A laser detection system based on traffic overloading control equipment and its detection method

Abstract

The present invention discloses a laser detection system based on traffic overload control equipment and its detection method. It includes a multi-wavelength laser detection module, a feedback adaptive laser power control module, a data fusion processing module, a vehicle load estimation module, and a laser vision assisted calibration module. Through multi-wavelength laser modulation and feedback control mechanism, the system realizes precise detection of vehicle load under different environmental conditions, significantly improving the signal separation degree in complex traffic flows. Combining with the adaptive weighted algorithm, the system can effectively isolate multi-lane interference, improving the accuracy of single-vehicle overload detection. At the same time, a low-cost vision sensor performs real-time image calibration on the vehicle bottom, eliminating laser reflection errors and ensuring the detection accuracy in the edge area. In addition, through multi-dimensional data fusion and dynamic load estimation, the present invention optimizes the overload detection speed and accuracy, reduces the detection delay caused by traffic congestion or errors, and ensures efficient and safe traffic management.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic monitoring and overloading detection, and particularly to a laser detection system based on traffic overloading control equipment and a detection method thereof. Background Art

[0002] Currently, traffic overloading detection systems are widely used in traffic overloading control points such as highways and urban roads to monitor whether the passing vehicles exceed the specified load limit. These systems usually use mechanical induction, pressure sensors, visual monitoring and other means to detect vehicle loads. Most traditional overloading detection methods rely on ground weighing technology, or calculate the load by combining laser ranging and vehicle speed monitoring. However, these methods have obvious deficiencies in complex environments such as multi-lane roads and highways.

[0003] First of all, in the prior art, mechanical induction and pressure sensors are often affected by multiple factors such as vehicle type, road surface condition and weather, resulting in low measurement accuracy. For example, in a multi-lane environment with high-density traffic flow, the signal interference from other vehicles significantly reduces the monitoring accuracy of single-vehicle load. In addition, although traditional laser detection technology has high accuracy in static situations, in dynamic and complex traffic flows and adverse weather conditions (such as rain, snow, wet and slippery), the attenuation problem of laser signals has not been effectively solved, which directly affects the stability and reliability of the system.

[0004] To address these problems, redundant devices (such as multiple cameras, pressure sensors) are often used in the industry to improve accuracy, but this increases the complexity and maintenance cost of the system. At the same time, due to the differences in vehicle speed, road surface condition and different vehicle shapes, the laser reflection signals at the bottom of the vehicle are often difficult to accurately capture, resulting in low detection accuracy in the edge area. In addition, in order to reduce signal interference, multiple sensors are usually combined, which can reduce interference, but it is still difficult to effectively isolate signals and improve accuracy in complex road conditions. Moreover, in response to complex and changing traffic environments, especially high-density traffic flows and complex terrains, traditional methods cannot provide sufficient real-time performance and accuracy. Summary of the Invention

[0005] The technical problem solved by the present invention is to provide a laser detection system based on traffic overloading control equipment and its detection method to solve the problems raised in the above background technology in view of the defects existing in the above-mentioned prior art. To solve the above technical problems, the technical solutions adopted by the present invention are as follows: A laser detection system based on traffic overloading control equipment includes: A multi-wavelength laser detection module for emitting multi-wavelength modulated laser beams to the bottom of the vehicle and the road surface area, wherein the multi-wavelength modulated laser beams include at least two laser beams with different wavelengths, and each wavelength is used to detect different vehicle structure characteristics and road surface reflection characteristics; The multi-wavelength laser detection module is configured to synchronously switch laser beams of different wavelengths through a preset modulation frequency and time sequence for differential detection of the bottom structure of the vehicle, the tire surface, the axle position, and the road surface; A reflected signal receiving module for receiving the return signals after the multi-wavelength laser beams are reflected by the bottom of the vehicle and the road surface; The reflected signal receiving module includes a multi-band photodetector, which can separate and mark reflected signals from different sources according to the intensity and delay time of reflected signals of different wavelengths received;

[0006] A feedback adaptive laser power control module, connected to the multi-wavelength laser detection module, for dynamically adjusting the emission power of the multi-wavelength laser beams according to the reflected intensities of different wavelengths received by the reflected signal receiving module. Among them, the dynamic adjustment of the laser power is comprehensively calculated based on the real-time ambient light intensity, vehicle type, and vehicle speed. The calculation formula is: , where is the laser emission power, is the real-time ambient light intensity, is the current driving speed of the vehicle, is the laser wavelength currently used, is the vehicle type, and the function is a non-linear function obtained by experimental fitting to ensure the stability of the reflected signal intensity under different environmental conditions;

[0007] A data fusion processing module for fusing the multi-wavelength reflected signal data received by the reflected signal receiving module with vehicle dynamic data to generate multi-dimensional signal features; wherein the vehicle dynamic data includes but is not limited to vehicle speed, axle position, and vehicle type. The data fusion processing module includes: A noise suppression unit for filtering random noise in the reflected signal based on a Kalman filter to suppress signal fluctuations caused by environmental interference and vehicle vibration; A feature extraction unit for extracting the axle load, tire characteristics, and vehicle dynamic load distribution of the vehicle according to the reflection characteristics of different components at the bottom of the vehicle; An adaptive weighted fusion unit for weighted fusion of the reflected data of each wavelength according to the reliability weights of different wavelength signals in the multi-wavelength reflected signals. Among them, the weight is calculated based on: , where is the signal intensity at wavelength , and is the sum of the signal intensities at all wavelengths. The influence degree of the reflected signals at different wavelengths on the final vehicle load detection result is determined through normalization processing;

[0008] The vehicle load dynamic estimation module is connected to the data fusion processing module and is used to estimate the real-time load state of the vehicle according to the multi-dimensional signal features after weighted fusion. Among them, the single-axle load of the vehicle The calculation formula is: , where is the reflection delay time of the laser beam, is the distance between the bottom of the vehicle and the road surface, is the weight of the reflected signals at each wavelength, and the function represents the axle load estimation model obtained through non-linear fitting; The laser vision-assisted calibration module is used to calibrate the edge conditions of the reflected signals during the laser detection process. The laser vision-assisted calibration module includes: a vision sensor unit for capturing real-time images of the bottom of the vehicle; an image-laser data alignment unit for performing edge correction on the laser detection results according to the geometric features of the bottom of the vehicle and the spatial positions of the reflected signals; a data consistency verification unit for comparing the calibrated data with the original laser reflection data for consistency. When the verification result exceeds the set error threshold, a feedback adjustment mechanism is triggered to re-adjust the laser emission power and signal processing parameters; a data storage and output module for storing all detection data, vehicle load states, and environmental data, and uploading the detection results to the traffic management system or the over-limit control center. Among them, the data storage and output module is configured to automatically synchronize and upload data at a predetermined time interval and send an alarm signal when an overloading situation is detected.

[0009] As a further solution of the present invention, the emission wavelength range of the multi-wavelength laser detection module includes the infrared band 850nm - 1550nm and the visible light band 400nm - 700nm, where different bands are used to detect the reflection characteristics of the metal and non-metal components at the bottom of the vehicle respectively.

[0010] As a further solution of the present invention, the feedback adaptive laser power control module further includes an ambient light intensity sensor and a vehicle speed sensor. The ambient light intensity sensor is used to measure the external light intensity, and the vehicle speed sensor is used to monitor the driving speed of the vehicle in real time, and dynamically adjust the laser emission power based on the data of both.

[0011] As a further solution of the present invention, the noise suppression unit in the data fusion processing module uses a combined algorithm of a Kalman filter and an adaptive mean filter, where the Kalman filter is used to dynamically eliminate high-frequency noise caused by vehicle vibration, and the adaptive mean filter is used to smooth low-frequency noise caused by environmental interference.

[0012] As a further solution of the present invention, the weight coefficient of the adaptive weighted fusion unit is also determined based on the stability score of each wavelength in the historical vehicle type data, where the stability score is calculated from past multiple detection results, and the formula is: , where is the wavelength at the th detection of the vehicle load value, is the deviation from the actual load value, is the number of historical detections.

[0013] As a further solution of the present invention, the non-linear fitting model for estimating the single-axle load of the vehicle in the vehicle load dynamic estimation module is obtained by training with machine learning methods, where the training data includes vehicle type, number of axles, tire contact area, vehicle speed, and road conditions, and the estimation formula is determined through a fitting model based on support vector regression.

[0014] As a further solution of the present invention, the image-laser data alignment unit of the laser vision assisted calibration module realizes spatial position alignment with the laser detection result by performing edge detection and shape feature matching on the bottom image of the vehicle, ensuring that the reflection signal deviation in the vehicle edge area is controlled within 1 millimeter.

[0015] As a further solution of the present invention, the predetermined time interval of the data storage and output module is dynamically adjusted according to the traffic flow. When the traffic flow is high, the predetermined time interval is automatically shortened to within 5 seconds. When the traffic flow is low, the time interval can be extended to 30 seconds.

[0016] As a further solution of the present invention, the system further includes a remote monitoring module communicating with the traffic management system. This remote monitoring module is used to real-time monitor the device status and detection results of all detection points, and automatically send a maintenance request when an abnormality occurs in the detection device.

[0017] A detection method for a laser detection system based on traffic overloading control equipment, characterized by comprising the following steps: S1, emitting a multi-wavelength modulated laser beam to the bottom of a traveling vehicle and the road surface through the multi-wavelength laser detection module, and capturing reflection signals at different wavelengths; S2, receiving the reflection signals through the reflection signal receiving module and preliminarily separating the reflection data at different wavelengths; S3, dynamically adjusting the laser emission power according to the real-time ambient light intensity and vehicle speed through the feedback adaptive laser power control module; S4, performing noise suppression, feature extraction, and weighted fusion on the received reflection signals through the data fusion processing module to obtain multi-dimensional signal features; S5, calculating the single-axis load of the vehicle according to the weighted fusion signal features and vehicle dynamic data through the vehicle load dynamic estimation module; S6, calibrating the laser detection results in the edge area through the laser vision assisted calibration module; S7, storing the estimated vehicle load data in the data storage and output module, and judging whether to trigger an overloading alarm signal according to the detection result.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Through multi-band data fusion processing, laser signals of different wavelengths are combined with multi-dimensional data such as the speed of the vehicle and the position of the axles, and an adaptive weighting algorithm is used for dynamic adjustment, effectively isolating the signal interference from vehicles in multiple lanes, improving the signal separation degree, and further enhancing the accuracy of single-vehicle overload detection. In the multi-lane parallel environment, the interference separation degree of the system is significantly improved, ensuring that the load detection of each vehicle is not interfered by other vehicles, and significantly enhancing the detection accuracy and system reliability in the complex situation of multi-lane parallel. And it solves the error problem that may occur in the traditional laser detection at the bottom edge area of the vehicle. By using a low-cost vision sensor to collect real-time images of the vehicle bottom and precisely aligning and calibrating them with the spatial position of the laser signal, the accuracy loss caused by uneven laser reflection is eliminated, which is conducive to ensuring that the detection results remain highly accurate in complex traffic flows or edge detection areas, enabling the system to provide stable overload detection results under any conditions. Especially in the environment of complex-shaped vehicles (such as trailers, trucks, etc.) and high-density traffic flows, the system can work accurately on vehicles with different heights and structures. 2. The dynamic load estimation module combines laser signals with vehicle dynamic data (such as speed, axle position, etc.) and uses an adaptive algorithm to perform dynamic fusion processing on multi-dimensional signals, effectively improving the accuracy and efficiency of real-time load estimation. Especially when high-density traffic flows or multi-axle heavy vehicles pass by, the system can estimate the load status of the vehicle in real time and accurately, so as to make an overload detection judgment more quickly. This technology can significantly improve the speed and accuracy of overload detection with almost no additional hardware cost, ensure the efficient processing of traffic flows without affecting the completion of the overload monitoring task, and further reduce traffic congestion caused by detection errors or delays in traditional methods. While ensuring the overload detection accuracy, the multi-dimensional signal fusion and feedback control system also fully considers the efficiency of the system. By adaptively adjusting the laser power and automatically optimizing the detection mode, the system can maintain a fast response during peak traffic hours, ensuring high-precision detection without adding unnecessary traffic congestion and delays. Different from traditional overload control devices, this solution can adjust the detection strategy according to the traffic flow and road conditions, avoiding redundant detection and over-reliance on equipment, greatly improving the working efficiency and traffic capacity during the overload control process, enabling vehicles to complete overload detection without stopping, thus improving the smoothness and management efficiency of the entire traffic flow. 3. Through the multi-wavelength modulated laser detection system, combined with feedback adaptive laser power control, the problem of detection accuracy caused by the inability of traditional single-wavelength laser detection to adapt to complex environments (such as different weather, lighting, slippery roads, etc.) is successfully solved. The system monitors the intensity of the reflected signal in real time and automatically adjusts the laser emission power according to environmental changes, reducing signal attenuation caused by environmental factors such as weather and lighting.This innovative integration of multi-wavelength laser signals and real-time environmental adjustment mechanisms has achieved an improvement in overload detection accuracy in complex environments, reducing the detection error of the system to within 3% in harsh environments such as rain, snow, and slippery conditions, far lower than the 10% error of existing technologies. This combined technology enhances environmental adaptability and signal stability while improving the overall system's accuracy and robustness, solving the problem of insufficient detection accuracy in dynamic and complex traffic environments.

[0019] Additional aspects and advantages of the present invention will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 It is the system architecture diagram of the present invention.

[0022] Figure 2 It is the working flow chart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] The following will clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0024] Please refer to Figure 1 and Figure 2, in the embodiment of the present invention, a laser detection system based on traffic overloading control equipment includes: a multi-wavelength laser detection module configured to emit a multi-wavelength modulated laser beam towards the bottom of the vehicle and the road surface area, wherein the multi-wavelength modulated laser beam includes at least two laser beams with different wavelengths, and each wavelength is used to detect different vehicle structure characteristics and road surface reflection characteristics; the multi-wavelength laser detection module is configured to synchronously switch laser beams with different wavelengths through a preset modulation frequency and time sequence for differential detection of the vehicle bottom structure, tire surface, axle position, and road surface; a reflected signal receiving module configured to receive the return signal after the multi-wavelength laser beam is reflected from the vehicle bottom and the road surface; the reflected signal receiving module includes a multi-band photodetector capable of separating and marking reflected signals from different sources according to the intensity and delay time of the received reflected signals with different wavelengths; a feedback adaptive laser power control module connected to the multi-wavelength laser detection module, configured to dynamically adjust the emission power of the multi-wavelength laser beam according to the reflected intensities of different wavelengths of laser received by the reflected signal receiving module, wherein the dynamic adjustment of the laser power is comprehensively calculated based on the real-time ambient light intensity, vehicle type, and vehicle speed, and the calculation formula is: , where is the laser emission power, is the real-time ambient light intensity, is the current driving speed of the vehicle, is the laser wavelength currently used, is the vehicle type, and the function is a non-linear function obtained by experimental fitting, used to ensure the stability of the reflected signal intensity under different environmental conditions; a data fusion processing module configured to fuse the multi-wavelength reflected signal data received by the reflected signal receiving module with vehicle dynamic data to generate multi-dimensional signal features; wherein the vehicle dynamic data includes but is not limited to vehicle speed, axle position, and vehicle type, and the data fusion processing module includes: a noise suppression unit configured to filter random noise in the reflected signal based on a Kalman filter to suppress signal fluctuations caused by environmental interference and vehicle vibration; a feature extraction unit configured to extract the axle load, tire characteristics, and vehicle dynamic load distribution of the vehicle according to the reflection characteristics of different components at the vehicle bottom; an adaptive weighted fusion unit configured to perform weighted fusion on the reflected data of each wavelength according to the reliability weights of different wavelength signals in the multi-wavelength reflected signal, wherein the weight is calculated based on: , where is the wavelength is the signal intensity at is the sum of signal intensities at all wavelengths, and determines the influence degree of reflected signals at different wavelengths on the final vehicle load detection result through normalization processing; the vehicle load dynamic estimation module, connected to the data fusion processing module, is used to estimate the real-time load state of the vehicle according to the multi-dimensional signal features after weighted fusion, where the single-axis load of the vehicle The calculation formula is: , where is the reflection delay time of the laser beam, is the distance between the bottom of the vehicle and the road surface, is the weight of the reflected signals at each wavelength, and the function represents the axle load estimation model obtained through non-linear fitting;

[0025] The laser vision-assisted calibration module is used to calibrate the edge conditions of the reflected signals during the laser detection process. The laser vision-assisted calibration module includes: a vision sensor unit for capturing real-time images of the bottom of the vehicle; an image-laser data alignment unit for performing edge correction on the laser detection results according to the geometric features of the bottom of the vehicle and the spatial position of the reflected signals; a data consistency verification unit for comparing the calibrated data with the original laser reflection data for consistency. When the verification result exceeds the set error threshold, a feedback adjustment mechanism is triggered to readjust the laser emission power and signal processing parameters; a data storage and output module for storing all detection data, vehicle load status, and environmental data, and uploading the detection results to the traffic management system or the over-limit control center. Among them, the data storage and output module is configured to automatically synchronize and upload data at a predetermined time interval, and send an alarm signal when an overloading situation is detected.

[0026] As a further solution of the present invention, the emission wavelength range of the multi-wavelength laser detection module includes the infrared band 850nm - 1550nm and the visible light band 400nm - 700nm, where different bands are used to detect the reflection characteristics of metal and non-metal components at the bottom of the vehicle respectively.

[0027] As a further solution of the present invention, the feedback adaptive laser power control module further includes an ambient light intensity sensor and a vehicle speed sensor. The ambient light intensity sensor is used to measure the external light intensity, and the vehicle speed sensor is used to monitor the driving speed of the vehicle in real time, and dynamically adjust the laser emission power based on the data of both.

[0028] As a further solution of the present invention, the noise suppression unit in the data fusion processing module uses a combined algorithm of a Kalman filter and an adaptive mean filter, where the Kalman filter is used to dynamically eliminate high-frequency noise caused by vehicle vibration, and the adaptive mean filter is used to smooth low-frequency noise caused by environmental interference.

[0029] As a further solution of the present invention, the weight coefficient of the adaptive weighted fusion unit is also determined based on the stability score of each wavelength in the historical vehicle type data, where the stability score is calculated from the past multiple detection results, and the formula is: , where is the wavelength at the th detection of the vehicle load value is the deviation from the actual load value is the number of historical detections

[0030] As a further solution of the present invention, the non-linear fitting model used in the vehicle load dynamic estimation module to estimate the single-axle load of the vehicle is obtained by training with machine learning methods, where the training data includes vehicle type, number of axles, tire contact area, vehicle speed, and road conditions, and the estimation formula is determined through a fitting model based on support vector regression

[0031] As a further solution of the present invention, the image laser data alignment unit of the laser vision assisted calibration module realizes the spatial position alignment with the laser detection result by performing edge detection and shape feature matching on the bottom image of the vehicle, ensuring that the reflection signal deviation in the vehicle edge area is controlled within 1 millimeter

[0032] As a further solution of the present invention, the predetermined time interval of the data storage and output module is dynamically adjusted according to the traffic flow. When the traffic flow is high, the predetermined time interval is automatically shortened to within 5 seconds. When the traffic flow is low, the time interval can be extended to 30 seconds

[0033] As a further solution of the present invention, the system further includes a remote monitoring module communicating with the traffic management system, which is used to monitor the device status and detection results of all detection points in real time, and automatically send a maintenance request when an abnormality occurs in the detection device

[0034] A detection method for a laser detection system based on traffic overload control equipment, characterized by including the following steps: S1, emitting a multi-wavelength modulated laser beam to the bottom of the driving vehicle and the road surface through the multi-wavelength laser detection module, and capturing reflection signals at different wavelengths; S2, receiving the reflection signals through the reflection signal receiving module and preliminarily separating the reflection data of different wavelengths; S3, dynamically adjusting the laser emission power according to the real-time ambient light intensity and vehicle speed through the feedback adaptive laser power control module; S4, performing noise suppression, feature extraction, and weighted fusion on the received reflection signals through the data fusion processing module to obtain multi-dimensional signal features; S5, calculating the single-axis load of the vehicle according to the weighted fusion signal features and vehicle dynamic data through the vehicle load dynamic estimation module; S6, calibrating the laser detection results in the edge area through the laser vision assisted calibration module; S7, storing the estimated vehicle load data in the data storage and output module, and judging whether to trigger an overload alarm signal according to the detection result.

[0035] Embodiment 1: This embodiment details the implementation path of an overload detection method based on multi-wavelength laser detection and vision assisted calibration, solving problems such as low accuracy and large signal interference of traditional laser detection methods in complex traffic environments. Through multi-dimensional data fusion and adaptive adjustment technologies, this embodiment optimizes the accuracy and speed of overload detection, while avoiding errors caused by uneven laser signal reflection.

[0036] The multi-wavelength laser detection module is mainly used to emit multi-wavelength modulated laser beams to the bottom area of the vehicle and the road surface, and through synchronously switching laser beams of different wavelengths, perform differential detection for different vehicle components and road surface characteristics. The implementation steps are as follows: the laser module switches the laser wavelength according to a preset frequency, and the wavelength range covers the infrared band from 850nm to 1550nm and the visible light band from 400nm to 700nm. The reflection intensity and delay time of the laser signal are used to detect the metal components and non-metal components at the bottom of the vehicle respectively, ensuring that signals of different wavelengths are optimized respectively under different reflection characteristics.

[0037] The optimization of the adaptive laser power control module adjusts the laser emission power according to dynamic factors such as ambient light intensity and vehicle speed to ensure the stability and effectiveness of the signal in different environments. Its implementation steps: the dynamic adjustment of the laser power adopts the following formula: , where is the laser emission power, is the real-time ambient light intensity, is the current driving speed of the vehicle, is the currently used laser wavelength, is the vehicle type. In this formula, It is a non - linear function obtained by fitting according to actual experiments to ensure the stability of the intensity of the laser reflection signal under various environmental conditions.

[0038] The multi - dimensional data fusion and noise suppression module is responsible for integrating the signals from the laser detection module and vehicle dynamic data, and uses a combination of Kalman filter and adaptive mean filter for noise suppression. Its implementation steps are to remove the high - frequency noise caused by vehicle vibration through the Kalman filter, and smooth the low - frequency noise caused by environmental interference using the adaptive mean filter. Based on the intensity and delay time of the multi - wavelength reflection signals, each wavelength signal is weighted to ensure that the contributions of signals of different wavelengths to the load detection result are weighted according to their reliability. The weight calculation formula is: , where, is the wavelength is the signal intensity at the wavelength. The vehicle load dynamic estimation module estimates the single - axle load of the vehicle in real - time based on the weighted - fused signal features and vehicle dynamic data. The implementation steps are to calculate the vehicle single - axle load using the following formula: where, is the vehicle single - axle load, is the laser reflection delay time, is the distance between the bottom of the vehicle and the road surface, is the wavelength is the signal weight at the wavelength. The function in this formula represents the axle load estimation model obtained by non - linear fitting.

[0039] The laser vision - assisted calibration module is used to calibrate the edge region of the laser detection signal to eliminate the error caused by uneven laser signal reflection. Its implementation steps are to use a vision sensor to capture the real - time image of the bottom of the vehicle, combine the image data with the laser reflection signal, and achieve precise calibration of the edge region through the image - laser data alignment unit. The calibrated data will be compared with the original laser reflection signal for consistency. When the error exceeds the set threshold, the system will trigger an adjustment mechanism to dynamically adjust the laser power and signal processing parameters.

[0040] The data storage and output module is responsible for storing all detected data and uploading the results to the traffic management system or the overloading control center. The implementation steps are as follows: when the traffic flow is high, the data synchronization upload time interval is automatically shortened to within 5 seconds to ensure timely response; when the traffic flow is low, the upload time interval can be extended to 30 seconds to save bandwidth and resources. This embodiment realizes high-precision overloading detection through multi-wavelength laser detection and visual calibration technology. Especially in complex traffic environments and adverse weather conditions, the system shows strong stability and accuracy. Through adaptive laser power control and multi-dimensional data fusion, the system can effectively isolate signal interference in multiple lanes and ensure accurate detection of the load of a single vehicle under high-density traffic flow conditions.

[0041] Embodiment 2: To solve the problems of signal interference, error processing, and dynamic adjustment in the existing laser detection system in complex traffic environments, the system architecture of this embodiment includes a multi-wavelength laser detection module, a reflected signal receiving module, a feedback adaptive laser power control module, a data fusion processing module, a vehicle load dynamic estimation module, and a laser vision assisted calibration module. The hardware configuration, function description, and data flow path of each module are as follows. Multi-wavelength laser detection module: This module emits multi-wavelength laser beams, and the wavelength range includes the infrared band of 850nm - 1550nm and the visible light band of 400nm - 700nm, which are used to detect the metal and non-metal structures at the bottom of the vehicle respectively. The laser beams of each wavelength are switched synchronously, and through modulation frequency and time series adjustment, signal capture is optimized under different detection targets.

[0042] Reflected signal receiving module: This module includes multi-band photodetectors to receive the reflected laser signals. By analyzing the signal intensity and delay time, different reflection sources are distinguished and the signal sources are marked to ensure that signals in different lanes do not get confused. Feedback adaptive laser power control module: This module dynamically adjusts the laser emission power through real-time monitoring of ambient light intensity, vehicle speed, and laser wavelength. Specifically, the adjustment process is supported by the following formula: , where is the laser emission power, is the ambient light intensity, is the vehicle speed, is the laser wavelength used. The function is a non-linear function fitted based on experimental data to ensure the stability of the laser reflection signal under different lighting and weather conditions. In the data fusion processing module, first, high-frequency noise is removed through a Kalman filter to eliminate the fluctuations caused by vehicle vibration, and then low-frequency noise caused by environmental interference (such as wind, rain, etc.) is smoothed through an adaptive mean filter. The fused multi-wavelength signal data is weighted according to reliability, and the calculation basis of the weight coefficient is: , where is the signal intensity at the wavelength , and is the sum of the signal intensities of all wavelengths. Through this weight coefficient, it is ensured that the contributions of different wavelengths to the final detection result are dynamically adjusted according to their reliability. By processing the weighted and fused signal data, the single-axle load of the vehicle is estimated using the following formula: , where is the single-axle load of the vehicle, is the delay time of the laser reflection, is the distance between the bottom of the vehicle and the road surface, is the signal weight of the corresponding wavelength. The function represents the load estimation model obtained based on non-linear fitting, and this model is dynamically adjusted according to the bottom structure of the vehicle, the tire contact area, and the vehicle speed. To eliminate the errors in the laser detection process, especially in the edge area, a laser vision-assisted calibration module is added. This module captures the real-time image of the bottom of the vehicle through the vision sensor unit, and aligns the image with the laser data. Through the image-laser data alignment unit, precise calibration is performed according to the geometric features of the bottom structure and the spatial position of the reflection signal. When the error of the calibration result exceeds the predetermined threshold, the system automatically triggers a feedback mechanism to readjust the laser emission power and signal processing parameters to ensure the calibration accuracy. And to improve the response ability of the system under high-density traffic flow, the data storage and output module adjusts the upload frequency according to the traffic flow. When the traffic flow is high, the upload interval is shortened to 5 seconds, and when the traffic flow is low, the upload interval can be extended to 30 seconds. In addition, the system is equipped with a remote monitoring module, which can monitor the device status in real time and automatically send a maintenance request when a failure occurs.

[0043] In this embodiment, a multi-wavelength laser detection module and a vision sensor are configured to ensure that it can emit laser light within a predetermined wavelength range and receive the reflection signal. A feedback adaptive laser power control module is installed to ensure that it can dynamically adjust the laser power according to the ambient light intensity, vehicle speed, etc. A data fusion processing module is configured and a noise suppression and signal weighted fusion algorithm is implemented to ensure the accuracy of the fused data. A vehicle load estimation module is implemented and the load condition is estimated according to the real-time state of the vehicle through a dynamic algorithm. The visual-laser alignment algorithm is used for precise calibration of the reflection signal to minimize the error in the edge area. A data storage and output module is configured to ensure that the system can adaptively adjust the data upload frequency under high traffic flow conditions, and the device status is monitored in real time through the remote monitoring module. The implementation ensures that in a complex and changing traffic environment, the system can stably and efficiently perform overloading detection, significantly improving the detection accuracy and system stability, and at the same time having high operability and maintainability during the implementation process.

[0044] Example 3: To further simplify the process of load estimation, especially without relying on complex calculation models, this example proposes a simple and effective load estimation method based on vehicle dynamic characteristics and laser detection signals. The goal of this example is to improve the simplicity and efficiency of system implementation by reducing the computational burden, ensuring that the vehicle load can be quickly and accurately obtained in practical applications.

[0045] In the system structure and core modules, the laser detection module: emits multi-wavelength laser beams to scan the bottom of the vehicle and receives the reflected signals. The wavelength and reflection intensity of the laser beams are used to infer relevant data of the vehicle load. The data acquisition module: collects data in combination with the vehicle speed and vehicle type information obtained in real time. The role of this module is to provide basic data for subsequent estimation. The load estimation module: calculates the vehicle load through a simplified estimation model based on the dynamic characteristic data such as the vehicle speed, the number of axles, and the vehicle type, combined with the intensity of the laser reflection signal.

[0046] Load estimation method: This example uses a simplified model to avoid using complex non-linear fitting or a large amount of historical data for calculation. Through a real-time data processing module with fast response, using the known vehicle speed and vehicle characteristics (such as the number of axles, vehicle type), combined with the laser reflection intensity, a linear regression model is used to estimate the load. The simplified load estimation formula is as follows: , where: is the single-axle load of the vehicle, is the current speed of the vehicle, is the number of axles of the vehicle, is the intensity of the reflection signal, is the weight coefficient adjusted through experiments or actual data. In this example, the weight coefficient is obtained through on-site experiments or data fitting. These coefficients play a crucial role in the load estimation process and determine the contribution degree of different parameters to the load estimation result. The specific acquisition method is as follows: In data collection, such as experimental environment setting: First, conduct a series of experiments in different traffic scenarios, such as different vehicle speeds and different vehicle types. When the vehicle is driving, real-time dynamic data of the vehicle and its corresponding actual load (for example, the actual load measured by a road surface pressure sensor) are collected through the laser detection module, vehicle speed sensor, axle number identification, etc.

[0047] Range of data collection: The experimental data should cover the reflection intensities under different types of vehicles (such as small cars, large trucks, etc.), different vehicle speeds (such as low speed, medium speed, high speed), and different road conditions. These data will be used to fit the weight coefficients and ensure that the fitting model has a wide range of adaptability. Data processing and fitting, Preliminary data processing: According to the collected data, remove outliers and standardize the data. For each set of experimental data, the vehicle speed , the number of axles and the reflection intensity are extracted as input features, and the load value is used as the target output. Selection of fitting method: Use a regression analysis method (such as the least squares method) for fitting to calculate the contribution ratio of each feature to the load, thereby obtaining the corresponding weight coefficients , where represents the influence of vehicle speed on the load. Generally, the higher the vehicle speed, the more linear the relationship between the load and the speed; represents the influence of the number of axles on the load. Vehicles with more axles usually have a heavier load, and the weight coefficient reflects this influence; represents the influence of the reflection intensity on the load. The stronger the reflection intensity, the higher the load on the bottom structure of the vehicle, and the coefficient reflects this linear relationship. Then, fit the collected data through the least squares method model. For example, when the vehicle speed is 70 km / h, the number of axles is 3, and the reflection intensity is 1.6, the calculated load is: . By comparing the actual load in the experiment with the estimated load, verify the accuracy of the model. If a large deviation is found, the model can be further optimized by adjusting the parameters in the fitting process or adding more experimental data to improve the weight coefficients. If it is highly sensitive to certain features (such as vehicle speed), the weight coefficients can be appropriately adjusted to make the model more accurate and efficient. The test data is as follows:

[0048]

[0049] And the vehicle speed : The driving speed of the vehicle is obtained in real time through a vehicle speed sensor. The number of axles : Based on the vehicle type, a predefined number of axles value is set in the system. For most vehicle types, this parameter can be directly obtained during license plate recognition. The reflection signal intensity : It is directly obtained through the intensity of the laser reflection signal, and this intensity data reflects the relevant information of the vehicle bottom structure and load. Its implementation steps, Step 1: Collect the reflection signal at the bottom of the vehicle through the laser detection module and obtain the vehicle speed data in real time. Step 2: Automatically extract the number of axles according to the vehicle type or manually input vehicle information. Step 3: According to the formula , calculate the uniaxial load. Step 4: The system determines whether the vehicle is overloaded according to the calculation result and outputs the result.

[0050] Embodiment 4: In this embodiment, first, a multi-wavelength laser detection module emits multiple laser beams with different wavelengths to the bottom of the vehicle and the road surface area. Each wavelength of the laser beam is used to detect different components of the vehicle. For example, the long-wavelength laser beam is used to detect the reflection characteristics of the metal components of the vehicle, while the short-wavelength laser beam detects the non-metal components. The reflection signal receiving module separates the reflection signals with different wavelengths to ensure that the reflection signals from different components can be accurately recorded. During this process, the intensity and delay time of the reflection signals are the main basis for signal processing, which are used to judge the load characteristics of the vehicle.

[0051] To improve the accuracy of signal processing, this embodiment adopts an algorithm that combines Kalman filtering and adaptive mean filtering to suppress the noise of the signal. The Kalman filter is used to process the high-frequency noise caused by vehicle vibration, while the adaptive mean filter mainly smooths the low-frequency noise caused by environmental interference (such as wind, rain, etc.). By combining these two algorithms, the interference caused by external environmental changes and vehicle dynamic factors can be effectively reduced, ensuring that the signal data transmitted to the data fusion processing module is more accurate.

[0052] Subsequently, the data fusion processing module combines the multi-wavelength reflection signals with the vehicle dynamic data (such as vehicle speed, axle position, etc.) through a weighted fusion method. The calculation of the weighting coefficient is based on the reliability and intensity of the signal. Specifically, by normalizing the signal intensity, the influence degree of each wavelength signal on the final detection result is obtained. After weighted fusion, the reliability of the signal is optimized to ensure that the estimation result of the vehicle load is more accurate.

[0053] In view of the problem of laser signal attenuation caused by environmental changes, this embodiment proposes a dynamic power control mechanism based on real-time ambient light intensity, vehicle driving speed, and laser wavelength. The system adjusts the laser emission power in real time according to the received reflection signal intensity and delay time. First, the external light intensity is measured in real time by an ambient light intensity sensor, and an optimal laser emission power is calculated by combining the real-time speed information of the vehicle. Specifically, the relationship among the laser power, ambient light intensity, vehicle speed, and wavelength is dynamically adjusted through a non-linear function fitted by experiments, so as to ensure the stability of the reflection signal in different environments. In this way, the system can avoid detection errors caused by light changes or bad weather, thereby improving the overall overloading detection accuracy. And in order to further improve the detection accuracy in the edge area, this embodiment combines a laser vision-assisted calibration technology. The vision sensor is used to capture the image data of the vehicle bottom in real time, and the reflection result of the laser signal is spatially aligned with the image data. Through the image laser data alignment unit, based on the geometric features of the vehicle bottom and the positional relationship of the reflection signal, the laser detection result is accurately calibrated. If the calibration error exceeds the set threshold, the system will automatically trigger a feedback adjustment mechanism to readjust the laser emission power and signal processing parameters. Through this calibration process, the system can eliminate the laser reflection error caused by the vehicle shape or road unevenness, ensuring high-precision load detection. Finally, based on the signal features and vehicle dynamic data after weighted fusion, the vehicle load dynamic estimation module calculates the single-axis load of the vehicle in real time. Through the cooperation of the real-time laser signal delay time, the distance between the vehicle bottom and the road surface, and the signal weight, the system can accurately estimate the load condition of the vehicle. For the influence degree of each wavelength signal on the load estimation, the weight coefficient is optimized through historical detection data, so as to improve the accuracy of the load estimation. During this process, the system combines dynamic data for overloading detection. When the vehicle load exceeds the set standard value, an overloading alarm signal is automatically triggered, and the detection result is uploaded to the traffic management system or the overloading control center for further processing by relevant personnel.

[0054] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above-described exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to embrace all changes within the meaning and scope of the equivalent elements of the claims in the present invention.

Claims

1. A laser detection system based on traffic overload control equipment, characterized in that: include: A multi-wavelength laser detection module is used to emit a multi-wavelength modulated laser beam to the bottom of the vehicle and the road surface area, wherein the multi-wavelength modulated laser beam is used to detect different vehicle structural features and road surface reflection characteristics; a reflection signal receiving module is used to receive the return signal of the multi-wavelength laser beam after being reflected from the bottom of the vehicle and the road surface; a feedback adaptive laser power control module is connected to the multi-wavelength laser detection module and is used to dynamically adjust the emission power of the multi-wavelength laser beam according to the different wavelength laser reflection intensities received by the reflection signal receiving module, wherein the dynamic adjustment of the laser power is comprehensively calculated based on the real-time ambient light intensity, vehicle type and vehicle speed, and the calculation formula is: ,in, is the laser emission power, is the real-time ambient light intensity, is the current speed of the vehicle, is the laser wavelength currently used, is the vehicle type, function The nonlinear function obtained by experimental fitting is used to ensure the stability of the reflected signal intensity under different environmental conditions; The data fusion processing module is used to fuse the multi-wavelength reflection signal data received by the reflection signal receiving module with the vehicle dynamic data to generate multi-dimensional signal features; the data fusion processing module includes: a noise suppression unit, which is used to filter out random noise in the reflection signal based on the Kalman filter, and suppress signal fluctuations caused by environmental interference and vehicle vibration; a feature extraction unit, which is used to extract the vehicle's axle load, tire characteristics and vehicle dynamic load distribution according to the reflection characteristics of different components at the bottom of the vehicle; an adaptive weighted fusion unit, which is used to perform weighted fusion on the reflection data of each wavelength according to the reliability weights of different wavelength signals in the multi-wavelength reflection signal, wherein the weight The calculation basis is: ,in, The wavelength The signal strength under is the sum of the signal intensities at all wavelengths, and the influence of different wavelength reflection signals on the final vehicle load detection results is determined through normalization processing; The vehicle load dynamic estimation module is connected to the data fusion processing module and is used to estimate the real-time load state of the vehicle based on the multi-dimensional signal characteristics after weighted fusion, wherein the single-axle load of the vehicle The calculation formula is: ,in, is the reflection delay time of the laser beam, is the distance between the bottom of the vehicle and the road surface, is the weight of the reflected signal at each wavelength, function It represents the axle load estimation model obtained by nonlinear fitting; the laser vision-assisted calibration module is used to calibrate the edge conditions of the reflected signal during the laser detection process; the data storage and output module is used to store all detection data, vehicle load status and environmental data, and upload the detection results to the traffic management system or the overload control center, wherein the data storage and output module is configured to automatically and synchronously upload data at predetermined time intervals, and send an alarm signal when an overload condition is detected.

2. A laser detection system based on traffic overload control equipment according to claim 1, characterized in that: The emission wavelength range of the multi-wavelength laser detection module includes the infrared band 850nm-1550nm and the visible light band 400nm-700nm.

3. The laser detection system based on traffic overload control equipment according to claim 1 is characterized in that: The feedback adaptive laser power control module also includes an ambient light intensity sensor and a vehicle speed sensor. The ambient light intensity sensor is used to measure the external light intensity, and the vehicle speed sensor is used to monitor the vehicle's running speed in real time.

4. The laser detection system based on traffic overload control equipment according to claim 1 is characterized in that: The noise suppression unit in the data fusion processing module uses a combination algorithm of a Kalman filter and an adaptive mean filter, where the Kalman filter is used to dynamically eliminate high-frequency noise caused by vehicle vibration, and the adaptive mean filter is used to smooth low-frequency noise caused by environmental interference.

5. The laser detection system based on traffic overload control equipment according to claim 1 is characterized in that: Weight coefficient of adaptive weighted fusion unit It is also determined based on the stability score of each wavelength in the historical vehicle type data, wherein the stability score is calculated through multiple past test results, and the formula is: ,in, The wavelength Next The vehicle load value detected for the first time, is the deviation from the actual load value, The number of historical detections.

6. The laser detection system based on traffic overload control equipment according to claim 1 is characterized in that: The nonlinear fitting model used to estimate the single-axle load of the vehicle in the vehicle load dynamic estimation module is obtained through machine learning training. The training data includes vehicle type, number of axles, tire contact area, vehicle speed and road conditions, and the estimation formula is determined by a fitting model based on support vector regression.

7. The laser detection system based on traffic overload control equipment according to claim 1 is characterized in that: The image laser data alignment unit of the laser vision-assisted calibration module performs edge detection and shape feature matching on the vehicle bottom image to achieve spatial position alignment with the laser detection result, ensuring that the deviation of the reflected signal in the edge area of ​​the vehicle is controlled within 1 mm.

8. The laser detection system based on traffic overload control equipment according to claim 1 is characterized in that: The scheduled time interval of the data storage and output module is dynamically adjusted according to the traffic volume. When the traffic volume is high, the scheduled time interval is automatically shortened to less than 5 seconds. When the traffic volume is low, the time interval can be extended to 30 seconds.

9. The laser detection system based on traffic overload control equipment according to claim 1 is characterized in that: The system also includes a remote monitoring module that communicates with the traffic management system. The remote monitoring module is used to monitor the equipment status and detection results of all detection points in real time, and automatically issue maintenance requests when the detection equipment is abnormal.

10. The laser detection system based on traffic overload control equipment according to claim 1 is characterized in that: The detection system includes the following detection method steps: S1, emitting a multi-wavelength modulated laser beam to the bottom and road surface of a moving vehicle through the multi-wavelength laser detection module to capture reflection signals at different wavelengths; S2, receiving the reflection signal through the reflection signal receiving module, and preliminarily separating the reflection data of different wavelengths; S3, dynamically adjusting the laser emission power according to the real-time ambient light intensity and vehicle speed through the feedback adaptive laser power control module; S4, performing noise suppression, feature extraction and weighted fusion on the received reflection signal through the data fusion processing module to obtain multi-dimensional signal features; S5, calculating the single-axle load of the vehicle according to the weighted fused signal features and vehicle dynamic data through the vehicle load dynamic estimation module; S6, calibrating the laser detection results of the edge area through the laser vision-assisted calibration module; S7, storing the estimated vehicle load data in the data storage and output module, and judging whether to trigger an overload alarm signal based on the detection results.

Images