Road defect judgment and map making method and device, electronic equipment and medium

Abstract

This invention provides a method, apparatus, electronic device, and medium for road surface defect assessment and map creation. The method includes: clustering abnormal event data uploaded by multiple vehicle terminals to obtain abnormal event clusters and confidence levels; determining the road surface defect level based on vehicle speed and weather data, peak vehicle acceleration in the vertical direction of the road surface, and road surface material type corresponding to abnormal event clusters with confidence levels greater than a preset confidence threshold; locating the location of emergency avoidance behavior based on vehicle avoidance data and vehicle geographic coordinates to determine the geographic coordinates of the road surface defect and its lane; and generating a road map containing road surface defect information based on the road surface defect type, the road surface defect level, the geographic coordinates of the road surface defect, and its lane. This invention solves the problem that existing solutions cannot achieve all-weather, full-network, low-cost, high-real-time, and high-precision dynamic monitoring and map creation of road surface defects in complex road environments.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, specifically to a method, apparatus, electronic device, and medium for road surface defect assessment and map creation. Background Technology

[0002] my country's urbanization and highway network expansion have led to increased road traffic load and frequent road surface defects, threatening driving safety and shortening the service life of roads.

[0003] Currently, there are two main technical approaches to road defect detection: one is intelligent methods based on visual recognition, which analyze road images using deep learning models such as YOLO and EfficientDet. Through attention mechanisms, lightweight network design, and multi-scale feature fusion optimization, these methods have achieved significant results in terms of accuracy and efficiency in crack and defect detection. However, these methods are highly dependent on the quality of image acquisition, and their recognition performance drops significantly in complex environments such as nighttime, rain, snow, and occlusion. The other is non-destructive testing methods based on specialized equipment, with typical technologies including ground-penetrating radar, ultrasonic detection, and laser point cloud scanning. These methods can assess the internal structure and material strength of the road surface and have high detection accuracy. However, they suffer from problems such as expensive equipment, specialized operation, complex processes, and low efficiency. They are only suitable for periodic sampling inspections and cannot meet the needs of large-scale, high-frequency dynamic monitoring.

[0004] In recent years, the development of vehicle-to-everything (V2X) and intelligent sensing technologies has made it possible for ordinary vehicles and even smartphones to dynamically perceive road conditions using sensors such as accelerometers and GPS, providing new opportunities for low-cost, wide-coverage crowdsourced detection models. Existing research has attempted to identify road conditions through tire noise and vehicle vibration signals, initially demonstrating the potential of mobile sensing. However, most solutions are limited to a single sensor or data source, and there are still significant shortcomings in data reliability and real-time judgment accuracy.

[0005] In summary, existing technologies have failed to resolve the core contradiction between high precision and wide coverage in road defect detection. Neither visual recognition nor professional non-destructive testing can achieve all-weather, full-network, low-cost, high-real-time, and high-precision dynamic monitoring and map creation of road defects in complex road environments. Summary of the Invention

[0006] In view of this, it is necessary to provide a method, device, electronic device and medium for road surface defect assessment and mapping, in order to solve the technical problem that existing solutions cannot achieve all-weather, full-network, low-cost, high real-time and high-precision dynamic monitoring and mapping of road surface defects in complex road environments.

[0007] To address the aforementioned problems, in a first aspect, the present invention provides a method for assessing road surface defects and creating maps, comprising:

[0008] Clustering is performed on abnormal event data uploaded by multiple vehicle terminals within the target area to obtain abnormal event clusters and their confidence levels. Based on the vehicle speed and weather data corresponding to the clusters of abnormal events with a confidence level greater than the preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface, and the road material type, the road defect level is determined. Based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, the location of the emergency avoidance behavior is located, and the geographic coordinates of the road defect and the lane in which it is located are determined. Based on the type of road surface defect, the level of the road surface defect, the geographical coordinates of the road surface defect and the lane where it is located, a road map containing road surface defect information is generated by matching it with a preset road map. The abnormal event data is obtained by the vehicle terminal based on the following steps: Based on vehicle speed, vehicle acceleration, road surface data, and weather data, anomaly type identifiers are determined; the anomaly type identifiers include: road surface defect type and vehicle avoidance data. Abnormal event data is constructed based on the vehicle's geographic coordinates, vehicle speed, vehicle acceleration, road surface data, weather data, and anomaly type identifier.

[0009] In one possible implementation, the vehicle acceleration is obtained by the vehicle terminal based on the following steps: Collect vehicle terminal acceleration data and vehicle terminal motion direction data; Based on the vehicle terminal acceleration data and the vehicle terminal motion direction data, the relative pose relationship between the vehicle terminal and the vehicle is determined. Based on the relative pose relationship, the vehicle terminal acceleration data is converted into vehicle acceleration data.

[0010] In one possible implementation, the road defect level is determined based on vehicle speed and weather data corresponding to a cluster of abnormal events with a confidence level greater than a preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface, and the road material type, including: For the vehicle speed and weather data corresponding to the clusters of abnormal events with a confidence level greater than a preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface is standardized; the peak vehicle acceleration in the vertical direction of the road surface is extracted based on the abnormal event data. The road defect level is determined based on the average of the standardized peak vehicle acceleration and the type of road material.

[0011] In one possible implementation, the location of the emergency avoidance behavior is determined based on vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, including: Based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, the location of the emergency avoidance behavior is located to obtain the location of the emergency avoidance behavior. Based on the location of the emergency avoidance action and the direction of the vehicle's lane change, the geographical coordinates of the road defect and its lane are determined.

[0012] In one possible implementation, an anomaly type identifier is determined based on vehicle speed, vehicle acceleration, road surface data, and weather data, including: Based on the vehicle speed, the road surface material type, and the weather data, an environmental adaptive dynamic threshold is constructed. Based on the relationship between the changes in vehicle acceleration in the vertical direction of the road surface and the magnitude of the adaptive dynamic threshold of the environment, an anomaly type identifier is determined.

[0013] In one possible implementation, the road surface defect assessment and mapping method also includes: Upon receiving abnormal event data uploaded by the terminals of subsequent vehicles, the system verifies the geographical coordinates of the road surface defects and their lanes determined based on the abnormal event data uploaded by the terminals of the preceding vehicles, and generates a verification report. The road map containing pavement defect information is updated based on the verification report.

[0014] In one possible implementation, the method for assessing road surface defects and creating maps... Based on the abnormal event data uploaded by subsequent vehicles, the geographical coordinates and lane locations of road defects determined from the abnormal event data uploaded by the terminals of previous vehicles are verified, and a verification report is generated, including: If it is determined that a subsequent vehicle has evasive behavior or has run over a road surface defect based on abnormal event data uploaded by subsequent vehicles, the confidence level of the geographical coordinates of the road surface defect and the lane it is located in, determined based on the abnormal event data uploaded by the terminal of the preceding vehicle, is increased, and a corresponding verification report is generated.

[0015] Secondly, the present invention also provides a road surface defect assessment and map making device, comprising: The clustering module is used to cluster abnormal event data uploaded by multiple vehicle terminals within the target area to obtain abnormal event clusters and their confidence levels. The defect identification module is used to determine the road defect level based on the vehicle speed and weather data corresponding to the abnormal event clusters with a confidence level greater than a preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface, and the road material type. The defect location module is used to locate the location of the emergency avoidance behavior based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, and to determine the geographic coordinates of the road defect and the lane in which it is located. The map drawing module is used to match the road surface defect type, the road surface defect level, the geographical coordinates of the road surface defect and the lane it is located with a preset road map to generate a road map containing road surface defect information. The abnormal event data is obtained by the vehicle terminal based on the following steps: Based on vehicle speed, vehicle acceleration, road surface data, and weather data, anomaly type identifiers are determined; the anomaly type identifiers include: road surface defect type and vehicle avoidance data. Abnormal event data is constructed based on the vehicle's geographic coordinates, vehicle speed, vehicle acceleration, road surface data, weather data, and anomaly type identifier.

[0016] Thirdly, the present invention also provides an electronic device, including a memory and a processor, wherein, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps of the road surface defect assessment and map making method as described in any of the above.

[0017] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the road surface defect assessment and map making method as described in any of the preceding claims.

[0018] The beneficial effects of adopting the above implementation method are as follows: The road surface defect assessment and map making method, device, electronic device and medium provided by the present invention acquire abnormal event data uploaded by multiple vehicle terminals, and cluster the abnormal event data. Based on the abnormal event clusters and their confidence levels, as well as the peak vehicle acceleration in the vertical direction of the road surface and the road material type, the road surface defect level is determined. Then, the location of the emergency avoidance behavior is located by the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, the geographic coordinates of the road surface defect and the lane in which it is located are determined, and then it is marked on a preset lane map to create a map containing the road surface defect type, road surface defect level, road surface defect geographic coordinates and the lane in which it is located.

[0019] The abnormal event data comes from multiple vehicle terminals, and the abnormal events uploaded by each terminal are constructed based on the vehicle speed, vehicle acceleration, road surface data and weather data collected by the vehicle terminal itself. Compared with the data collected by a single sensor of a single vehicle, the abnormal event data provided by this invention has a wider range of sources, and the source is not just one vehicle, but multiple vehicles driving on the road, which can realize all-weather, all-road network, low cost and high real-time data collection.

[0020] The abnormal event data provided by this invention has a wide coverage, and the abnormal event data is clustered for analysis. The road surface defect level is determined based on the abnormal event clusters with high confidence. Compared with the analysis results obtained from data collected from a single vehicle, the results obtained by this invention are more accurate.

[0021] Furthermore, the abnormal event data collected by this invention can be obtained based on vehicle speed and acceleration without relying on cameras or optical sensors. This effectively overcomes the interference of complex environments such as nighttime, rain, snow, fog, and obstruction on traditional visual methods. It can operate stably under various weather and lighting conditions, significantly improving the robustness of the system.

[0022] Therefore, the method provided by this invention solves the technical problem that existing solutions cannot achieve all-weather, full-network, low-cost, high-real-time, and high-precision dynamic monitoring and map production of road surface defects in complex road environments. Attached Figure Description

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

[0024] Figure 1 A flowchart of an embodiment of the road surface defect assessment and map creation method provided by the present invention; Figure 2 A flowchart of another embodiment of the road surface defect assessment and map creation method provided by the present invention; Figure 3 A schematic diagram of an embodiment of the road surface defect assessment and map making device provided by the present invention; Figure 4 A schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0026] In the description of the embodiments of this application, unless otherwise stated, "a plurality of" means two or more.

[0027] In this embodiment of the invention, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, apparatus, product or device that includes a series of steps or modules is not necessarily limited to those steps or modules that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product or device.

[0028] The naming or numbering of steps in the embodiments of the present invention does not mean that the steps in the method flow must be executed in the time / logical order indicated by the naming or numbering. The execution order of the named or numbered process steps can be changed according to the technical purpose to be achieved, as long as the same or similar technical effect can be achieved.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] This invention provides a method, apparatus, electronic device, and medium for assessing road surface defects and creating maps, which will be described below.

[0031] This invention provides a method for assessing road surface defects and creating maps. This method can be implemented by executing an application on a cloud platform, i.e., a cloud server. Figure 1 As shown, the method includes: S101. Cluster the abnormal event data uploaded by multiple vehicle terminals in the target area to obtain abnormal event clusters and their confidence levels.

[0032] The abnormal event data is obtained by the vehicle terminal based on the following steps: Based on vehicle speed, vehicle acceleration, road surface data, and weather data, anomaly type identifiers are determined; the anomaly type identifiers include: road surface defect type and vehicle avoidance data. Abnormal event data is constructed based on the vehicle's geographic coordinates, vehicle speed, vehicle acceleration, road surface data, weather data, and anomaly type identifier.

[0033] Understandably, a vehicle terminal, that is, a terminal installed on a vehicle, can be a mobile phone terminal or other in-vehicle terminal.

[0034] For abnormal event data uploaded by multiple vehicle terminals in the same geographical area (e.g., within a 10-meter radius) and within 2 hours, with the same driving direction, a spatiotemporal clustering algorithm is used to generate high-confidence event clusters.

[0035] S102. Based on the vehicle speed and weather data corresponding to the abnormal event clusters with confidence levels greater than the preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface, and the road material type, determine the road defect level.

[0036] Understandably, the mean value of the peak vehicle acceleration along the Z-axis (i.e., the direction perpendicular to the road surface) after standardization within the cluster is calculated, and the defect level judgment rule is dynamically adjusted in combination with the road surface material type of the road section: when the mean value is less than 5 m / s², it is considered a minor defect; when the mean value is not less than 5 m / s² and not greater than 8 m / s², it is considered a moderate defect; and when the mean value is greater than 8 m / s², it is considered a severe defect.

[0037] S103. Based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, locate the location of the emergency avoidance behavior and determine the geographic coordinates of the road defect and the lane in which it is located.

[0038] Understandably, by statistically clustering all avoidance directions and combining the number of lanes on the road segment with GPS positioning accuracy, the direction that most vehicles would detour in can be inferred, and the lane number where the defect actually is located can be determined.

[0039] S104. Based on the type of road surface defect, the level of the road surface defect, the geographical coordinates of the road surface defect and the lane where it is located, match it with a preset road map to generate a road map containing road surface defect information.

[0040] Understandably, the assessment results are stored in a structured manner, including the lane, defect level, geographical coordinates, station number, driving direction, and defect type (concave / convex). Based on this, a road defect map is generated that is updated every 5–15 minutes. On the electronic map, the risk level is represented by a color gradient: red indicates severe defects (high risk), orange indicates moderate defects (medium risk), and yellow indicates minor defects (low risk).

[0041] In some embodiments, the vehicle acceleration is obtained by the vehicle terminal based on the following steps: Collect vehicle terminal acceleration data and vehicle terminal motion direction data; Based on the vehicle terminal acceleration data and the vehicle terminal motion direction data, the relative pose relationship between the vehicle terminal and the vehicle is determined. Based on the relative pose relationship, the vehicle terminal acceleration data is converted into vehicle acceleration data.

[0042] It is understandable that the vehicle terminal's acceleration data is collected through sensors built into the vehicle terminal, including: the three-axis linear acceleration output by the accelerometer (unit: m / s²); and the three-axis angular velocity output by the gyroscope (unit: rad / s).

[0043] In addition, the GPS module of the vehicle terminal can also output location, speed and heading information (i.e. direction of movement); at the same time, it can obtain real-time vehicle speed and road material type (such as asphalt, cement) through map navigation API (application interface), and obtain current weather conditions (such as sunny, rainy, snowy) through meteorological service API.

[0044] On the mobile device, a complementary filtering algorithm is used to fuse three-axis acceleration and three-axis angular velocity data, and combine them with the motion direction provided by GPS to calculate the mobile device's attitude angle (i.e., pose data) relative to the vehicle.

[0045] In some embodiments, the road defect level is determined based on vehicle speed and weather data corresponding to a cluster of abnormal events with a confidence level greater than a preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface, and the road material type, including: For the vehicle speed and weather data corresponding to the clusters of abnormal events with a confidence level greater than a preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface is standardized; the peak vehicle acceleration in the vertical direction of the road surface is extracted based on the abnormal event data. The road defect level is determined based on the average of the standardized peak vehicle acceleration and the type of road material.

[0046] Understandably, the standardization phase involves standardizing the peak vehicle acceleration along the original Z-axis (vertical direction of the road surface) for each reported data point within the cluster, based on its corresponding vehicle speed and weather conditions.

[0047] Grading stage: Calculate the mean value of the standardized Z-axis acceleration peak within the cluster, and dynamically adjust the defect level judgment rules in combination with the pavement material type of the road section: when the mean value is less than 5 m / s², it is identified as a minor defect; when the mean value is not less than 5 m / s² and not greater than 8 m / s², it is identified as a moderate defect; when the mean value is greater than 8 m / s², it is identified as a severe defect.

[0048] In some embodiments, the location of the emergency avoidance behavior is determined based on vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, including: Based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, the location of the emergency avoidance behavior is located to obtain the location of the emergency avoidance behavior. Based on the location of the emergency avoidance action and the direction of the vehicle's lane change, the geographical coordinates of the road defect and its lane are determined.

[0049] Understandably, lane-level positioning is achieved using Y-axis avoidance behavior. For vehicles deemed to have made a valid avoidance maneuver, the lateral lane-changing direction is determined by the sign of their peak Y-axis acceleration—a Y-axis acceleration greater than zero indicates a right lane change, and less than zero indicates a left lane change. The number of vehicles avoiding to the left (NL) and the number of vehicles avoiding to the right (NR) are counted, and combined with the total number of lanes K of the road segment (which can be provided by high-precision maps, traffic management data, or roadside units), the lane where the defect is located is inferred based on the distribution characteristics of the avoidance direction: if NL and NR are both significant and approximately equal, the defect is located in the middle lane; if NL is significantly greater than NR, it indicates that vehicles mainly detour to the left, and the defect is located in the rightmost lane; if NR is significantly greater than NL, the defect is located in the leftmost lane. For multi-lane scenarios, the system outputs the relative lane position identifier of the defect (such as "left edge lane," "middle lane," or "right edge lane") based on the basic principle that "boundary lanes lead to one-way avoidance, and middle lanes allow multi-way avoidance," achieving lane-level positioning without relying on absolute lane numbers.

[0050] In some embodiments, an anomaly type identifier is determined based on vehicle speed, vehicle acceleration, road surface data, and weather data, including: Based on the vehicle speed, the road surface material type, and the weather data, an environmental adaptive dynamic threshold is constructed. Based on the relationship between the changes in vehicle acceleration in the vertical direction of the road surface and the magnitude of the adaptive dynamic threshold of the environment, an anomaly type identifier is determined.

[0051] Understandably, the environmentally adaptive dynamic threshold is dynamically adjusted based on real-time vehicle speed, road material type, and weather conditions: for example, the higher the vehicle speed, the larger the threshold; different road materials correspond to different baseline settings; and the threshold can also be adaptively corrected under special weather conditions such as rain and snow. When the Z-axis acceleration suddenly drops downward (negative peak), it is initially identified as a depression-type road defect (such as potholes or collapses); when the Z-axis acceleration suddenly rises upward (positive peak), it is initially identified as a bulge-type road defect (such as manhole cover protrusions, misalignments, or bulges). If the absolute value of the peak exceeds the current dynamic threshold and the duration is within the range of 0.1–0.5 seconds, it is marked as a candidate event for road defect.

[0052] In some embodiments, the method for assessing road surface defects and creating maps further includes: Upon receiving abnormal event data uploaded by the terminals of subsequent vehicles, the system verifies the geographical coordinates of the road surface defects and their lanes determined based on the abnormal event data uploaded by the terminals of the preceding vehicles, and generates a verification report. The road map containing pavement defect information is updated based on the verification report.

[0053] Understandably, two types of typical driving behaviors are defined. The first is avoidance behavior: when a vehicle approaches a suspected defect area, there is no significant impact on the Z-axis, but the Y-axis acceleration shows a short-term significant peak, indicating that the vehicle bypasses the defect by changing lanes laterally and does not actually run over it. The second is running over behavior: when a vehicle drives directly over a suspected defect area, the Z-axis acceleration shows a sharp pulse (lasting 0.1–0.5 seconds), while the Y-axis acceleration remains stable, indicating that the wheels actually contact and run over the road surface defect.

[0054] By using the differences in the aforementioned sensing characteristics, it is possible to effectively distinguish whether a vehicle has actually come into contact with a defect.

[0055] Cross-validation mechanism: If both avoidance behavior records and crushing behavior records exist within the same event cluster formed by spatial clustering, they constitute strong complementary evidence. Avoidance behavior reflects the driver's subjective intention to avoid, while crushing behavior provides objective physical impact signals. The co-occurrence of the two significantly increases the confidence that there is a real road defect at that location and helps to eliminate false alarms caused by sensor noise or non-defect disturbances (such as speed bumps or manhole covers passing normally).

[0056] Only clusters with a confidence level ≥ 0.7 are retained as valid pavement defect events.

[0057] In some embodiments, based on abnormal event data uploaded by subsequent vehicles, the geographic coordinates of road defects and their lanes determined based on abnormal event data uploaded by previous vehicle terminals are verified, and a verification report is generated, including: If it is determined that a subsequent vehicle has evasive behavior or has run over a road surface defect based on abnormal event data uploaded by subsequent vehicles, the confidence level of the geographical coordinates of the road surface defect and the lane it is located in, determined based on the abnormal event data uploaded by the terminal of the preceding vehicle, is increased, and a corresponding verification report is generated.

[0058] Understandably, if the co-occurrence of avoidance and compaction behaviors continues to be observed at a location already identified as a road defect, the reliability of the defect at that location is further reinforced; conversely, if there are no compaction signals at that location for an extended period, the reliability is reduced, and the defect information may even be deleted. This mechanism enables dynamic verification of defect events and continuous updating and optimization of the defect map.

[0059] Enhanced verification: During an enhanced monitoring period (set to 24 hours) following the initial identification, if new reported events continue to be received at this location, the following rules will apply: If a new event contains both "crushing behavior" and "avoidance behavior" patterns, it is considered a strong verification signal. The platform will increase the overall confidence score of this defect and update its defect level based on the latest standardized acceleration mean. Simultaneously, in map rendering, its risk warning color can be maintained or deepened.

[0060] If a new incident involves only "crushing behavior" without "avoidance behavior," it is still considered valid evidence of the defect and can be used to slightly increase the confidence level of that point.

[0061] If the new event only involves "avoidance behavior," it serves as supplementary evidence to maintain the confidence level of that point.

[0062] Decay and Clear: The system is configured with a state decay period (set to 72 hours). Within the period: If the defect does not receive any new "rolling behavior" signals, its confidence level will decay linearly or exponentially over time.

[0063] At the same time, the system checks whether there is also a lack of new "avoidance behavior" signals at that point. If the "avoidance behavior" signals continue to be missing, it indicates that the driver may no longer perceive a defect at that location that needs to be avoided.

[0064] When the confidence level of a point decays to below the preset clearing threshold and exceeds the decay period, the platform automatically marks its status as "cleared" or "presumed to be repaired".

[0065] In some embodiments, this invention proposes a crowdsourced method for road defect assessment and mapping based on vibration characteristics of group vehicles, aiming to address the prominent problems of existing road defect detection technologies in terms of reliability, cost, timeliness, and accuracy, specifically including: (1) Visual recognition methods are heavily dependent on lighting conditions, and their performance degrades significantly in complex environments such as night, rain, snow, fog, or obstruction.

[0066] (2) Detection schemes that rely on specialized equipment (such as ground-penetrating radar and laser scanners) are costly and complex to operate, making it difficult to achieve large-scale and high-frequency monitoring.

[0067] (3) Single sensor or isolated event analysis is susceptible to noise interference, resulting in a high false alarm rate and insufficient positioning accuracy.

[0068] (4) The traditional “fixed-point and regular” inspection mode has a long update cycle and cannot support the creation of real-time road surface defect maps.

[0069] To overcome the above problems, this embodiment provides the following technical solution: A crowdsourced method for assessing road defects and creating maps based on vibration characteristics of group vehicles comprises two main parts: (1) a navigation APP installed on the user's mobile phone (as a sensing terminal, i.e., a vehicle terminal), and (2) a platform-side data analysis system deployed on a cloud server; the two work together through the Internet to form a complete closed loop of "sensing-assessment-map creation".

[0070] The system specifically includes the following steps: 1. Mobile device (navigation app) data collection and anomaly event capture: S201. After the user opens the navigation APP, the system continuously collects raw data from the phone's built-in accelerometer, gyroscope and GPS module, and obtains information on road surface material type and weather conditions through map and meteorological interfaces.

[0071] S202. Based on the complementary filtering algorithm, multi-sensor data is fused to complete the attitude estimation of the mobile phone relative to the vehicle coordinate system on the APP, and the acceleration data is uniformly converted to the vehicle coordinate system with the vehicle's forward direction as the X-axis, the right side as the Y-axis, and the vertical upward direction as the Z-axis.

[0072] S203. Monitor the Z-axis acceleration signal in real time, and construct an environmentally adaptive dynamic threshold by combining vehicle speed, road material type and weather conditions: if the Z-axis acceleration exceeds the threshold, it is initially determined to be a candidate event for road defect.

[0073] S204. Simultaneously monitor the Y-axis acceleration. If a short-term, large lateral acceleration occurs without a steering command, it is directly determined as an emergency avoidance event.

[0074] S205. When an abnormal vibration event on the Z-axis or an emergency avoidance event on the Y-axis is detected, the APP records data such as timestamp, acceleration segment, GPS location, driving direction, vehicle speed, and abnormality type identifier (defect / avoidance), and uploads it to the platform in a structured and intelligent manner.

[0075] 2. Platform-side (cloud-based) analysis and defect map creation: S206. The platform receives abnormal events reported from a large number of mobile terminals. First, it preliminarily distinguishes between raised or sunken road surface defects based on the polarity (positive / negative) of the Z-axis acceleration for candidate defect events.

[0076] S207. For abnormal events in the same geographical area and time range (e.g., within a 10-meter radius or within 2 hours) and in the same driving direction, use a spatiotemporal clustering algorithm to generate event clusters.

[0077] S208. Calculate the comprehensive confidence score for each event cluster, taking into account the consistency of features within the cluster, the historical reliability of the reporting equipment, and the matching degree of environmental factors. Use Y-axis avoidance behavior data to assist in verification, and retain only valid defect event clusters with a confidence score > 0.7.

[0078] S209. Introduce an environmental adaptive grading mechanism. For each valid defect event cluster, standardize the corresponding Z-axis acceleration peak value based on the vehicle speed and weather conditions in each reported data within the cluster.

[0079] Based on the mean value of the peak Z-axis acceleration after standardization within the cluster, and combined with the type of road material (asphalt / cement, etc.), the defect level (mild / moderate / severe) is dynamically classified.

[0080] S2010. The assessment results are stored in a structured manner, including: lane, risk level, geographical coordinates, road marker, driving direction, and defect type (concave / convex). Based on this, map matching is performed to generate a road defect map that is updated in real time every 5–15 minutes, with color gradients mapping the risk level.

[0081] S2011. If the co-occurrence of avoidance and compaction behaviors continues to be observed at a location identified as a pavement defect, the reliability of the defect at that location is further strengthened; if there are no compaction signals at that location for a long period of time, the reliability is reduced or the defect information may even be deleted. This mechanism enables dynamic verification of defect events and continuous updating and optimization of the defect map.

[0082] Preferably, the sensor data includes triaxial acceleration values, triaxial angular velocity values, and GPS location information.

[0083] Preferably, the vehicle coordinate system is defined as follows: the Z-axis is perpendicular to the ground and pointing upwards, the X-axis is along the vehicle's forward direction, and the Y-axis is perpendicular to the X-axis and points to the right.

[0084] Preferably, the attitude calculation is performed using a complementary filtering algorithm based on the three-axis acceleration values, three-axis angular velocity values, and GPS data, and then the three-axis acceleration data in the mobile phone coordinate system is converted into data in the vehicle coordinate system.

[0085] Preferably, the dynamic threshold increases with vehicle speed and changes accordingly in combination with weather conditions and road material type to adapt to vibration characteristics under different driving conditions.

[0086] Preferably, the platform utilizes Y-axis obstacle avoidance behavior data to achieve dual functions of auxiliary defect verification and lane-level positioning: when a vehicle approaches a suspected defect area, if an obstacle avoidance maneuver is performed, a significant peak in Y-axis acceleration occurs while there is no obvious impact on the Z-axis, indicating that the defect was not actually run over; if the vehicle passes directly through, a sharp pulse occurs on the Z-axis while the Y-axis remains stable. Based on this, the platform distinguishes between obstacle avoidance behavior and running over behavior, and can cross-verify the existence of defects.

[0087] Specifically, during the cloud-based data analysis platform's assessment phase, the platform aggregates abnormal event data reported by multiple mobile navigation apps and identifies two typical response patterns within a pre-defined geofence area: Avoidance behavior pattern: The driver actively changes lanes to avoid the defect before approaching it, and there is a significant peak in Y-axis acceleration (symbols indicate left / right lane change direction), while there is no significant impact in Z-axis acceleration; Crushing behavior pattern: The vehicle drives directly over the defect, and a sharp pulse appears in the Z-axis acceleration, while the Y-axis acceleration does not change significantly.

[0088] Preferably, the spatiotemporal clustering algorithm performs spatial proximity judgment based on latitude and longitude, and temporal proximity judgment based on the timestamps of defect candidate events.

[0089] Preferably, the color mapping rule of the road defect map is as follows: red represents high-risk / dense defects, yellow represents medium-risk defects, and green represents low-risk defects or no abnormalities.

[0090] Compared with the prior art, the present invention has the following advantages: (1) Achieve dynamic monitoring of road surface defects with wide coverage across the entire road network. This invention relies on a large number of users' mobile phones as mobile sensing terminals. Without the need to deploy dedicated detection equipment or rely on fixed-point inspections, it can achieve high-frequency, wide-range, and continuous defect monitoring of various road networks such as urban roads and highways, truly achieving real-time sensing capabilities with "full network coverage".

[0091] (2) It has all-weather and all-environment applicability. This invention does not rely on images or optical sensors, but is based on multi-axis acceleration signals caused by vehicle vibration for identification. It effectively overcomes the interference of complex environments such as night, rain, snow, fog, haze, and obstruction on traditional vision methods. It can operate stably under various weather and lighting conditions, significantly improving the robustness of the system.

[0092] (3) A defect classification mechanism is implemented by introducing three environmental factors: vehicle speed, weather, and road material type. In the data analysis platform assessment stage, this invention first standardizes the peak Z-axis acceleration for each reported data point within each valid event cluster based on its corresponding vehicle speed and weather conditions. This eliminates vibration amplitude deviations caused by differences in driving speed or weather conditions (such as slippery conditions or snow accumulation), thereby obtaining a comparable standardized impact intensity. Subsequently, based on the average of the standardized Z-axis acceleration peak values ​​within the cluster, and combined with the road material type (such as asphalt or cement) of the road section, the defect level judgment threshold is dynamically adjusted, ensuring that the same impact intensity corresponds to a reasonable risk level under different road conditions. This mechanism ensures that the classification results reflect both the physical impact nature and the actual impact on driving safety.

[0093] (4) A method for creating and updating road surface defect maps in real time is provided. Real-time updating of road surface defect maps can improve the accuracy of road surface defect information. At the same time, the created maps can not only be used by navigation apps to realize early warning of road surface defects and improve driving safety, but also be provided to road maintenance departments for road surface defect repair, which has great social and economic benefits.

[0094] (6) Extremely low deployment cost and strong compatibility. It makes full use of common sensors such as accelerometers, gyroscopes and GPS in users' existing mobile phones. There is no need to purchase additional professional equipment or modify vehicles. Ordinary car owners can participate in road defect monitoring after installing the APP, which greatly reduces the threshold for system promotion and the total social cost.

[0095] (7) Adopting a mobile-platform collaborative computing architecture improves efficiency and reduces resource consumption. The mobile device completes attitude calculation, coordinate system transformation and initial screening of defect candidate events locally, and only uploads key fragment data to the platform, which greatly reduces invalid data transmission and platform computing power burden, ensuring real-time performance and further controlling system operating costs, and supporting large-scale concurrent processing capabilities.

[0096] In summary, the main idea of ​​this invention is to construct a method for assessing and creating road surface defects, consisting of a navigation app and a cloud-based data analysis platform. The first step involves using a mobile app to collect vehicle vibration, attitude, positioning, and environmental data, performing coordinate system transformation and initial screening of abnormal events locally, and uploading structured event fragments. The second step involves integrating multi-vehicle reported data on the platform, and through spatiotemporal clustering, environmental adaptive standardization, and grading, achieving type identification, existence verification, and risk rating of road surface defects, thereby enabling map matching, creation of road surface defect maps, and real-time updates.

[0097] In some embodiments, a crowdsourced method and medium for assessing and mapping road defects based on the vibration characteristics of group vehicles, such as... Figure 2 As shown, it includes the following steps: ① Mobile data collection and preliminary identification: S301. After the user opens the navigation APP described in this invention, the system continuously collects raw data from the phone's built-in sensors at a sampling frequency of 10Hz, including: the three-axis linear acceleration output by the accelerometer (unit: m / s²); the three-axis angular velocity output by the gyroscope (unit: rad / s); the position, speed, and heading information output by the GPS module; at the same time, it obtains the road surface material type (such as asphalt, cement) through the map navigation API and the current weather conditions (such as sunny, rainy, snowy) through the meteorological service API.

[0098] S302. On the mobile phone, a complementary filtering algorithm is used to fuse three-axis acceleration and three-axis angular velocity data, and combined with the motion direction provided by GPS, to calculate the attitude angle of the mobile phone relative to the vehicle, and to uniformly convert the acceleration data to the vehicle coordinate system: where the X-axis is along the direction of vehicle movement, the Y-axis points to the right, and the Z-axis is perpendicular to the ground and upward.

[0099] S303. Real-time monitoring of the Z-axis acceleration signal in the vehicle coordinate system, and construction of an environment-adaptive dynamic threshold: This threshold is dynamically adjusted according to real-time vehicle speed, road material type, and weather conditions—for example, the higher the vehicle speed, the larger the threshold; different road materials correspond to different baseline settings; and the threshold can also be adaptively corrected under special weather conditions such as rain and snow. When the peak value of the Z-axis acceleration exceeds the dynamic threshold, it is determined as a candidate event for road defect.

[0100] S304. Synchronous monitoring of Y-axis acceleration signal: If a short-term large lateral acceleration occurs on the Y-axis without a steering command, it is determined to be an emergency avoidance event.

[0101] S305. When an abnormal vibration event on the Z-axis or an emergency avoidance event on the Y-axis is detected, the APP generates a structured abnormal event record, which includes the following fields: unique device ID, timestamp, GPS latitude and longitude, driving direction, vehicle speed, road surface material type, weather conditions, abnormality type identifier (defect / avoidance), and corresponding Z-axis and Y-axis acceleration data fragments; the record is encapsulated in JSON format and intelligently selects the upload time according to the current network status (such as Wi-Fi priority, mobile network traffic throttling), transmitting only structured metadata and necessary fragments, avoiding continuous uploading of raw sensor stream data.

[0102] ② Cloud-based data analysis platform for assessment, road defect map creation and real-time updates: S306. The platform receives abnormal events reported from a large number of mobile terminals and first distinguishes between raised or sunken road surface defects based on the polarity (positive / negative) of the Z-axis acceleration.

[0103] S307. For abnormal events in the same geographical area (within a radius of 10 meters) and the same time range (within 2 hours) with the same driving direction, use the improved DBSCAN algorithm for spatiotemporal clustering, and set the minimum number of cluster samples to 2.

[0104] S308. Calculate a comprehensive confidence score (maximum 1.0) for each event cluster, taking into account the following factors: (1) Characteristic consistency of Z / Y axis acceleration signals within the cluster (e.g., standard deviation of peak distribution); (2) The historical reliability of the reporting equipment (based on the accuracy of its past feedback); (3) Environmental factor matching degree (such as whether the weather and road surface material types in the various reported records of the same event are consistent); (4) Utilizing Y-axis avoidance behavior to achieve auxiliary defect verification, specifically including: (i) Behavioral pattern discrimination: Two types of typical driving behaviors are defined. The first is avoidance behavior: when the vehicle approaches the suspected defect area, there is no significant impact on the Z-axis, but the Y-axis acceleration shows a short-term significant peak, indicating that the vehicle bypasses the defect by changing lanes laterally and does not actually run over it. The second is running over behavior: when the vehicle drives directly over the suspected defect area, the Z-axis acceleration shows a sharp pulse (lasting 0.1–0.5 seconds), while the Y-axis acceleration remains stable, indicating that the wheel actually contacts and runs over the road surface defect.

[0105] By using the differences in the aforementioned sensing characteristics, it is possible to effectively distinguish whether a vehicle has actually come into contact with a defect.

[0106] (ii) Cross-validation mechanism: If both avoidance behavior records and crushing behavior records exist within the same event cluster formed by spatial clustering, they constitute strong complementary evidence. Avoidance behavior reflects the driver's subjective intention to avoid, while crushing behavior provides objective physical impact signals. The co-occurrence of the two significantly increases the confidence that there is a real road defect at that location and helps to eliminate false alarms caused by sensor noise or non-defect disturbances (such as speed bumps or manhole covers passing normally).

[0107] Only clusters with a confidence level ≥ 0.7 are retained as valid pavement defect events.

[0108] S309. For each valid event cluster, implement an environment-adaptive hierarchical mechanism, comprising two phases: (1) Standardization stage: For each reported data in the cluster, the original Z-axis acceleration peak value is standardized according to its corresponding vehicle speed and weather conditions.

[0109] (2) Grading stage: Calculate the mean value of the peak value of the Z-axis acceleration after standardization within the cluster, and dynamically adjust the defect level judgment rule in combination with the pavement material type of the road section: when the mean value is less than 5 m / s², it is identified as a minor defect; when the mean value is not less than 5 m / s² and not greater than 8 m / s², it is identified as a moderate defect; when the mean value is greater than 8 m / s², it is identified as a severe defect.

[0110] S3010. The assessment results are stored in a structured manner, including the defect level, geographical coordinates, station number, driving direction, and defect type (concave / convex). Based on this, a road defect map is generated that is updated every 5–15 minutes. The risk level is represented by a color gradient on the electronic map: red indicates severe defects (high risk), orange indicates moderate defects (medium risk), and yellow indicates minor defects (low risk).

[0111] S3011. If the co-occurrence of avoidance and compaction behaviors continues to be observed at a location identified as a pavement defect, the reliability of the defect at that location is further strengthened; if there are no compaction signals at that location for a long period of time, the reliability is reduced or the defect information may even be deleted. This mechanism enables dynamic verification of defect events and continuous updating and optimization of the defect map.

[0112] (1) Enhanced verification: During an enhanced monitoring period (set to 24 hours) following the initial identification, if new reported events are continuously received at this point, they shall be handled according to the following rules: If a new event contains both "crushing behavior" and "avoidance behavior" patterns, it is considered a strong verification signal. The platform will increase the overall confidence score of this defect and update its defect level based on the latest standardized acceleration mean. Simultaneously, in map rendering, its risk warning color can be maintained or deepened.

[0113] If a new incident involves only "crushing behavior" without "avoidance behavior," it is still considered valid evidence of the defect and can be used to slightly increase the confidence level of that point.

[0114] If the new event only involves "avoidance behavior," it serves as supplementary evidence to maintain the confidence level of that point.

[0115] (2) Decay and Clearance: The system is set with a state decay period (set to 72 hours). Within the period: If the defect does not receive any new "rolling behavior" signals, its confidence level will decay linearly or exponentially over time.

[0116] At the same time, the system checks whether there is also a lack of new "avoidance behavior" signals at that point. If the "avoidance behavior" signals continue to be missing, it indicates that the driver may no longer perceive a defect at that location that needs to be avoided.

[0117] When the confidence level of a point decays to below the preset clearing threshold and exceeds the decay period, the platform automatically marks its status as "cleared" or "presumed to be repaired".

[0118] (3) Map updates and feedback: The platform dynamically updates the road defect map display based on the real-time status (active / decaying / cleared) and confidence level of the defect points. For example, "active" high-confidence points are highlighted in red, while "decaying" points can be made semi-transparent or gray.

[0119] Defects whose status changes to "cleared" will be removed from the real-time road defect map, and their historical records can be archived for querying.

[0120] This dynamic map updates every 5-15 minutes, synchronized with the analysis cycle, ensuring that users always receive a map that reflects the latest road conditions.

[0121] like Figure 3 As shown, the present invention also provides a road surface defect assessment and map making device 300, comprising: Clustering module 301 is used to cluster abnormal event data uploaded by multiple vehicle terminals within the target area to obtain abnormal event clusters and the confidence level of abnormal event clusters. The defect identification module 302 is used to determine the road defect level based on the vehicle speed and weather data corresponding to the abnormal event clusters with confidence levels greater than a preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface, and the road material type. The defect location module 303 is used to locate the location of the emergency avoidance behavior based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, and to determine the geographic coordinates of the road defect and the lane in which it is located. The map drawing module 304 is used to match the road surface defect type, the road surface defect level, the geographical coordinates of the road surface defect and the lane where it is located with a preset road map to generate a road map containing road surface defect information. The abnormal event data is obtained by the vehicle terminal based on the following steps: Based on vehicle speed, vehicle acceleration, road surface data, and weather data, anomaly type identifiers are determined; the anomaly type identifiers include: road surface defect type and vehicle avoidance data. Abnormal event data is constructed based on the vehicle's geographic coordinates, vehicle speed, vehicle acceleration, road surface data, weather data, and anomaly type identifier.

[0122] The road surface defect assessment and map making device provided in the above embodiments can realize the technical solutions described in the above road surface defect assessment and map making method embodiments. The specific implementation principles of each module or unit can be found in the corresponding content in the above road surface defect assessment and map making method embodiments, which will not be repeated here.

[0123] like Figure 4 As shown, the present invention also provides an electronic device 400. The electronic device 400 includes a processor 401, a memory 402, and a display 403. Figure 4 Only some components of the electronic device 400 are shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead.

[0124] In some embodiments, memory 402 may be an internal storage unit of electronic device 400, such as a hard disk or memory of electronic device 400. In other embodiments, memory 402 may also be an external storage device of electronic device 400, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on electronic device 400.

[0125] Furthermore, the memory 402 may include both internal storage units of the electronic device 400 and external storage devices. The memory 402 is used to store application software and various types of data installed on the electronic device 400.

[0126] In some embodiments, processor 401 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 402 or process data, such as the road surface defect assessment and map making method of the present invention.

[0127] In some embodiments, display 403 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 403 is used to display information from electronic device 400 and to display a visual user interface. Components 401-403 of electronic device 400 communicate with each other via a system bus.

[0128] In some embodiments of the present invention, when the processor 401 executes the road surface defect assessment and map creation program in the memory 402, the following steps can be implemented: Clustering is performed on abnormal event data uploaded by multiple vehicle terminals within the target area to obtain abnormal event clusters and their confidence levels. Based on the vehicle speed and weather data corresponding to the clusters of abnormal events with a confidence level greater than the preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface, and the road material type, the road defect level is determined. Based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, the location of the emergency avoidance behavior is located, and the geographic coordinates of the road defect and the lane in which it is located are determined. Based on the type of road surface defect, the level of the road surface defect, the geographical coordinates of the road surface defect and the lane where it is located, a road map containing road surface defect information is generated by matching it with a preset road map. The abnormal event data is obtained by the vehicle terminal based on the following steps: Based on vehicle speed, vehicle acceleration, road surface data, and weather data, anomaly type identifiers are determined; the anomaly type identifiers include: road surface defect type and vehicle avoidance data. Abnormal event data is constructed based on the vehicle's geographic coordinates, vehicle speed, vehicle acceleration, road surface data, weather data, and anomaly type identifier.

[0129] It should be understood that when the processor 401 executes the road defect assessment and map creation program in the memory 402, in addition to the functions mentioned above, it can also perform other functions, as can be found in the description of the corresponding method embodiments above.

[0130] Furthermore, the embodiments of the present invention do not specifically limit the type of electronic device 400 mentioned. Electronic device 400 can be a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, laptop computer, or other portable electronic device. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic device can also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, electronic device 400 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0131] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the road surface defect assessment and map making methods provided by the above methods, the method comprising: Clustering is performed on abnormal event data uploaded by multiple vehicle terminals within the target area to obtain abnormal event clusters and their confidence levels. Based on the vehicle speed and weather data corresponding to the clusters of abnormal events with a confidence level greater than the preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface, and the road material type, the road defect level is determined. Based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, the location of the emergency avoidance behavior is located, and the geographic coordinates of the road defect and the lane in which it is located are determined. Based on the type of road surface defect, the level of the road surface defect, the geographical coordinates of the road surface defect and the lane where it is located, a road map containing road surface defect information is generated by matching it with a preset road map. The abnormal event data is obtained by the vehicle terminal based on the following steps: Based on vehicle speed, vehicle acceleration, road surface data, and weather data, anomaly type identifiers are determined; the anomaly type identifiers include: road surface defect type and vehicle avoidance data. Abnormal event data is constructed based on the vehicle's geographic coordinates, vehicle speed, vehicle acceleration, road surface data, weather data, and anomaly type identifier.

[0132] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0133] The road surface defect assessment and mapping method, device, electronic equipment, and medium provided by the present invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for assessing road surface defects and creating maps, characterized in that, include: Clustering is performed on abnormal event data uploaded by multiple vehicle terminals within the target area to obtain abnormal event clusters and their confidence levels. Based on the vehicle speed and weather data corresponding to the clusters of abnormal events with a confidence level greater than the preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface, and the road material type, the road defect level is determined. Based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, the location of the emergency avoidance behavior is located, and the geographic coordinates of the road defect and the lane in which it is located are determined. Based on the type of road surface defect, the level of the road surface defect, the geographical coordinates of the road surface defect and the lane where it is located, a road map containing road surface defect information is generated by matching it with a preset road map. The abnormal event data is obtained by the vehicle terminal based on the following steps: Based on vehicle speed, vehicle acceleration, road surface data, and weather data, anomaly type identifiers are determined; the anomaly type identifiers include: road surface defect type and vehicle avoidance data. Abnormal event data is constructed based on the vehicle's geographic coordinates, vehicle speed, vehicle acceleration, road surface data, weather data, and anomaly type identifier.

2. The method for road surface defect assessment and map production according to claim 1, characterized in that, The vehicle acceleration is obtained by the vehicle terminal based on the following steps: Collect vehicle terminal acceleration data and vehicle terminal motion direction data; Based on the vehicle terminal acceleration data and the vehicle terminal motion direction data, the relative pose relationship between the vehicle terminal and the vehicle is determined. Based on the relative pose relationship, the vehicle terminal acceleration data is converted into vehicle acceleration data.

3. The method for road surface defect assessment and map production according to claim 1, characterized in that, Based on vehicle speed and weather data corresponding to anomalous event clusters with confidence levels greater than a preset confidence threshold, peak vehicle acceleration in the vertical direction of the road surface, and road material type, the road defect level is determined, including: For the vehicle speed and weather data corresponding to the clusters of abnormal events with a confidence level greater than a preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface is standardized; the peak vehicle acceleration in the vertical direction of the road surface is extracted based on the abnormal event data. The road defect level is determined based on the average of the standardized peak vehicle acceleration and the type of road material.

4. The method for road surface defect assessment and map production according to claim 1, characterized in that, Based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, the location of the emergency avoidance behavior is determined, including the geographic coordinates of the road defect and its lane. Based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, the location of the emergency avoidance behavior is located to obtain the location of the emergency avoidance behavior. Based on the location of the emergency avoidance action and the direction of the vehicle's lane change, the geographical coordinates of the road defect and its lane are determined.

5. The method for road surface defect assessment and map production according to claim 1, characterized in that, Based on vehicle speed, vehicle acceleration, road surface data, and weather data, anomaly type identifiers are determined, including: Based on the vehicle speed, the road surface material type, and the weather data, an environmental adaptive dynamic threshold is constructed. Based on the relationship between the changes in vehicle acceleration in the vertical direction of the road surface and the magnitude of the adaptive dynamic threshold of the environment, an anomaly type identifier is determined.

6. The method for assessing road surface defects and creating maps according to any one of claims 1-5, characterized in that, Also includes: Upon receiving abnormal event data uploaded by the terminals of subsequent vehicles, the system verifies the geographical coordinates of the road surface defects and their lanes determined based on the abnormal event data uploaded by the terminals of the preceding vehicles, and generates a verification report. The road map containing pavement defect information is updated based on the verification report.

7. The method for road surface defect assessment and map production according to claim 6, characterized in that, Based on the abnormal event data uploaded by subsequent vehicles, the geographical coordinates and lane locations of road defects determined from the abnormal event data uploaded by the terminals of previous vehicles are verified, and a verification report is generated, including: If it is determined that a subsequent vehicle has evasive behavior or has run over a road surface defect based on abnormal event data uploaded by subsequent vehicles, the confidence level of the geographical coordinates of the road surface defect and the lane it is located in, determined based on the abnormal event data uploaded by the terminal of the preceding vehicle, is increased, and a corresponding verification report is generated.

8. A device for assessing road surface defects and creating maps, characterized in that, include: The clustering module is used to cluster abnormal event data uploaded by multiple vehicle terminals within the target area to obtain abnormal event clusters and their confidence levels. The defect identification module is used to determine the road defect level based on the vehicle speed and weather data corresponding to the abnormal event clusters with a confidence level greater than a preset confidence threshold, the peak vehicle acceleration in the vertical direction of the road surface, and the road material type. The defect location module is used to locate the location of the emergency avoidance behavior based on the vehicle avoidance data and vehicle geographic coordinates in the abnormal event data, and to determine the geographic coordinates of the road defect and the lane in which it is located. The map drawing module is used to match the road surface defect type, the road surface defect level, the geographical coordinates of the road surface defect and the lane it is located with a preset road map to generate a road map containing road surface defect information. The abnormal event data is obtained by the vehicle terminal based on the following steps: Based on vehicle speed, vehicle acceleration, road surface data, and weather data, anomaly type identifiers are determined; the anomaly type identifiers include: road surface defect type and vehicle avoidance data. Abnormal event data is constructed based on the vehicle's geographic coordinates, vehicle speed, vehicle acceleration, road surface data, weather data, and anomaly type identifier.

9. An electronic device, characterized in that, Including memory and processor, among which, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps of the road surface defect assessment and map making method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the road surface defect assessment and map making method as described in any one of claims 1 to 7.

Images