built physical response models to obtain the initial weight for each mode. The error correction module is used to correct the initial weight of each mode by combining vehicle speed parameters, temperature parameters and road condition parameters, and using error compensation methods to obtain the corrected weight assessment result and reliability assessment result for the corresponding mode. The results output module integrates the corrected weight assessment results and reliability assessment results of each mode, and performs weighted fusion using a dynamic weight fusion method to obtain the final vehicle weight monitoring result.

[0021] Thirdly, the present invention provides an electronic device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the vehicle weighing device of the first aspect or any corresponding embodiment described above.

[0022] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the vehicle weighing device of the first aspect or any corresponding embodiment described above. Attached Figure Description

[0023] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the first process of a vehicle weighing method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of acquiring multimodal weight monitoring data in a vehicle weighing method according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a second process for a vehicle weighing method according to an embodiment of the present invention; Figure 4This is a schematic diagram of the third process of the vehicle weighing method according to an embodiment of the present invention; Figure 5 This is a structural block diagram of a vehicle weighing device according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0027] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0028] In the vehicle weighing methods disclosed in related technologies, the pressure signal when the vehicle passes through a sensor is collected, and the vehicle's weight is estimated by inverting the pressure signal. However, the vehicle weighing methods disclosed in related technologies have at least the following drawbacks: 1. The different speeds of the vehicle being tested as it passes the sensor will cause changes in the peak value, duration, and phase relationship of the signal collected by the sensor, which in turn will lead to a large error between the estimated weight of the vehicle being tested and the actual axle load and total weight of the vehicle. 2. When the sensor is used in the external environment, it is subject to temperature drift and long-term zero drift, requiring frequent manual calibration by staff.

[0029] To overcome the shortcomings of vehicle weighing methods disclosed in related technologies, this embodiment provides a vehicle weighing method. First, it aligns multimodal sensor data with vehicle images using a time synchronization method. Then, it independently calculates the initial weight of each mode based on the physical response models of piezoelectric sensors, strain gauges, and fiber optic sensors, preserving the physical characteristics of each mode. Next, it performs error compensation on the initial weight of each mode by combining vehicle speed, temperature, and road surface conditions, reducing environmental and dynamic interference in the actual scenario. Finally, it performs dynamic weighted fusion based on the reliability results of each mode and outputs the vehicle weight monitoring result, effectively suppressing errors from using a single sensor to monitor weight and significantly improving the accuracy and robustness of vehicle weight detection.

[0030] According to an embodiment of the present invention, a vehicle weighing method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0031] This embodiment provides a vehicle weighing method that can be used in a vehicle weight monitoring server. Figure 1 This is a flowchart of a vehicle weighing method according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps: S101, based on the collected multimodal weight monitoring data, combined with the vehicle images and acquisition time captured by the camera, time synchronization is performed using a time synchronization method to obtain the vehicle speed parameters and weight monitoring data of multiple modes after time synchronization; the multimodal weight monitoring data includes piezoelectric sensing monitoring data, strain gauge sensing monitoring data and fiber optic sensing monitoring data.

[0032] Multimodal weight monitoring data refers to raw signals related to vehicle weight collected simultaneously by multiple different types of sensors, including: piezoelectric sensing monitoring data, which uses piezoelectric sensors to collect monitoring data reflecting tire impact and dynamic load changes; strain gauge sensing monitoring data, which uses strain gauge sensors to collect monitoring data reflecting local bending or deformation of the sensor pad; and fiber optic sensing monitoring data, which uses fiber Bragg grating sensors to collect strain and temperature information reflecting when the vehicle passes by.

[0033] Reference Figure 2A multimodal sensor pad integrating piezoelectric sensors, strain gauge sensors, and fiber optic sensors is laid on the surface of the lane to be inspected, ensuring close contact between the sensor pad and the road surface. It is then secured with ground spikes, clamps, edge strips, or detachable fasteners. A camera is installed in front of or to the side of the sensor pad, and the camera is calibrated to ensure that the image coordinates correspond to the sensor pad coordinates. When a vehicle approaches the sensor pad, the piezoelectric sensor first generates a weak response. After the camera detects that the vehicle has entered the preset area, it begins to continuously acquire image frames. As the vehicle tires pass through the sensor pad in sequence, the piezoelectric sensor, strain gauge sensor, and fiber optic sensor simultaneously generate multiple signals, which serve as multimodal weight monitoring signals.

[0034] The time synchronization method is a way to align data collected by different types of sensors and cameras on the time axis, so that the data collected by different types of sensors and cameras have a unified time coordinate. For example, a synchronization reference is established by setting multiple rows of piezoelectric trigger times, strain signal rising edges, camera frame timestamps, etc.

[0035] The vehicle speed parameter is the vehicle speed determined by the time difference of the front and rear sensor triggers or the vehicle displacement in consecutive frames of camera images. It is used for subsequent signal normalization and speed compensation.

[0036] By precisely aligning multimodal weight monitoring data and image information in time, the temporal consistency of subsequent feature extraction and fusion calculation is ensured, avoiding phase errors caused by signal misalignment. At the same time, by acquiring vehicle speed parameters, a quantitative basis is provided for subsequent speed compensation and time scale normalization, solving the defect that multimodal weight monitoring data cannot be directly fused, and facilitating improved adaptability to subsequent vehicle speed changes.

[0037] S102, based on the weight monitoring data after time synchronization of each mode, process them using their respective pre-built physical response models to obtain the initial weight of each mode.

[0038] The physical response model is a mathematical model constructed based on the piezoelectric effect, resistance strain effect, or fiber Bragg grating strain and temperature coupling effect of different types of sensors. It is used to convert the electrical signals collected by each type of sensor into intermediate data corresponding to the vehicle weight monitoring data represented by each sensor.

[0039] The initial weight is a preliminary weight value estimated independently based on individual types of weight monitoring data, including piezoelectric initial weight, strain initial weight, and fiber initial weight, which serves as a reference value for subsequent error compensation and fusion.

[0040] Specifically, when the weight monitoring data after time synchronization includes piezoelectric sensing monitoring data, the above S102 can be implemented as follows: a1, based on the time-synchronized piezoelectric sensing monitoring data, uses the piezoelectric physical response model to extract the time-domain integral area and maximum peak value, and outputs them as piezoelectric response characteristic quantities; a2, based on the piezoelectric response characteristic quantities and combined with the pre-calibrated piezoelectric physical response model coefficients, is used to calculate the initial weight corresponding to the piezoelectric branch.

[0041] For example, the above S102 can be implemented as follows:

[0042]

[0043]

[0044] in, This indicates the piezoelectric branch corresponding to the test vehicle's first... The initial weight of each axle; The first element extracted from the piezoelectric physical response model represents the... The action time of each axle to The integral area between them; The first element extracted from the piezoelectric physical response model represents the... The action time of each axle to The maximum peak value between; This indicates the corresponding number of piezoelectric sensing monitoring data monitored. Each axle is in The response voltage at time; , and This represents the coefficients of the piezoelectric physical response model after calibration using a vehicle with a known axle load.

[0045] First, the time-domain integral area and maximum peak value are extracted from the time-synchronized piezoelectric sensor monitoring data as piezoelectric response feature quantities, which effectively characterize the intensity and impact characteristics of the piezoelectric signal generated when the vehicle axle load passes the sensor. Then, the initial weight is calculated by combining the pre-calibrated physical model coefficients to ensure that a quantifiable physical mapping is established between the piezoelectric response feature quantities and the actual weight of the vehicle. This fully utilizes the sensitivity of piezoelectric sensors to dynamic loads and provides accurate and reliable piezoelectric branch weight estimation for subsequent multimodal fusion.

[0046] Specifically, when the weight monitoring data after time synchronization includes strain gauge sensing monitoring data, the above S102 can be implemented as follows: b1, Based on the strain gauge sensing and monitoring data after time synchronization, the strain gauge strain response is obtained by conversion using the first feature conversion method; b2, based on the strain gauge strain response, uses the strain physical response model to extract the time-domain integral area and maximum peak value, and outputs them as strain response characteristic quantities; b3, based on strain response characteristic quantities and combined with pre-calibrated strain physical response model coefficients, calculates the initial weight corresponding to the strain branch.

[0047] The first feature conversion method refers to the calculation method that converts the resistance change of strain gauge sensing data into a physical strain value. It can be obtained based on the physical principle that the resistance change of the strain gauge is proportional to the strain. The initial resistance is the resistance value when there is no load, and the strain factor (GF) is determined by the sensor manufacturing parameters.

[0048] For example, the above S102 satisfies:

[0049]

[0050]

[0051]

[0052] in, This indicates the branch of the strain gauge corresponding to the first test vehicle. The initial weight of each axle; The first expression extracted from the strain physical response model The action time of each axle to The integral area between them; The first expression extracted from the strain physical response model The action time of each axle to The maximum peak value between; , and These represent the coefficients of the strain physical response model after calibration using a vehicle with a known axle load. This indicates the corresponding number of strain gauge sensing monitoring data monitored. Each axle is in The strain response at any moment; This represents the change in resistance in the data monitored by the strain gauge. This represents the initial resistance of the strain gauge sensor; This represents the strain coefficient.

[0053] First, the strain gauge sensing data is converted into strain gauge strain response using the first feature conversion method, eliminating the influence of electrical signal drift and bridge nonlinearity on the original data. Then, the time-domain integral area and maximum peak value are extracted from the strain gauge strain response to characterize the strain accumulation effect and peak load characteristics, respectively, providing two-dimensional feature input for subsequent weight estimation. Finally, the initial weight is calculated by combining the pre-calibrated strain physical response model coefficients, establishing a close physical mapping between the strain gauge strain response and the axle load of the vehicle under test, providing a stable and reliable strain branch weight estimate for subsequent multimodal fusion.

[0054] Specifically, when the weight monitoring data after time synchronization includes fiber optic sensing monitoring data, the above S102 can be implemented as follows: c1, based on the time-synchronized fiber optic sensing monitoring data, is converted using the second feature conversion method, and the output is fiber optic mechanical strain. c2, based on fiber mechanical strain, uses the fiber physical response model to extract the time-domain integral area and maximum peak value, and outputs them as fiber response characteristic quantities; c3, based on the fiber response characteristic quantities and combined with the pre-calibrated fiber physical response model coefficients, calculates the initial weight corresponding to the fiber branch.

[0055] The second feature conversion method is to convert the fiber Bragg wavelength offset into mechanical strain after temperature decoupling based on the pre-calibrated "fiber strain characteristic-load" mapping relationship.

[0056] For example, the above S102 can be implemented as follows:

[0057]

[0058]

[0059]

[0060] in, This indicates the first branch of the fiber optic cable corresponding to the vehicle under test. The initial weight of each axle; The first element extracted from the fiber optic physical response model represents the... The action time of each axle to The integral area between them; The first element extracted from the fiber optic physical response model represents the... The action time of each axle to The maximum peak value between; , and All of these represent the coefficients of the fiber optic physical response model after calibration using a vehicle with a known axle load. This indicates the corresponding number of fiber optic sensing monitoring data monitored. Each axle is in The mechanical strain response of the optical fiber at time t; The first fiber optic sensor monitoring data monitoring Bragg wavelength offset under the action of a single axle; Indicates the initial Bragg wavelength; The temperature change monitored by fiber optic sensing data is obtained using a temperature compensation grating, an independent temperature sensor, or a reference grating not subjected to wheel load. Indicates the strain sensitivity coefficient; This represents the temperature sensitivity coefficient.

[0061] First, the fiber optic sensing monitoring data is converted into fiber optic mechanical strain using the second feature conversion method, effectively eliminating the influence of non-load factors such as light intensity fluctuations and coupling losses on measurement accuracy. Then, the time-domain integral area and maximum peak value are extracted from the fiber optic mechanical strain to characterize the strain energy accumulation and instantaneous deformation extreme value, respectively, providing two-dimensional feature inputs for subsequent weight estimation. Finally, the initial weight is calculated by combining the pre-calibrated fiber optic physical response model coefficients, realizing a precise physical mapping between the fiber optic strain response characteristic quantity and the axle load of the measured vehicle, providing a stable fiber optic branch weight estimate for subsequent multi-mode fusion.

[0062] The sensor data for each modality are processed separately through a physical response model to obtain an interpretable initial weight estimate. This ensures that the path for estimating vehicle weight from the weight monitoring data of each modality is clearly visible, facilitating verification and traceability. At the same time, by independently calculating the initial weight of each modality, a reliable input is provided for subsequent consistency assessment and dynamic weight allocation, thereby ensuring the accuracy of the subsequent vehicle weight monitoring results.

[0063] S103, based on the initial weight of each mode, combined with vehicle speed parameters, temperature parameters and road surface condition parameters, the error compensation method is used to correct the weight and reliability assessment results of the corresponding modes.

[0064] The temperature parameter is derived from the temperature information decoupled from the fiber optic sensor or from an independent temperature sensor, and is used to correct data errors caused by the temperature in the actual scene by the piezoelectric sensor or strain gauge sensor.

[0065] Road surface condition parameters are indicators used to reflect the contact quality between the sensor pad and the road surface, the road surface slope, or local unevenness. They are obtained by detecting the internal acceleration signals of sensors in each mode, the response differences of adjacent sensors, or attitude sensors.

[0066] Error compensation methods are the process of correcting the initial weight for interference factors such as speed, temperature, and road surface conditions in real-world scenarios. These methods include speed compensation, temperature compensation, and road surface condition compensation.

[0067] The reliability assessment results comprehensively consider factors such as the signal-to-noise ratio of each mode, the deviation from the estimates of other modes, and the amount of historical drift, and output a confidence score or reliability level to quantify the reliability of the corrected weight estimate for each mode.

[0068] The interference factors of speed, temperature and road surface are compensated in turn to keep the weight estimate of each mode stable under different working conditions. By calculating the reliability assessment results, the credibility of the current output result of each mode is identified, effectively suppressing the impact of single sensor anomaly or environmental disturbance on the final result.

[0069] S104. The corrected weight assessment results and reliability assessment results of each mode are combined and weighted by a dynamic weight fusion method to obtain the final vehicle weight monitoring result.

[0070] The dynamic weight fusion method dynamically allocates the weight assessment results of each modality in the final weight calculation based on the reliability assessment results of the weight assessment results after modality correction, and performs a weighted summation of the corrected weights of each modality. At the same time, if the reliability assessment result of a certain modality is lower than the threshold, the weight of the corrected weight assessment result of its corresponding modality is reduced or removed.

[0071] For example, the dynamic weight fusion method in S104 above satisfies:

[0072]

[0073] in, This indicates the final vehicle weight monitoring result; Indicates the first The weights of the modal-corrected weight assessment results; Indicates the first Modal-corrected weight assessment results; This represents a nonlinear residual compensation term, which is obtained by training a regression model, gradient boosting tree, support vector regression, shallow neural network, convolutional network, recurrent network or other trainable model by minimizing the residual between the actual weight of the calibrated vehicle and the weight fused from the physical model as the optimization objective. Indicates the first Reliability assessment results of the modal-corrected weight assessment results; This represents a set of reliability assessments for the corrected weight assessment results across multiple modes. This represents a set of weight monitoring data across multiple modalities.

[0074] By utilizing weight assessment results from multiple modalities and achieving adaptive fusion through dynamic weighting, when a sensor is affected by noise, drift, or local damage, its weight is automatically reduced, while the contribution of high-reliability sensors increases, thereby ensuring the accuracy and robustness of the final weight result. This approach is suitable for actual operating scenarios under different road surfaces, temperatures, and vehicle speeds.

[0075] In this embodiment, a vehicle weight monitoring method is provided. First, multimodal sensor data and vehicle images are aligned using a time synchronization method. Then, the initial weight of each mode is calculated independently based on the physical response models of piezoelectric sensors, strain gauges, and fiber optic sensors, preserving the physical characteristics of each mode. Next, error compensation is performed on the initial weight of each mode in combination with vehicle speed, temperature, and road conditions to reduce environmental and dynamic interference in the actual scene. Finally, dynamic weighted fusion is performed based on the reliability results of each mode, and the vehicle weight monitoring result is output. This effectively suppresses the error of using a single sensor to monitor weight and significantly improves the accuracy and robustness of vehicle weight detection.

[0076] This embodiment provides a vehicle weighing method that can be used in a vehicle weight monitoring server. Figure 3 This is a flowchart of a vehicle weighing method according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps: S301, based on the collected multimodal weight monitoring data, combined with the vehicle images and acquisition time captured by the camera, uses a time synchronization method to synchronize the time, and obtains the vehicle speed parameters and weight monitoring data of multiple modes after time synchronization; the multimodal weight monitoring data includes piezoelectric sensing monitoring data, strain gauge sensing monitoring data and fiber optic sensing monitoring data.

[0077] Specifically, S301 includes: S3011 uses the peak trigger time of piezoelectric sensing monitoring data, the rising edge time of strain gauge sensing monitoring data, and the timestamp of camera image frames to align them using the same time coordinate, thus obtaining the alignment result of each modal data on a unified time axis.

[0078] S3012, based on the trigger time difference of piezoelectric sensor monitoring data set adjacent to each other along the vehicle's forward direction or the vehicle's displacement in consecutive frame images from a camera, the vehicle speed parameters are obtained using a vehicle speed evaluation method.

[0079] Specifically, the above S2012 can be implemented as follows:

[0080] in, Indicates vehicle speed parameters; This indicates the interval between two adjacent piezoelectric sensors positioned along the vehicle's direction of travel. This indicates the peak moment when any tire of the tested vehicle triggers the second row of piezoelectric sensors; This indicates the peak moment when any tire of the tested vehicle triggers the first row of piezoelectric sensors.

[0081] S3013, based on vehicle speed parameters, uses a time-scale normalization method to map the duration of multiple modes to the same time window, obtaining time-synchronized multimodal weight monitoring data, and outputs time-synchronized vehicle speed parameters and weight monitoring data of multiple modes.

[0082] Time synchronization is achieved by using the peak time of the piezoelectric signal, the rising edge time of the strain signal, and the tire contact position detected by the camera. Meanwhile, when the estimated speed of the peak time difference of the piezoelectric sensors set in multiple rows is inconsistent with the estimated speed of the image acquisition time obtained by the camera, a reliable speed is selected based on the consistency score of multiple modes, and the weight monitoring data of the inconsistent modes is output as an abnormal state record.

[0083] By using the peak moments of piezoelectric sensing data, the rising edge moments of strain gauge sensing data, and the timestamps of camera image frames to achieve unified timeline alignment, the consistency of multi-source data in event timing is ensured. Then, vehicle speed is calculated based on the time difference between adjacent piezoelectric triggers or image displacement, providing parameter basis for subsequent speed correction. Finally, the duration of multimodal data is normalized using vehicle speed to eliminate differences in vehicle passage time caused by speed variations, making the monitoring data of each modality comparable within a unified time window. This lays a precise time reference for subsequent physical response modeling and dynamic fusion for each modal branch, significantly improving the accuracy of the final weight detection results.

[0084] S302, based on the weight monitoring data after time synchronization of each mode, processes the data using its respective pre-built physical response model to obtain the initial weight for each mode. For details, please refer to [link to relevant documentation]. Figure 1 S102 of the illustrated embodiment will not be described again here.

[0085] S303, based on the initial weight of each mode, and combined with vehicle speed parameters, temperature parameters, and road condition parameters, uses error compensation methods to correct for each mode, obtaining the corrected weight assessment result and reliability assessment result for that mode. For details, please refer to [link to relevant documentation]. Figure 1 S103 of the illustrated embodiment will not be described again here.

[0086] S304: The corrected weight assessment results and reliability assessment results for each mode are combined and weighted using a dynamic weight fusion method to obtain the final vehicle weight monitoring result. For details, please refer to [link to relevant documentation]. Figure 1 S104 of the illustrated embodiment will not be described again here.

[0087] This embodiment provides a vehicle weighing method that can be used in a vehicle weight monitoring server. Figure 4 This is a flowchart of a vehicle weighing method according to an embodiment of the present invention, such as... Figure 4 As shown, the process includes the following steps: S401, based on the acquired multimodal weight monitoring data, combined with vehicle images captured by a camera and the acquisition time, uses a time synchronization method to achieve time synchronization, obtaining the synchronized vehicle speed parameters and weight monitoring data for multiple modes; the multimodal weight monitoring data includes piezoelectric sensing monitoring data, strain gauge sensing monitoring data, and fiber optic sensing monitoring data. For details, please refer to... Figure 1 S101 of the illustrated embodiment will not be described again here.

[0088] S402, based on the weight monitoring data after time synchronization of each mode, processes it using its own pre-built physical response model to obtain the initial weight of each mode.

[0089] S403, based on the initial weight of each mode, combined with vehicle speed parameters, temperature parameters and road condition parameters, uses error compensation methods to correct the weight and reliability of the corresponding mode.

[0090] Specifically, S403 includes: S4031, based on the initial weight of each mode, combined with the vehicle speed parameters, the speed correction coefficient is used to correct the weight of the corresponding mode after speed correction.

[0091] For example, S4031 above includes:

[0092]

[0093] in, This represents the weight after velocity correction for any mode; This indicates a piezoelectric branch, strain gauge branch, or optical fiber branch. This represents the velocity correction factor for the corresponding mode; Indicates vehicle speed parameters; , and This indicates the velocity calibration parameters for the corresponding mode.

[0094] S4032, based on the velocity-corrected weight of each mode, combined with temperature parameters, and corrected using the temperature drift coefficient, to obtain the temperature-corrected weight of the corresponding mode.

[0095] S4033, based on the temperature-corrected weight of each mode, combined with road surface condition parameters, and corrected using road surface correction terms, obtains the corrected weight assessment result for the corresponding mode.

[0096] For example, the corrected weight evaluation results for each mode in the above 4033 satisfy the following:

[0097]

[0098] in, Indicates the first Corrected weight assessment results for each mode; Indicates the first Temperature compensation term for each mode; Indicates the current temperature; Indicates the calibrated temperature; Indicates the first Temperature drift coefficient of each mode; This represents the road surface correction term, used to compensate for errors caused by insufficient contact between the sensor and the road surface, road slope, local unevenness, or displacement of the pad. It is obtained based on the collected acceleration signal, the difference in response of adjacent sensors, baseline disturbance when the vehicle passes, or the output of the installation attitude sensor.

[0099] S4034. Based on the corrected weight assessment results for each mode, combined with the signal-to-noise ratio of the corresponding mode, the deviation between the weight assessment results of other modes, and the historical drift, the reliability assessment results of the corrected weight assessment results for the corresponding mode are obtained.

[0100] Specifically, S4034 above can be implemented as follows: d1, based on the corrected weight assessment results of each mode and combined with the corrected weight assessment results of other modes, the inter-modal deviation of the corresponding mode is obtained by using the deviation evaluation method respectively; d2, based on the intermodal deviation of each mode, combined with the signal-to-noise ratio of the corresponding mode and the historical drift, the original confidence level of the mode is obtained by using a weighted summation or nonlinear mapping method; d3, based on the original confidence level of each modality, compares it with the preset reliability level threshold, determines the reliability level or normalized confidence score of the modality, and outputs the reliability assessment result of the corrected weight assessment result of the corresponding modality.

[0101] For example, S4034 above can be implemented as follows:

[0102] in, Indicates the first Reliability assessment results of the modal-corrected weight assessment results; Indicates the first The signal-to-noise ratio of the modal; Indicates the first The deviation between the modified weight estimate and the median estimate of other modes; Indicates the first Modal history offset; Indicates the first Temperature sensitivity index of modality; Indicates the first Modal condition matching degree; The confidence function can be implemented as a rule function, a lookup table function, or a regression function trained from calibration data.

[0103] By utilizing a bias evaluation method and comparing the corrected weight assessment results of different modalities, abnormal modalities can be identified in a timely manner. Then, by integrating the signal-to-noise ratio, historical drift, and inter-modal bias of each modality, the original reliability is obtained through weighted summation or nonlinear mapping, thus achieving a comprehensive quantification of the quality of the weight assessment results. Finally, by using threshold grading to output reliability levels or normalized reliability scores, clear quality indicators are provided for each modality, enabling subsequent dynamic weight fusion to automatically allocate weights based on reliability, significantly improving the robustness and accuracy of vehicle weighing under complex working conditions.

[0104] First, a speed correction coefficient is used to eliminate the impact of vehicle speed changes on the vehicle weight assessment of each mode. Then, a temperature drift coefficient is used to compensate for sensor zero drift caused by ambient temperature. Next, a road surface correction term is used to correct the additional load introduced by road surface unevenness, thereby obtaining the corrected weight of each mode. Finally, the reliability assessment results of the corrected weight assessment results of the quantification mode are obtained by combining the signal-to-noise ratio, inter-modal deviation, and historical drift. The interference caused by vehicle speed, temperature, and road surface on the vehicle weight estimation results of each mode is eliminated step by step, and a credibility index is constructed for each mode, laying a high-quality data foundation for subsequent dynamic fusion.

[0105] S404 combines the corrected weight assessment results and reliability assessment results from each mode, and uses a dynamic weight fusion method to perform weighted fusion to obtain the final vehicle weight monitoring result. For details, please refer to [link to relevant documentation]. Figure 1 S104 of the illustrated embodiment will not be described again here.

[0106] This embodiment also provides a vehicle weighing device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0107] This embodiment provides a vehicle weighing device, such as... Figure 5 As shown, it includes: The time synchronization module 510 is used to synchronize the vehicle speed parameters and multiple modes of weight monitoring data based on the collected multimodal weight monitoring data, combined with the vehicle images and acquisition time captured by the camera, using a time synchronization method. The multimodal weight monitoring data includes piezoelectric sensing monitoring data, strain gauge sensing monitoring data, and fiber optic sensing monitoring data. The initial evaluation module 520 is used to process the weight monitoring data after time synchronization of each mode using their respective pre-built physical response models to obtain the initial weight of each mode. The error correction module 530 is used to correct the weight of each mode based on the initial weight, combined with vehicle speed parameters, temperature parameters and road surface condition parameters, using error compensation methods to obtain the corrected weight assessment result and reliability assessment result for the corresponding mode. The result output module 540 is used to integrate the corrected weight assessment results and reliability assessment results of each mode, and perform weighted fusion using a dynamic weight fusion method to obtain the final vehicle weight monitoring result.

[0108] In some alternative implementations, the time synchronization module 510 includes: The timing alignment unit is used to align the peak trigger time of the piezoelectric sensing monitoring data, the rising edge time of the strain gauge sensing monitoring data, and the timestamp of the camera image frame using the same time coordinate, so as to obtain the alignment result of each modal data on a unified time axis. The vehicle speed evaluation unit is used to obtain vehicle speed parameters based on the trigger time difference of piezoelectric sensor monitoring data set adjacent to each other along the vehicle's forward direction or the vehicle's displacement in continuous frame images from a camera, using a vehicle speed evaluation method. The time synchronization unit is used to map the duration of multiple modes to the same time window based on the vehicle speed parameters and using a time scale normalization method, so as to obtain time-synchronized multimodal weight monitoring data. The output is the time-synchronized vehicle speed parameters and the weight monitoring data of multiple modes.

[0109] In some alternative implementations, the error correction module 530 includes: The first correction unit is used to correct the initial weight of each mode by combining the vehicle speed parameters and using the speed correction coefficient to obtain the speed-corrected weight of the corresponding mode. The second correction unit is used to correct the weight after velocity correction for each mode by combining the temperature parameters and using the temperature drift coefficient to obtain the weight after temperature correction for the corresponding mode. The integrated correction unit is used to correct the weight based on the temperature correction for each mode, combined with the road surface condition parameters, using the road surface correction term to obtain the corrected weight assessment result for the corresponding mode. The reliability assessment unit is used to obtain the reliability assessment result of the corrected weight assessment result of the corresponding mode based on the corrected weight assessment result of each mode, combined with the signal-to-noise ratio of the corresponding mode, the deviation between the weight assessment results of other modes, and the historical drift amount.

[0110] In some optional implementations, the reliability assessment unit is specifically used for: Based on the corrected weight assessment results of each mode, and combined with the corrected weight assessment results of other modes, the inter-modal deviation of the corresponding mode is obtained by using the deviation evaluation method. Based on the intermodal deviation of each mode, combined with the signal-to-noise ratio of the corresponding mode and the historical drift, the original confidence level of the mode is obtained by using a weighted summation or nonlinear mapping method. Based on the original confidence level of each modality, a preset reliability level threshold is used for comparison to determine the reliability level or normalized confidence score of the modality, and the output is the reliability assessment result of the corrected weight assessment result of the corresponding modality.

[0111] In some alternative implementations, when the time-synchronized weight monitoring data includes piezoelectric sensing monitoring data, the initial evaluation module 520 is specifically used for: Based on the time-synchronized piezoelectric sensing monitoring data, the time-domain integral area and maximum peak value are extracted using the piezoelectric physical response model and output as piezoelectric response characteristic quantities. Based on the piezoelectric response characteristics, the initial weight corresponding to the piezoelectric branch is calculated using pre-calibrated piezoelectric physical response model coefficients.

[0112] In some alternative implementations, when the time-synchronized weight monitoring data includes strain gauge sensing monitoring data, the initial evaluation module 520 is specifically used for: Based on the strain gauge sensing and monitoring data after time synchronization, the strain response of the strain gauge is obtained by conversion using the first feature conversion method. Based on the strain gauge strain response, the time-domain integral area and maximum peak value are extracted using the strain physical response model and output as strain response characteristic quantities; Based on the strain response characteristics, the initial weight corresponding to the strain branch is calculated using pre-calibrated strain physical response model coefficients.

[0113] In some alternative implementations, when the time-synchronized weight monitoring data includes fiber optic sensing monitoring data, the initial evaluation module 520 is specifically used for: Based on the time-synchronized fiber optic sensing monitoring data, the second feature conversion method is used to convert the data, and the output is fiber optic mechanical strain. Based on optical fiber mechanical strain, the time-domain integral area and maximum peak value are extracted using the optical fiber physical response model and output as optical fiber response characteristic quantities. Based on the fiber response characteristics, the initial weight of the fiber branch is calculated using pre-calibrated fiber physical response model coefficients.

[0114] The vehicle weighing device provided in this embodiment of the invention can execute the vehicle weighing method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units described above are the same as in the corresponding embodiments described above, and will not be repeated here.

[0115] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0116] The following is a detailed reference. Figure 6 This diagram illustrates a suitable structural design for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 601, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 602 or a program loaded from memory 608 into random access memory (RAM) 603. RAM 603 also stores various programs and data required for the operation of the electronic device. The processor 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0117] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 6Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0118] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 609, or installed from a memory 608, or installed from a ROM 602. When the computer program is executed by the processor 601, it performs the functions defined in the vehicle weighing method of the embodiments of the present invention.

[0119] Figure 6 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0120] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the vehicle weighing method shown in the above embodiments is implemented.

[0121] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0122] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for weighing vehicles, characterized in that, The method includes: Based on the collected multimodal weight monitoring data, combined with the vehicle images and acquisition time captured by the camera, time synchronization is performed using a time synchronization method to obtain the vehicle speed parameters and weight monitoring data of multiple modes after time synchronization; the multimodal weight monitoring data includes piezoelectric sensing monitoring data, strain gauge sensing monitoring data and fiber optic sensing monitoring data. Based on the weight monitoring data after time synchronization of each mode, the data is processed using their respective pre-built physical response models to obtain the initial weight for each mode. Based on the initial weight of each mode, combined with vehicle speed parameters, temperature parameters and road condition parameters, the error compensation method is used to correct the weight and reliability of the corresponding mode. By combining the corrected weight assessment results and reliability assessment results for each mode, a weighted fusion method is used to obtain the final vehicle weight monitoring result.

2. The method according to claim 1, characterized in that, The initial weight for each mode is corrected using error compensation methods, taking into account vehicle speed parameters, temperature parameters, and road condition parameters, to obtain the corrected weight assessment result and reliability assessment result for the corresponding mode, including: Based on the initial weight of each mode, combined with the vehicle speed parameters, the speed correction coefficient is used to correct the weight of the corresponding mode after speed correction. Based on the velocity-corrected weight for each mode, combined with temperature parameters, the temperature drift coefficient is used for correction to obtain the temperature-corrected weight for the corresponding mode. Based on the temperature-corrected weight of each mode, combined with road surface condition parameters, the road surface correction term is used for correction to obtain the corrected weight assessment result for the corresponding mode. Based on the corrected weight assessment results for each mode, combined with the signal-to-noise ratio of the corresponding mode, the deviation between the weight assessment results of other modes, and the historical drift, the reliability assessment results of the corrected weight assessment results for the corresponding mode are obtained.

3. The method according to claim 2, characterized in that, The revised weight assessment result for each mode, combined with the signal-to-noise ratio of the corresponding mode, the deviation between the weight assessment results of other modes, and the historical drift, yields a reliability assessment result for the revised weight assessment result of the corresponding mode, including: Based on the corrected weight assessment results of each mode, and combined with the corrected weight assessment results of other modes, the inter-modal deviation of the corresponding mode is obtained by using the deviation evaluation method. Based on the intermodal deviation of each mode, combined with the signal-to-noise ratio of the corresponding mode and the historical drift, the original confidence level of the mode is obtained by using a weighted summation or nonlinear mapping method. Based on the original confidence level of each modality, a preset reliability level threshold is used for comparison to determine the reliability level or normalized confidence score of the modality, and the output is the reliability assessment result of the corrected weight assessment result of the corresponding modality.

4. The method according to claim 1, characterized in that, When the weight monitoring data after time synchronization includes piezoelectric sensing monitoring data, the weight monitoring data based on each mode after time synchronization is processed using its respective pre-built physical response model to obtain the initial weight corresponding to each mode, including: Based on the time-synchronized piezoelectric sensing monitoring data, the time-domain integral area and maximum peak value are extracted using the piezoelectric physical response model and output as piezoelectric response characteristic quantities. Based on the piezoelectric response characteristics, and combined with the pre-calibrated piezoelectric physical response model coefficients, the initial weight corresponding to the piezoelectric branch is calculated.

5. The method according to claim 1, characterized in that, When the weight monitoring data after time synchronization includes strain gauge sensing monitoring data, the weight monitoring data based on each modality after time synchronization is processed using its respective pre-built physical response model to obtain the initial weight corresponding to each mode, including: Based on the strain gauge sensing and monitoring data after time synchronization, the strain response of the strain gauge is obtained by conversion using the first feature conversion method. Based on the strain response of the strain gauge, the time-domain integral area and maximum peak value are extracted using the strain physics response model and output as strain response characteristic quantities. Based on the strain response characteristics, the initial weight corresponding to the strain branch is calculated using the pre-calibrated strain physical response model coefficients.

6. The method according to claim 1, characterized in that, When the weight monitoring data after time synchronization includes fiber optic sensing monitoring data, the weight monitoring data based on each mode after time synchronization is processed using its respective pre-built physical response model to obtain the initial weight corresponding to each mode, including: Based on the time-synchronized fiber optic sensing monitoring data, the second feature conversion method is used to convert the data, and the output is fiber optic mechanical strain. Based on the optical fiber mechanical strain, the time-domain integral area and maximum peak value are extracted using the optical fiber physical response model and output as optical fiber response characteristic quantities. Based on the fiber response characteristics, the initial weight of the fiber branch is calculated using the pre-calibrated fiber physical response model coefficients.

7. The method according to claim 1, characterized in that, The multimodal weight monitoring data collected is combined with vehicle images captured by a camera and the acquisition time. A time synchronization method is used to synchronize the data, resulting in synchronized vehicle speed parameters and multimodal weight monitoring data, including: Based on the peak trigger time of piezoelectric sensing monitoring data, the rising edge time of strain gauge sensing monitoring data, and the timestamp of camera image frames, the data are aligned using the same time coordinate to obtain the alignment result of each modal data on a unified time axis. Based on the trigger time difference of piezoelectric sensor monitoring data set adjacent to each other along the vehicle's direction of travel or the vehicle's displacement in consecutive frames of camera images, the vehicle speed parameters are obtained using a vehicle speed evaluation method. Based on the vehicle speed parameters, the duration of multiple modes is mapped to the same time window using a time scale normalization method to obtain time-synchronized multimodal weight monitoring data. The output is time-synchronized vehicle speed parameters and weight monitoring data of multiple modes.

8. A vehicle weighing device, characterized in that, The device includes: The time synchronization module is used to synchronize the collected multimodal weight monitoring data with the vehicle images and acquisition time captured by the camera using a time synchronization method, so as to obtain the time-synchronized vehicle speed parameters and weight monitoring data of multiple modes; the multimodal weight monitoring data includes piezoelectric sensing monitoring data, strain gauge sensing monitoring data and fiber optic sensing monitoring data. The initial evaluation module is used to process the weight monitoring data after time synchronization of each mode using their respective pre-built physical response models to obtain the initial weight for each mode. The error correction module is used to correct the initial weight of each mode by combining vehicle speed parameters, temperature parameters and road condition parameters, and using error compensation methods to obtain the corrected weight assessment result and reliability assessment result for the corresponding mode. The results output module integrates the corrected weight assessment results and reliability assessment results of each mode, and performs weighted fusion using a dynamic weight fusion method to obtain the final vehicle weight monitoring result.

9. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory storing computer instructions, and the processor executing the computer instructions to perform the vehicle weighing method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the vehicle weighing method according to any one of claims 1 to 7.

Images