Monitoring and early warning system for monitoring expressway surface defects by adopting radar

Abstract

The invention discloses a monitoring and early warning system for monitoring expressway surface defects by adopting radar. The monitoring and early warning system comprises a signal acquisition module, a signal processing module and a signal processing module, wherein the signal acquisition module is mounted on an expressway cross bar and used for acquiring real-time reflected wave signals of vehicles running on an expressway surface; the reflected wave signal extraction module is used for obtaining real-time wave change characteristics when a vehicle runs non-stably, and the highway surface defect map characteristic library collects the wave change characteristics to form the highway surface defect map characteristic library; the wave change feature comparison module is used for comparing real-time wave deformation features with the expressway surface defect map feature library, and the early warning device sends an alarm to a vehicle driving to the area; according to the invention, the abnormal waveform of the vehicle driving to the road surface is obtained by using the radar, the road surface abnormal condition can be rapidly obtained through the vehicle body attitude, and the vehicle is early warned in time; road surface monitoring in a large area can be achieved by monitoring vehicle shaking, and the system has the advantages of being low in cost and suitable for popularization.

Description

Technical Field

[0001] The invention relates to a monitoring and early warning system for monitoring high-speed road surface defects using radar, and belongs to the technical field of radio wave detection. Background Art

[0002] With the increase in the service life of highways and the changes in natural conditions, the road surface is prone to road surface defects such as settlement, cracking, and potholes. Especially in the past two years, due to the influence of extreme weather, road surface defects mainly caused by foundation settlement have become an important factor affecting traffic safety, posing a huge risk to people’s lives and property safety. Therefore, monitoring and diagnosis of highway pavement plays an important role in ensuring vehicle driving safety, reducing traffic congestion, increasing actual road traffic flow, and enhancing the supervision and maintenance capabilities of transportation and related departments.

[0003] The investigation and assessment report on the "May 1" landslide disaster on the Chayang section of the Meida Expressway mentioned that this landslide disaster had the characteristics of strong suddenness, clear sliding boundaries, orderly accumulation before and after, and steep rear wall of the landslide. Comprehensively judged, this landslide disaster first slid as a whole in the middle and lower part of the embankment along the north-south groove direction, and then caused the landslide of the fill on the upper part of the embankment. The main lesson mentioned that there is no standard specification for monitoring and early warning of expressways. There is no monitoring of the stability of key roadbed slopes, and the failure to detect and warn landslides in the first time is far from the requirements of the "five-fast mechanism" of rapid discovery, rapid evacuation, rapid linkage, rapid self-inspection, and rapid intervention. In terms of prevention and rectification measures, it is recommended to promote science and technology to promote safety, improve the risk monitoring and control capabilities of expressways, improve the intelligent comparison of key projects and key parts, strengthen scientific and technological information means, strengthen and improve the disaster monitoring and early warning capabilities of embankment slope projects throughout the life cycle, and provide timely reminder services such as traffic anomalies and emergencies to drivers and passengers and related units in conjunction with the navigation software platform.

[0004] The existing commonly used road monitoring mostly uses road detection vehicles, which use radar, optical sensors, mechanical sensors, etc. to detect the road surface. For example, the patent name is a geological radar detection vehicle, patent number 201920843507.1, which discloses that a partition is provided in the vehicle body to divide the vehicle body into a cab at the front of the vehicle, a control room in the middle and a detection device placement room at the rear of the vehicle, and a control management device, a display device and a power supply device are provided in the control room. A placement rack is provided in the detection device placement room, and a detection vehicle for detection and data collection is provided on the placement rack. The detection vehicle is provided with a four-channel ground penetrating radar device, a high-definition video acquisition device, a GPS positioning device and a transmission device for transmitting data to the control management device, and a rotary encoder connected to the control management device is provided on the wheel hub of the vehicle body. Although this road detection vehicle for detection can effectively discover the actual road conditions and hidden injuries of the road, its detection is periodic and cannot obtain the road conditions of the highway in real time, especially the real-time conditions of the road that affect driving safety.

[0005] Existing highway pavement defect detection mainly relies on the following technologies: Manual inspection and inspection vehicles: long cycle (≥14 days / time), poor timeliness, and inability to provide real-time feedback on sudden defects. Embedded sensors: require large-scale deployment (average cost ≥2 million yuan / km), long construction cycle, and difficult maintenance. Optical video monitoring: significantly affected by light, rain and fog (false alarm rate in bad weather >30%), and limited detection capabilities at night. Safety hazards and economic losses: According to the Ministry of Transport's 2023 statistics, traffic accidents caused by road defects accounted for 18%, with annual direct economic losses exceeding 5 billion yuan.

[0006] Through searching, there are also many technologies that can monitor the road surface in real time, such as the patent name of a highway road surface settlement monitoring and early warning system, patent application number 201811325798.1, which uses a settlement detection unit, auxiliary detection equipment, relay equipment and monitoring center. The settlement detection unit includes multiple displacement sensors, which are all located at the lower end of the road surface and located in the roadbed. The auxiliary detection equipment includes video acquisition equipment, car weighing equipment, vehicle counting equipment and earthquake detection equipment. The earthquake detection equipment is located at the lower end of the road surface and located in the roadbed. The relay equipment includes a data acquisition unit, a wireless transmission unit and an emergency communication device, which can monitor the settlement of the highway road surface in real time. Multi-data collection makes it easier for staff to analyze the causes of road surface settlement in many aspects, and realizes the comprehensive understanding of the monitoring center on the monitored road section, which can timely eliminate the safety hazards caused by the settlement of the highway. Although this technology can monitor the internal conditions in real time, it needs to set up a large number of sensors on the road surface. Due to the long mileage of the highway, a large number of equipment needs to be set up to achieve this, which has the problems of slow construction and high cost.

[0007] There are also methods for road detection based on existing big data, such as a real-time road disease detection method based on radar, video and data analysis, patent application number 202410044080.4, which includes the following steps: establishing a disease detection neural network architecture; obtaining a training sample set; training the disease detection neural network architecture; the inspection vehicle inspects the road surface, inputs the radar scanning point cloud map and video image frame with the same acquisition position information into the trained disease detection neural network architecture, and outputs the disease recognition result. The present invention applies radar, video, GPS and data analysis technology to real-time detection of road diseases, and realizes real-time monitoring and abnormal detection capabilities of road conditions through the steps of multimodal data collection, feature extraction, data fusion and machine learning, which helps to improve the safety, availability and maintenance efficiency of roads. The present invention can accurately and efficiently realize large-scale detection of road diseases, thereby solving the shortcomings of manual detection. Although this method can realize large-scale detection with high efficiency, it needs to be based on video, and has high requirements for video clarity, and also has the problem of high cost. At the same time, video data is seriously affected by fog, rainy days and dark days, and real-time monitoring cannot be realized. Summary of the invention

[0008] In view of this, the purpose of the present invention is to provide a monitoring and early warning system for monitoring highway road surface defects using radar. The system abandons the existing idea of ​​directly monitoring the road surface by video and radar, and judges road surface defects by detecting abnormal fluctuations of the vehicle to solve the problems mentioned in the background technology.

[0009] The monitoring and early warning system for highway road surface defects using radar includes the following functional modules: The signal acquisition module includes a radar and Beidou timing, which is installed on the horizontal bar above the highway or on the vertical bar on the roadside. The radar includes a transmitter, a transmitting antenna, a receiver, a receiving antenna and a signal processing unit, which is used to obtain the real-time reflected wave signal of vehicles traveling on the highway. Beidou timing realizes multi-radar time synchronization; A reflected wave signal extraction module, which is connected to the signal acquisition module, is used to obtain real-time reflected wave signals and their position information, and uses waveform data processing technology to extract features from the real-time reflected wave signals of vehicles traveling on high-speed roads, so as to obtain real-time wave change features of vehicles when non-steady driving occurs, and transmit the acquired real-time wave change features and position information to the wave change feature comparison module, wherein the waveform data processing technology includes a. pre-processing denoising, b. fast Fourier transform extraction of 50-200Hz frequency band, c. synchronous squeezing wavelet transform precise positioning, d. adaptive threshold filtering denoising, e. impact feature enhancement to improve signal-to-noise ratio, and f. three-dimensional feature coding adaptation model; A highway pavement defect map feature library, a. collecting known highway pavement defects that affect vehicle driving to establish a pavement defect library; b. searching for highway pavement defects in the pavement defect library on existing highway sections or replicating highway pavement defects in the pavement defect library on temporary sections; c. setting a signal acquisition module at the pavement defect found or replicated in step b, using the signal acquisition module to obtain wave change characteristics of the vehicle during non-steady driving when it reaches the highway pavement defect found or replicated in step b, collecting the wave change characteristics to form a highway pavement defect map feature library; and grading the risk level of the highway pavement defect map feature library, and setting a wave change characteristic threshold; The wave change feature comparison module is connected to the highway pavement defect map feature library, the reflected wave signal extraction module and the early warning device. After receiving the real-time wave deformation feature and position information obtained from the reflected wave signal extraction module, it is analyzed and compared with the wave change feature in the highway pavement defect map feature library. After the analysis and comparison, if the real-time wave deformation feature is at the level threshold required for alarm, a signal is sent to the early warning device at the corresponding position; A warning device, which is installed next to or 30-50 meters in front of the signal acquisition module, sends warnings to vehicles traveling to the area through warning lights, loudspeakers and / or radio voice warnings; The aforementioned monitoring and early warning system for monitoring highway pavement defects using radar is that the system also includes a supervision center, which can obtain the real-time wave change characteristics and their position information obtained by the reflected wave signal extraction module, perform waveform processing and analysis on the real-time wave change characteristics, and use the highway pavement defect map feature library to invert the pavement defects corresponding to the real-time wave change characteristics, and send the obtained pavement defects and the location information of the pavement defects to the pavement maintenance department for maintenance; the supervision center can also obtain the signal sent by the wave change characteristic comparison module to the early warning device, and send the signal sent to the early warning device to the road administration, traffic police and pavement maintenance departments.

[0010] The aforementioned monitoring and early warning system for monitoring highway road surface defects using radar is that the waveform data processing technology includes: aPre-processing denoising includes denoising and signal alignment: separating high-frequency impact components from low-frequency vibrations through dynamic noise suppression to eliminate environmental noise; beamforming optimization: using spatial filtering of the MIMO radar array to suppress interference signals from lateral vehicles and retain echoes in the main lobe direction; data normalization: normalizing the amplitude of radar intermediate frequency signals to eliminate the influence of distance attenuation and time domain alignment to compensate for phase shifts caused by vehicle motion; b. Fast Fourier transform extraction of 50-200Hz frequency band includes coarse-grained frequency band screening: perform 2048-point FFT on the preprocessed signal, analyze the signal spectrum, and extract the 50-200Hz frequency band related to driving safety; frequency domain energy threshold segmentation: set the energy threshold, filter the low-frequency lateral motion component <20Hz and the ultra-high frequency noise >200Hz, and retain the potentially harmful frequency band; spectrum baseline calibration: highlight the sudden impact signal through background spectrum difference; c Synchronous squeezed wavelet transform precise positioning includes time-frequency resolution improvement: Synchronous squeezed wavelet transform is performed on the signal after FFT screening, and the fuzzy energy of traditional wavelet transform is redistributed to the instantaneous frequency trajectory to achieve 0.2ms / 5Hz high-resolution time-frequency analysis; Transient impact positioning: Separate short-term impacts of <30ms from continuous vibrations, and locate the precise time points of events such as road potholes through instantaneous frequency; d Adaptive threshold filtering noise reduction includes dynamic noise suppression: calculating the instantaneous energy of the signal based on the Teager energy operator, setting an adaptive threshold, and filtering out more than 90% of residual interference; impact event screening: retaining energy surge areas and marking them as potential road hazard events; e. Impact feature enhancement to improve the signal-to-noise ratio includes energy focusing: performing Gaussian energy redistribution on the screened impact signals, and by weighted aggregation of neighborhood energy, the impact energy concentration is increased by 45%, the signal-to-noise ratio is increased by more than 22dB, and the waveform details are highlighted; waveform morphology repair: compensating for the attenuation distortion during signal transmission and restoring the true impact waveform; f. The three-dimensional feature encoding adaptation model includes multi-dimensional feature fusion: encoding the time domain, frequency domain, and energy domain features into three-dimensional vectors, expanding them into high-order features to form a 256-dimensional feature matrix; Dimensionality reduction and classification adaptation: using principal component analysis to compress the features to 18 dimensions for adaptation to subsequent machine learning models.

[0011] The aforementioned monitoring and early warning system using radar to monitor highway pavement defects is that the risk levels of the highway pavement defect map feature library are divided into a general-level highway pavement defect map feature library, a deceleration-level highway pavement defect map feature library, and a dangerous-level highway pavement defect map feature library.

[0012] The aforementioned monitoring and early warning system for highway pavement defects using radar is that a deceleration-level highway pavement defect map feature library and a dangerous-level highway pavement defect map feature library are installed in the wave change feature comparison module. If the real-time wave deformation feature is in the deceleration-level highway pavement defect map feature library, a deceleration prompt signal is sent to the early warning device; if the real-time wave deformation feature is at the lowest threshold of the high-risk level, an emergency stop prompt signal is sent to the early warning device.

[0013] The above-mentioned monitoring and early warning system for monitoring highway road surface defects using radar is that the radar uses a multi-input multi-output MIMO antenna array with a frequency of 10-80GHz, and uses beamforming technology to concentrate the main lobe energy in the direction of vehicle travel and suppress lateral interference signals.

[0014] The aforementioned monitoring and early warning system for monitoring highway pavement defects using radar is that the wave change characteristics of the non-steady driving of the vehicle when it reaches the highway pavement defect found or reproduced in step b obtained by the signal acquisition module include the wave change characteristics formed by different types of vehicles and speeds, and the collected highway pavement defect map feature library includes the wave change characteristics of different types of vehicles at different speeds.

[0015] The above-mentioned monitoring and early warning system using radar to monitor highway pavement defects is that the highway pavement defect map feature library is based on pavement defects when grading risk levels, and the grading is performed manually through test data.

[0016] The aforementioned monitoring and early warning system using radar to monitor highway road surface defects is that the early warning device also includes a text display screen or an electronic speed limit sign.

[0017] The above-mentioned monitoring and early warning system using radar to monitor highway pavement defects is that the highway pavement defects include pavement settlement, pavement collapse and pavement potholes, pavement cracks and pavement congestion.

[0018] Compared with the prior art, the beneficial effects of the present invention are: 1. The present invention changes the idea of ​​directly monitoring the actual condition of the road surface in the existing road survey, and uses 10-80GHz radar to monitor vehicles on the road surface. When a vehicle encounters road defects during driving, it will inevitably produce obvious abnormal vibration, shaking or body tilt. When the body is abnormal, the abnormality will produce a waveform that is significantly different from normal driving for the radar detecting the vehicle. Therefore, the abnormal condition of the road surface can be quickly obtained by monitoring the body posture of the vehicle using radar, and then the passing vehicles can be reminded in time and the abnormal road surface can be repaired; because the monitoring distance of 10-80GHz is 150-4750 meters, it can realize road surface monitoring in a larger area, and darkness has no effect on its detection, and the influence of rain and fog on it can also be controlled, which has the advantages of low cost and suitable for promotion.

[0019] 2. Waveform data processing technology can obtain the required waveform faster and more accurately, reduce the amount of waveform data processing and improve processing efficiency through a combination of preprocessing, fast Fourier transform extraction of 50-200Hz frequency band, or even 100-200Hz frequency band, synchronous squeezing wavelet transform, adaptive threshold filtering, impact feature enhancement and three-dimensional feature encoding.

[0020] 3. Based on radar monitoring of vehicles traveling on highways, a highway road defect map feature library is constructed. It searches for or replicates road defects on highways and uses radar collector waveforms to form a defect map feature library. The feature library is gradually improved with use, and the risk level of the defect map feature library is graded. The real-time wave change characteristics of vehicle movements collected in real time can be compared with the defect map feature library, so that the computer can determine the risk level of the road defect and issue a quick warning. At the same time, the supervision center in the background can also infer the road defect situation through the real-time wave change characteristics and quickly propose countermeasures, such as preparation of repair materials.

[0021] 4. Because different vehicle speeds and vehicle models will produce different wave change characteristics when facing different road defects, the wave change characteristics of different vehicle models at different speeds are collected when constructing the high-speed road defect map feature library, which can realize early warning judgment more accurately and quickly.

[0022] 5. The early warning device is installed next to the signal acquisition module. Its early warning includes the most direct and eye-catching combination of sound, light and electricity. It also includes the ability to automatically occupy the vehicle radio 87.5-108MHz frequency band to play alarm voice, so that it can more comprehensively remind nearby vehicles and reduce the occurrence of accidents.

[0023] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a logic block diagram of Embodiment 1 of the present invention; Figure 2 This is a logic block diagram of embodiment 2 of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the tables in the embodiments of the present invention to 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 embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Example 1. A monitoring and early warning system for highway road surface defects using radar, wherein the highway road surface defects include road surface settlement, road surface collapse, road surface potholes, road surface cracks, and road surface congestion. To detect the road surface defects, Figure 1 , the system consists of the following functional modules: 1. A signal acquisition module, including a radar and a Beidou timing service, which is installed on a horizontal bar above a highway or on a vertical bar on the roadside. The radar includes a transmitter, a transmitting antenna, a receiver, a receiving antenna and a signal processing unit, and is used to obtain real-time reflected wave signals of vehicles traveling on the highway. Beidou timing service realizes multi-radar time synchronization. In the present invention, the signal acquisition module is installed on a gantry above the highway pavement or on the top of a vertical bar on the roadside. The gantry is preferably arranged in a high backfill area prone to settlement, with a vertical height greater than 6 meters. The position information of each horizontal bar or roadside vertical bar installed with the signal acquisition module is calibrated, that is, the coordinate information of the signal acquisition module. The real-time reflected wave signal of the vehicle traveling on the highway is obtained by using the radar, and the Doppler effect is used to realize the speed measurement of the vehicle. Beidou timing service realizes multi-radar time synchronization, so that the real-time reflected wave signal of the vehicle has coordinate information and timestamp.

[0027] The radar requirements are as follows: ‌1.1 Radar frequency band and modulation method‌ ‌Recommended frequency band‌:‌ 24GHz (K band) This frequency band has low atmospheric attenuation characteristics in long-distance detection, supports long-distance signal penetration, and the RF chip technology is mature. ‌FMCW (Frequency Modulated Continuous Wave) modulation is required to improve the ranging range by dynamically adjusting the chirp slope.

[0028] 1.2. Core performance parameters, Detection distance: ≥1 km: High transmit power (≥20dBm) is required, and MIMO antenna array is used to improve signal-to-noise ratio. Vertical resolution: Pitch angle resolution: ≤5°, using multi-antenna beamforming technology to capture vehicle up and down shaking (amplitude ≥5cm).

[0029] 1.3. Antenna and hardware design Antenna configuration: MIMO array (e.g. 8 transmit and 16 receive), which expands spatial coverage and enhances vertical resolution.

[0030] ‌Beam width‌: horizontal ≤24°, pitch ≤15°, focusing on the target area to reduce interference, ‌Transmit power‌: ≥20dBm, with a low noise amplifier (LNA) to improve the echo signal strength.

[0031] 2. A reflected wave signal extraction module, which is connected to the signal acquisition module, is used to obtain real-time reflected wave signals and their position information and use waveform data processing technology to extract features from the real-time reflected wave signals of vehicles traveling on high-speed roads, so as to obtain the real-time wave change characteristics of vehicles when non-steady driving occurs, and transmit the obtained real-time wave change characteristics and position information to the wave change feature comparison module. The waveform data processing technology includes a. pre-processing denoising, b. fast Fourier transform extraction of 50-200Hz frequency band, c. synchronous squeezing wavelet transform precise positioning, d. adaptive threshold filtering denoising, e. impact feature enhancement to improve signal-to-noise ratio, f. three-dimensional feature coding adaptation model, a Preprocessing denoising ‌Noise Removal and Signal Alignment‌: Separate high-frequency impact components from low-frequency vibrations and eliminate environmental noise (such as crosswind and electromagnetic interference) through dynamic noise suppression (such as complementary lumped empirical mode decomposition (CEEMD)).

[0032] ‌Beamforming Optimization‌: Utilizes spatial filtering of the MIMO radar array to suppress interference signals from lateral vehicles (targets outside of ±5°) and retain echoes in the main lobe direction (longitudinal vehicles).

[0033] ‌Data Standardization‌: The radar intermediate frequency signal is amplitude normalized (to eliminate the influence of distance attenuation) and time-domain aligned (to compensate for the phase shift caused by vehicle motion).

[0034] b Fourier transform feature extraction 50-200Hz frequency band‌ ‌Coarse-grained frequency band screening‌: Perform a ‌2048-point FFT‌ on the preprocessed signal, analyze the signal spectrum, and extract the ‌50-200Hz frequency band‌ related to driving safety (corresponding to the high-frequency characteristics of impact events such as road potholes and tire bounce).

[0035] ‌Frequency domain energy threshold segmentation‌: Set the energy threshold (such as >3% of total energy), filter out low-frequency lateral motion components (<20Hz) and ultra-high frequency noise (>200Hz), and retain potentially harmful frequency bands.

[0036] ‌Spectrum baseline calibration‌: By differentiating the background spectrum (current frame spectrum - historical average spectrum), sudden impact signals (such as sudden appearance of foreign objects on the road) are highlighted.

[0037] c Synchronous squeeze wavelet transform accurate positioning Improved time-frequency resolution: Perform synchronous squeezed wavelet transform on the FFT-filtered signal, redistribute the fuzzy energy of traditional wavelet transform to the instantaneous frequency trajectory, and achieve high-resolution time-frequency analysis of 0.2ms / 5Hz.

[0038] ‌Transient impact location‌: Separates short-term impacts (<30ms) from continuous vibrations (such as engine vibrations), and locates the precise time point of events such as road potholes through the instantaneous frequency f0(t,a) (error <0.5ms).

[0039] ‌Lateral Interference Suppression‌: Low-frequency energy corresponding to lateral motion (such as <20Hz vibration caused by lane change) is automatically filtered out during the SWT reallocation process because the instantaneous frequency deviates from the target frequency band.

[0040] dAdaptive threshold filtering noise reduction‌ Dynamic noise suppression: Based on the Teager Energy Operator (TEO): E(t)=x2(t)−x(t−1)x(t+1), the instantaneous energy of the signal is calculated, and the adaptive threshold Threshold=2.5×median(|E(t)|) is set to filter out more than 90% of the residual interference.

[0041] ‌Initial screening of impact events‌: retain areas with sudden energy increases (such as peaks where the TEO amplitude exceeds the threshold) and mark them as potential road hazard events (such as potholes and speed bumps).

[0042] e-shock feature enhancement improves signal-to-noise ratio‌ ‌Energy Focusing‌: Perform ‌Gaussian Energy Redistribution‌ on the filtered impact signals: Eenhanced(t,f)= (4σ Gaussian window weighted) By 4σ weighted aggregation of neighborhood energy, the impact energy concentration is increased by 45%, the signal-to-noise ratio is improved by more than 22dB‌, and the waveform details (such as rising edge slope and oscillation decay rate) are highlighted. Eenhanced(t,f) or E(t,f): This represents the enhanced signal or energy, which is a function of time and frequency. : means to sum all integer values ​​of variable k from -4 to 4. This range limits the value of frequency offset.

[0043] k: is the index variable for summation, indicating the index of the frequency offset. In this formula, the value range of k is from -4 to 4, which is used to traverse different frequency offsets.

[0044] Tswt(t,f+kΔf): This represents a certain conversion or transformation function at time t and frequency f+kΔf. This function may represent a certain characteristic or transformation result of the signal at a specific time and frequency.

[0045] :This is an exponential function used to weight Tswt(t,f+kΔf). As k increases or decreases, this weighting term will gradually decrease, thereby achieving different degrees of weighting for signals with different frequency offsets.

[0046] Δf: This represents the frequency interval or frequency resolution, which is the step size of the frequency offset. It determines the interval between the values ​​of the frequency offset kΔf during the summation process.

[0047] ‌Waveform morphology repair‌: Compensate for attenuation distortion during signal transmission and restore the true impact waveform (such as the force-time curve of a pothole).

[0048] f3D feature encoding adaptation model‌ ‌Multi-dimensional feature fusion‌: Encode the time domain, frequency domain, and energy domain features into a three-dimensional vector: F=[t,f,E(t,f)] (time, frequency, energy density) and expand them into high-order features (such as kurtosis, envelope spectral entropy, instantaneous frequency slope) to form a ‌256-dimensional feature matrix‌.

[0049] Dimensionality reduction and classification adaptation: Use principal component analysis (PCA) to compress features to 18 dimensions (retaining 95% of the information), and adapt subsequent machine learning models (such as SVM, CNN) to classify road conditions (flat / cracked / pitted) 3. Highway pavement defect map feature library, a. Collect known highway pavement defects that affect vehicle driving to establish a pavement defect library, b. Find highway pavement defects in the pavement defect library on existing highway sections or replicate highway pavement defects in the pavement defect library on temporary sections, c. Set a signal acquisition module at the pavement defect found or replicated in step b, use the signal acquisition module to obtain the wave change characteristics of non-steady driving when different types of vehicles travel at different speeds to the highway pavement defect found or replicated in step b, collect the wave change characteristics to form a highway pavement defect map feature library; and classify the risk level of the highway pavement defect map feature library, and set the wave change characteristic threshold; the highway pavement defect map feature library is based on the pavement defects when grading the risk level, and is manually graded through test data, and the same pavement defects on the driving road belong to the same risk level. Highway pavement defect map feature library storage architecture: MongoDB sharding cluster is used to store time series data, supporting PB-level data expansion; retrieval optimization: feature vector index is constructed based on Faiss framework, and retrieval delay is less than 5ms.

[0050] Data collection plan stage content Implementation Phase I Basic defect sample library 50 defects × 10 speeds × 8 vehicle types Stage II Environmental Enhancement Library Rain / fog / night scene simulation Stage III Online Learning Updates LSTM dynamically corrects feature thresholds The risk levels of the highway pavement defect map feature library are divided into general-level highway pavement defect map feature library, deceleration-level highway pavement defect map feature library and dangerous-level highway pavement defect map feature library. The general level refers to pavement defects that will cause vehicle body vibration and shaking but do not affect vehicle safety. The deceleration-level refers to pavement defects that can be safely passed below the specified speed. The dangerous level refers to pavement defects that may cause vehicle damage and accidents, such as pavement cracks and collapses. Deploy a deep convolutional network (ResNet-18 architecture), with a training sample size of ≥ 100,000 sets of abnormal vibration waveforms Risk Level Trigger conditions (example) Corresponding defect type General Amplitude standard deviation Δσ<0.3g Minor cracks (width < 2cm) Reduction stage 0.3g≤Δσ<0.8g Medium pits (depth > 5 cm) Danger level Δσ≥0.8g or tilt angle>5° for 2 seconds Collapsed area (area>1㎡) 4. A wave change feature comparison module is connected to the highway pavement defect map feature library, the reflected wave signal extraction module and the early warning device. After receiving the real-time wave deformation feature and position information from the reflected wave signal extraction module, it is analyzed and compared with the wave change feature in the highway pavement defect map feature library. After the analysis and comparison, if the real-time wave deformation feature is at the alarm level threshold, a signal is sent to the early warning device at the corresponding position; if the real-time wave deformation feature is within the general-level highway pavement defect map feature library area, no signal is sent to the early warning device, but only to the supervision center; if the real-time wave deformation feature is in the deceleration-level highway pavement defect map feature library, a deceleration prompt signal is sent to the early warning device at the corresponding position; if the real-time wave deformation feature is at the lowest threshold of the high-risk level, an emergency stop prompt signal is sent to the early warning device at the corresponding position.

[0051] The wave change feature comparison module uses real-time comparison technology, including dynamic time warping (DTW): matching real-time waveforms with feature library templates, and tolerating time axis expansion and speed differences. Deep learning assistance: training a 1D-CNN model (structure: Conv1D(64)-MaxPool-GRU(32)-Dense) to classify waveform segments, with a classification accuracy of ≥96%.

[0052] 5. The early warning device is installed next to the signal acquisition module or 30-50 meters in front of it. It sends warnings to vehicles traveling to the area through warning lights, loudspeakers and / or automatically occupies the vehicle radio 87.5-108MHz frequency band to play alarm voice; a text display screen or electronic speed limit sign can also be installed on the gantry where the radar is installed to provide direct and clear warnings.

[0053] 6. The supervision center can obtain the real-time wave change characteristics and their location information obtained by the reflected wave signal extraction module, perform waveform processing and analysis on the real-time wave change characteristics, and use the high-speed road surface defect spectrum feature library to invert the road surface defects corresponding to the real-time wave change characteristics, and send the obtained road surface defects and the location information of the road surface defects to the road maintenance department for maintenance; the supervision center can also obtain the signal sent by the wave change feature comparison module to the early warning device, and send the signal sent to the early warning device to the road administration and traffic police road maintenance department after confirmation. The calculation of the location of the road surface defect can be calculated through the location information of the signal acquisition module, the vehicle speed measured by the Doppler effect, and the time when the road surface defect occurred.

[0054] Example 2. Figure 2 The difference between the monitoring and early warning system for highway pavement defects using radar and Example 1 is that the highway pavement defect map feature library is divided into two categories. One is a 100% highway pavement defect map feature library installed in the supervision center, and the highway pavement defect map feature library will be supplemented according to new pavement defects. The other is a wave change feature comparison module installed near the radar, which only has a deceleration-level highway pavement defect map feature library and a dangerous-level highway pavement defect map feature library. This can reduce the amount of comparison data of the road side wave change feature comparison module and improve the comparison data efficiency.

[0055] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A monitoring and early warning system for highway road surface defects using radar, characterized in that: The system includes the following functional modules: The signal acquisition module includes a radar and Beidou timing, which is installed on the horizontal bar above the highway or on the vertical bar on the roadside. The radar includes a transmitter, a transmitting antenna, a receiver, a receiving antenna and a signal processing unit, which is used to obtain the real-time reflected wave signal of vehicles traveling on the highway. Beidou timing realizes multi-radar time synchronization; A reflected wave signal extraction module, which is connected to the signal acquisition module, is used to obtain real-time reflected wave signals and their position information, and uses waveform data processing technology to extract features from the real-time reflected wave signals of vehicles traveling on high-speed roads, so as to obtain real-time wave change features of vehicles when non-steady driving occurs, and transmit the acquired real-time wave change features and position information to the wave change feature comparison module, wherein the waveform data processing technology includes a. pre-processing denoising, b. fast Fourier transform extraction of 50-200Hz frequency band, c. synchronous squeezing wavelet transform precise positioning, d. adaptive threshold filtering denoising, e. impact feature enhancement to improve signal-to-noise ratio, and f. three-dimensional feature coding adaptation model; A highway pavement defect map feature library, a. collecting known highway pavement defects that affect vehicle driving to establish a pavement defect library; b. searching for highway pavement defects in the pavement defect library on existing highway sections or replicating highway pavement defects in the pavement defect library on temporary sections; c. setting a signal acquisition module at the pavement defect found or replicated in step b, using the signal acquisition module to obtain wave change characteristics of the vehicle during non-steady driving when it reaches the highway pavement defect found or replicated in step b, collecting the wave change characteristics to form a highway pavement defect map feature library; and grading the risk level of the highway pavement defect map feature library, and setting a wave change characteristic threshold; The wave change feature comparison module is connected to the highway pavement defect map feature library, the reflected wave signal extraction module and the early warning device. After receiving the real-time wave deformation feature and position information obtained from the reflected wave signal extraction module, it is analyzed and compared with the wave change feature in the highway pavement defect map feature library. After the analysis and comparison, if the real-time wave deformation feature is at the level threshold required for alarm, a signal is sent to the early warning device at the corresponding position; The warning device is installed next to the signal acquisition module or 30-50 meters in front of it. It sends warnings to vehicles traveling to the area through warning lights, loudspeakers and / or radio voice warnings.

2. The monitoring and early warning system for highway road surface defects using radar according to claim 1 is characterized in that: The system also includes a supervision center, which can obtain the real-time wave change characteristics and their position information obtained by the reflected wave signal extraction module, perform waveform processing and analysis on the real-time wave change characteristics, and use the high-speed pavement defect spectrum feature library to invert the pavement defects corresponding to the real-time wave change characteristics, and send the obtained pavement defects and the location information of the pavement defects to the pavement maintenance department for maintenance; the supervision center can also obtain the signal sent by the wave change characteristic comparison module to the early warning device, and send the signal sent to the early warning device to the road administration, traffic police and pavement maintenance departments.

3. The monitoring and early warning system for highway road surface defects using radar according to claim 2 is characterized in that: The waveform data processing technology includes: aPre-processing denoising includes denoising and signal alignment: separating high-frequency impact components from low-frequency vibrations through dynamic noise suppression to eliminate environmental noise; beamforming optimization: using spatial filtering of the MIMO radar array to suppress interference signals from lateral vehicles and retain echoes in the main lobe direction; data normalization: normalizing the amplitude of radar intermediate frequency signals to eliminate the influence of distance attenuation and time domain alignment to compensate for phase shifts caused by vehicle motion; b. Fast Fourier transform extraction of 50-200Hz frequency band includes coarse-grained frequency band screening: perform 2048-point FFT on the preprocessed signal, analyze the signal spectrum, and extract the 50-200Hz frequency band related to driving safety; frequency domain energy threshold segmentation: set the energy threshold, filter the low-frequency lateral motion component <20Hz and the ultra-high frequency noise >200Hz, and retain the potentially harmful frequency band; spectrum baseline calibration: highlight the sudden impact signal through background spectrum difference; c Synchronous squeezed wavelet transform precise positioning includes time-frequency resolution improvement: Synchronous squeezed wavelet transform is performed on the signal after FFT screening, and the fuzzy energy of traditional wavelet transform is redistributed to the instantaneous frequency trajectory to achieve 0.2ms / 5Hz high-resolution time-frequency analysis; Transient impact positioning: Separate short-term impacts of <30ms from continuous vibrations, and locate the precise time points of events such as road potholes through instantaneous frequency; d Adaptive threshold filtering noise reduction includes dynamic noise suppression: calculating the instantaneous energy of the signal based on the Teager energy operator, setting an adaptive threshold, and filtering out more than 90% of residual interference; impact event screening: retaining energy surge areas and marking them as potential road hazard events; e. Impact feature enhancement to improve the signal-to-noise ratio includes energy focusing: performing Gaussian energy redistribution on the screened impact signals, and by weighted aggregation of neighborhood energy, the impact energy concentration is increased by 45%, the signal-to-noise ratio is increased by more than 22dB, and the waveform details are highlighted; waveform morphology repair: compensating for the attenuation distortion during signal transmission and restoring the true impact waveform; f. The three-dimensional feature encoding adaptation model includes multi-dimensional feature fusion: encoding the time domain, frequency domain, and energy domain features into three-dimensional vectors, expanding them into high-order features to form a 256-dimensional feature matrix; Dimensionality reduction and classification adaptation: using principal component analysis to compress the features to 18 dimensions for adaptation to subsequent machine learning models.

4. The monitoring and early warning system for highway road surface defects using radar according to claim 3 is characterized in that: The risk levels of the highway pavement defect map feature library are divided into a general-level highway pavement defect map feature library, a deceleration-level highway pavement defect map feature library, and a dangerous-level highway pavement defect map feature library.

5. The monitoring and early warning system for highway road surface defects using radar according to claim 4 is characterized in that: The wave change feature comparison module is equipped with a deceleration level highway pavement defect map feature library and a dangerous level highway pavement defect map feature library. If the real-time wave deformation feature is in the deceleration level highway pavement defect map feature library, a deceleration prompt signal is sent to the early warning device; if the real-time wave deformation feature is at the lowest threshold of the high-risk level, an emergency stop prompt signal is sent to the early warning device.

6. The monitoring and early warning system for monitoring highway road surface defects using radar according to any one of claims 1 to 5, characterized in that: The radar adopts a multiple-input multiple-output MIMO antenna array with a frequency of 10-80GHz, and uses beamforming technology to concentrate the main lobe energy in the direction of vehicle travel and suppress lateral interference signals.

7. The monitoring and early warning system for highway road surface defects using radar according to claim 6 is characterized in that: The signal acquisition module is used to obtain the wave change characteristics of the non-steady driving of the vehicle when it reaches the high-speed road surface defect found or reproduced in step b, which includes the wave change characteristics formed by different types of vehicles and speeds. The collected high-speed road surface defect map feature library includes the wave change characteristics of different types of vehicles at different speeds.

8. The monitoring and early warning system for highway road surface defects using radar according to claim 7 is characterized in that: The highway pavement defect map feature library is based on pavement defects when grading risk levels, and is graded manually through test data.

9. The monitoring and early warning system for highway road surface defects using radar according to claim 8 is characterized in that: The early warning device also includes a text display screen or an electronic speed limit sign.

10. The monitoring and early warning system for highway road surface defects using radar according to claim 9, characterized in that: The highway pavement defects include pavement subsidence, pavement collapse, pavement potholes, pavement cracks and pavement bumps.

Images