Road maintenance decision-making method and system

Abstract

Provided in the present disclosure are a road maintenance decision-making method and system. The method comprises: determining a PCI of a target road; determining a preliminary decision-making level for maintenance of the target road on the basis of the PCI; controlling a road inspection device to acquire inspection data of the target road; controlling a workstation to determine road damage data on the basis of the inspection data, wherein the workstation is a road damage identification and treatment platform based on a road damage model; determining a deep decision-making level for maintenance of the target road on the basis of the road damage data; and on the basis of the preliminary decision-making level for maintenance of the target road and the deep decision-making level for maintenance of the target road, determining a maintenance decision-making mode for the target road. In the present disclosure, road damage is automatically detected on the basis of a road inspection device and a workstation, such that a higher efficiency and a higher level of accuracy are achieved, and manpower resources and costs can be saved on; and a maintenance decision-making mode for a target road is determined on the basis of a preliminary decision-making level and a deep decision-making level, such that rational decision-making with regard to a road maintenance mode is achieved, so as to maintain road safety and comfort.

Description

Road maintenance decision-making method and system

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese patent application number 2024100224487, filed with the China Patent Office on January 8, 2024, entitled “A Road Maintenance Decision-Making Method and System,” the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to the technical field of road maintenance, and in particular to a road maintenance decision-making method and system. Background Art

[0004] With the rapid development and improvement of road construction, road mileage has continued to increase, and road traffic volume has gradually increased in recent years. This, coupled with overloaded roads, has directly led to an increasing number of road diseases, which have a significant impact on road performance. This not only directly affects the overall performance and lifespan of highways, but also directly affects driving safety, comfort, and economy. Therefore, road maintenance is a vital link in maintaining road quality.

[0005] The contradiction between traditional road maintenance and management methods and the current large-scale maintenance needs is becoming increasingly acute. Road inspection and repair work still primarily relies on inspectors using written descriptions of damage locations. This lacks a robust road repair inspection model and consumes significant time and human resources. Due to the extensive and complex scope of road repair projects, manual inspections are subject to significant subjectivity, and different inspectors may produce inconsistent results, complicating decision-making for road repairs.

[0006] With the advancement of road maintenance decision-making concepts, traditional manual inspection and decision-making methods have been unable to keep up with the tide of the times. Therefore, a reasonable road maintenance decision-making method is needed to prevent decision-makers from making blind maintenance decisions.

[0007] Summary of the Invention

[0008] The present disclosure provides a road maintenance decision-making method and system to address the low efficiency and accuracy of manual detection decision-making methods in the existing technology. The present disclosure performs automated detection of road defects based on road detection equipment and workstations, which has higher efficiency and accuracy, can save human resources and costs, and determines the target road maintenance decision model based on the preliminary decision level and the deep decision level, so as to achieve reasonable decision-making on road maintenance methods and maintain road safety and comfort.

[0009] The present disclosure provides a road maintenance decision-making method, comprising: determining the PCI of a target road; determining a preliminary decision level for target road maintenance based on the PCI; controlling a road detection device to acquire detection data of the target road; controlling a workstation to determine road damage data based on the detection data; the workstation being a road damage identification and processing platform based on a road damage model; determining a deep decision level for target road maintenance based on the road damage data; and determining a target road maintenance decision mode based on the preliminary decision level and the deep decision level for target road maintenance.

[0010] According to a road maintenance decision-making method provided by the present disclosure, controlling the road detection equipment to obtain detection data of the target road includes: controlling the road detection equipment to perform target road detection in a full inspection manner to obtain all image data of the target road.

[0011] According to a road maintenance decision-making method provided by the present disclosure, before the control workstation determines the road disease data based on the detection data, it also includes: controlling a temporary storage to store the detection data so as to import the detection data into the workstation.

[0012] According to a road maintenance decision-making method provided by the present disclosure, controlling a temporary memory to store the detection data includes: controlling the temporary memory to store all the image data; the temporary memory is a flash drive or a solid-state memory.

[0013] According to a road maintenance decision-making method provided by the present disclosure, the control workstation determines road disease data based on the detection data, including: controlling the workstation to analyze and identify all the image data based on a pre-trained deep learning neural network model to obtain characteristic information of road diseases; and determining the road benign repair rate and the new road disease rate based on the characteristic information.

[0014] According to a road maintenance decision-making method provided by the present disclosure, the characteristic information of the road damage is the information of the damage type and damage location annotated on the image data based on the road damage dataset; the damage types include irregular repairs, missed repairs, secondary development of damage after repairs, and new damages.

[0015] According to a road maintenance decision-making method provided by the present disclosure, the method of determining the road benign repair rate and the road disease new addition rate based on the characteristic information includes: determining the road benign repair rate based on a first preset formula according to the characteristic information; the first preset formula is: α=A / C

[0016] Among them, α is the benign repair rate of the road, A is the number of repairs that are repaired according to regulations and without secondary development of the disease after repair, and C is the total number of diseases in the last inspection;

[0017] According to the characteristic information, the new rate of road damage is determined based on a second preset formula; the second preset formula is: β=B / C

[0018] Among them, β is the rate of new road defects, B is the number of new defects, and C is the total number of defects in the last inspection.

[0019] According to a road maintenance decision-making method provided by the present disclosure, the preliminary decision level for target road maintenance is determined based on the PCI, including: the preliminary decision level is divided into five levels, excellent PCI ≥ 92, good PCI 80-92, medium PCI 70-80, inferior PCI 60-70, and poor PCI < 60.

[0020] According to a road maintenance decision-making method provided by the present disclosure, the depth decision level of target road maintenance is determined based on the road disease data, including: the road benign repair rate is divided into four levels, the excellent level δ I Road benign repair rate ≥ 90%, good grade δ II Road benign repair rate 80%-90%, intermediate δ III Road benign repair rate 60%-80%, poor grade δ IV The road benign repair rate is less than 60%; the new rate of road damage is divided into four levels, the excellent level is ε I The new rate of road damage is less than 10%, and the road is good II New rate of road damage 10%-20%, intermediate ε III New rate of road damage 20%-50%, poor level ε IV The new rate of road damage is greater than 50%. The depth decision level is divided into four levels. The depth decision level I includes δ I and ε I , δ I and ε II , δ II and ε I , δ II and ε II ; Depth decision level II includes δ III and ε I , δ III and ε II , δ IV and ε I , δ IV and ε II ; Depth decision level III includes δ I and ε III , δ I and ε IV , δ II and ε III , δII and ε IV ; Depth decision level IV includes δ III and ε III , δ III and ε IV , δ IV and ε IV , and ε III .

[0021] According to a road maintenance decision method provided by the present disclosure, the target road maintenance decision mode is determined according to the preliminary decision level of the target road maintenance and the deep decision level of the target road maintenance, including: the target road maintenance decision mode is divided into six modes, mode one corresponds to the preliminary decision level excellent PCI, good PCI and deep decision level I; mode two corresponds to the preliminary decision level excellent PCI, good PCI and deep decision levels II, III and IV; mode three corresponds to the preliminary decision level intermediate PCI and deep decision level I; mode four corresponds to The preliminary decision level should be intermediate PCI and the deep decision levels should be II, III and IV; mode five corresponds to the preliminary decision level of secondary PCI, poor PCI and deep decision level I; mode six corresponds to the preliminary decision level of secondary PCI, poor PCI and deep decision levels II, III and IV; mode one is daily inspection; mode two is daily inspection and special spot check; mode three is repair; mode four is repair and special rectification; mode five is cover sealing and milling; mode six is ​​cover sealing, milling, expert diagnosis and special rectification.

[0022] The present disclosure also provides a road maintenance decision system, including: a PCI determination module for determining the PCI of a target road; a preliminary decision level determination module for determining a preliminary decision level for target road maintenance based on the PCI; a detection equipment control module for controlling a road detection device to obtain detection data of a target road; a workstation control module for controlling a workstation to determine road damage data based on the detection data; the workstation is a road damage identification and processing platform based on a road damage model; a depth decision level determination module for determining a depth decision level for target road maintenance based on the road damage data; and a decision mode determination module for determining a target road maintenance decision mode based on the preliminary decision level and the depth decision level of the target road maintenance.

[0023] According to a road maintenance decision-making system provided by the present disclosure, it also includes a storage control module, which is used to control a temporary memory to store the detection data so as to import the detection data into the workstation.

[0024] The present disclosure provides a road maintenance decision-making method and system, which includes determining the PCI of a target road; determining a preliminary decision level for target road maintenance based on the PCI; controlling road inspection equipment to acquire inspection data for the target road; controlling a workstation to determine road damage data based on the inspection data; the workstation being a road damage identification and processing platform based on a road damage model; determining a deep decision level for target road maintenance based on the road damage data; and determining a target road maintenance decision model based on the preliminary decision level and the deep decision level of the target road maintenance. The present disclosure automates road damage detection based on road inspection equipment and a workstation, achieving higher efficiency and accuracy, saving human resources and costs. Furthermore, the system determines a target road maintenance decision model based on the preliminary decision level and the deep decision level, enabling rational decision-making on road maintenance methods to maintain road safety and comfort. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the present disclosure or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0026] FIG1 is a schematic diagram of a process flow of a road maintenance decision-making method provided by the present disclosure;

[0027] FIG2 is a schematic structural diagram of a road maintenance decision-making system provided by the present disclosure. DETAILED DESCRIPTION

[0028] To make the objectives, technical solutions, and advantages of this disclosure more clear, the technical solutions of this disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of this disclosure, not all of them. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of this disclosure without creative effort shall fall within the scope of protection of this disclosure.

[0029] In recent years, the total amount of road infrastructure has increased significantly, playing a vital role in socioeconomic development. This has not only continuously improved road transportation efficiency and capacity, but also significantly increased highway mileage. However, this presents an unavoidable problem: under the influence of various factors, including vehicle loads, natural factors, and human factors, roads gradually develop various damage, deformation, and other defects. Common defects include cracks, potholes, rutting, looseness, subsidence, and surface damage. For example, cracks in roads can undermine the integrity and continuity of the road structure and serve as a natural channel for surface water to intrude into the roadbed and road structure. If not promptly addressed, cracks can further extend, forming cracks that can cause road delamination, resulting in structural damage and compromising driving comfort and safety. Therefore, road defect detection, a key and challenging aspect of road maintenance, is crucial for determining road maintenance strategies.

[0030] Routine road inspections lack professional testing equipment, and defects, especially road damage, are usually determined through manual observation. Engineering quantity measurement and acceptance are also often determined through manual measurement. There are shortcomings such as trivial work, high work intensity, long time consumption, high labor costs, low efficiency, subjective misjudgment due to human factors, poor repeatability, limited accuracy, large management loopholes, serious safety hazards, and difficulty in meeting the needs of information timeliness.

[0031] Please refer to FIG1 , which is a flow chart of a road maintenance decision-making method provided by the present disclosure.

[0032] In order to solve the technical problems existing in the prior art, the present disclosure provides a road maintenance decision-making method, comprising:

[0033] 101: Determine the PCI of the target road;

[0034] 102: Determine the preliminary decision level for target road maintenance based on PCI;

[0035] 103: Control the road detection device to obtain detection data of the target road;

[0036] 104: The control workstation determines road damage data based on the detection data; the workstation is a road damage identification and processing platform based on the road damage model;

[0037] 105: Determine the depth decision level of target road maintenance based on road disease data;

[0038] 106: Determine a target road maintenance decision mode according to a preliminary decision level of target road maintenance and a deep decision level of target road maintenance.

[0039] Specifically, the PCI (Pavement Condition Index) of the target road is first determined. This can be determined by conducting pavement defect inspections with road inspection equipment or by other methods. A preliminary decision level for maintenance of the target road is then determined based on the PCI. Digital road information is collected by road inspection equipment to obtain inspection data for the target road, which can be image data. The road inspection equipment can be mounted on intelligent road inspection equipment. The inspection data can then be stored in a temporary memory and then imported into a workstation. Alternatively, the workstation can directly read the inspection data from the road inspection equipment. Based on the inspection data and a road defect model, the workstation identifies road defects and determines road defect data, such as the benign repair rate and the rate of new road defects. The road defect data can be categorized according to the severity of the defect. The in-depth decision level for maintenance of the target road is then determined based on the road defect data. The target road maintenance decision model is then determined based on the preliminary decision level and the in-depth decision level of the target road maintenance. This disclosure is conducive to analyzing the repair effect of the target road section, the occurrence of defects and the time period of recurrence, thereby helping road management departments and maintenance agencies to accurately grasp the maintenance effect and provide a basis for road maintenance management units to select the best maintenance plan.

[0040] Based on the above embodiment:

[0041] As an optional embodiment, controlling the road detection device to obtain detection data of the target road includes: controlling the road detection device to perform target road detection in a full inspection manner to obtain all image data of the target road.

[0042] Specifically, a portable road inspection device can be mounted on a vehicle and driven at an appropriate speed along the road to be inspected. The device can include a camera. It is important to note that the camera array on the vehicle is preferably arranged along the road's cross-section, allowing simultaneous imaging of a single cross-section, minimizing distortion caused by camera distortion. The road inspection device can perform a full inspection of the target road section, obtaining complete image data of the target road, saving time and human resources while achieving high efficiency and accuracy.

[0043] As an optional embodiment, before controlling the workstation to determine the road damage data based on the detection data, the method further includes: controlling a temporary memory to store the detection data so as to import the detection data into the workstation.

[0044] As an optional embodiment, controlling the temporary memory to store the detection data includes: controlling the temporary memory to store all image data; the temporary memory is a flash memory drive or a solid-state memory.

[0045] In this embodiment, a temporary storage device can be used to store the detection data. The temporary storage device is in communication with the road detection device and can store the road detection data collected by the road detection device in real time. The road image data in the temporary storage device can then be imported into the workstation. It is worth noting that the temporary storage device can be a built-in flash drive or solid-state memory, such as an eMMC (Embedded Multi Media Card), a UFS (Universal Flash Storage), or an SSD (Solid-State Disk).

[0046] As an optional embodiment, the control workstation determines road damage data based on the detection data, including: the control workstation analyzes and identifies all image data based on a pre-trained deep learning neural network model to obtain characteristic information of road damage; and determines the road benign repair rate and the new road damage rate based on the characteristic information.

[0047] As an optional embodiment, the characteristic information of road damage is the information of the damage type and damage location annotated on the image data based on the road damage dataset; the damage types include irregular repairs, missed repairs, secondary development of damage after repairs, and new damages.

[0048] As an optional embodiment, determining the road benign repair rate and the road disease new rate according to the characteristic information includes: determining the road benign repair rate based on the characteristic information based on a first preset formula; the first preset formula is: α=A / C

[0049] Among them, α is the benign repair rate of the road, A is the number of repairs that are repaired according to regulations and without secondary development of the disease after repair, and C is the total number of diseases in the last inspection;

[0050] According to the characteristic information, the new rate of road damage is determined based on the second preset formula; the second preset formula is: β=B / C

[0051] Among them, β is the rate of new road defects, B is the number of new defects, and C is the total number of defects in the last inspection.

[0052] In this embodiment, the workstation is an interactive road hazard identification and processing software platform based on a deep learning neural network equipped with computer processing hardware. Optionally, the deep learning neural network can be YOLO or other existing deep learning models.

[0053] The workstation automatically generates processed images of road damage and counts irregular repairs, missed repairs, secondary damage development after repairs, and new damage. The workstation then calculates the benign road repair rate and the new road damage rate based on the first and second preset formulas. It's worth noting that the calculations of both the benign road repair rate and the new road damage rate require data from the previous inspection. Therefore, using portable road inspection equipment for inspections can effectively reduce inspection costs while maintaining inspection frequency.

[0054] Among them, irregular repair means that the length and area of ​​the repair do not meet the highway technical condition assessment standards; missed repair means that no repair traces are detected compared with the last inspection; secondary development of defects after repair means that cracks, potholes and other defects are detected on the repair traces; new defects refer to the detection of defects that did not appear in the last inspection.

[0055] As an optional embodiment, the preliminary decision level for target road maintenance is determined based on PCI, including: the preliminary decision level is divided into five levels, excellent PCI>92, good PCI 80-92, medium PCI 70-80, inferior PCI 60-70, and poor PCI<60.

[0056] PCI is an important indicator of road surface damage and is closely related to vehicle safety and comfort. PCI is a standardized metric for evaluating and describing road surface conditions, measuring pavement quality, damage severity, and maintenance needs. PCI is typically expressed on a scale of 0 to 100, with higher values ​​representing better pavement conditions and lower values ​​indicating worse pavement conditions. Table 1 shows the preliminary decision levels for target road maintenance based on PCI.

[0057] Table 1 Preliminary decision-making level table

[0058] As an optional embodiment, determining the depth decision level of target road maintenance based on road damage data includes:

[0059] The road benign repair rate is divided into four levels, the excellent level δ I Road benign repair rate ≥ 90%, good grade δ II Road benign repair rate 80%-90%, intermediate δ III Road benign repair rate 60%-80%, poor grade δ IV The road benign repair rate is less than 60%;

[0060] The rate of new road damage is divided into four levels, the excellent level is ε I The new rate of road damage is less than 10%, and the road is good II New rate of road damage 10%-20%, intermediate ε IIINew rate of road damage 20%-50%, poor level ε IV The rate of new road damage is greater than 50%;

[0061] The depth decision level is divided into four levels. The depth decision level I includes δ I and ε I , δ I and ε II , δ II and ε I , δ II and ε II ; Depth decision level II includes δ III and ε I , δ III and ε II , δ IV and ε I , δ IV and ε II ; Depth decision level III includes δ I and ε III , δ I and ε IV , δ II and ε III , δ II and ε IV ; Depth decision level IV includes δ III and ε III , δ III and ε IV , δ IV and ε IV , and ε III .

[0062] The depth decision level of target road maintenance determined based on road disease data is shown in Table 2.

[0063] Table 2 Depth decision level table

[0064] As an optional embodiment, the target road maintenance decision mode is determined according to the preliminary decision level of the target road maintenance and the depth decision level of the target road maintenance, including: the target road maintenance decision mode is divided into six modes, mode one corresponds to the preliminary decision level excellent PCI, good PCI and depth decision level I; mode two corresponds to the preliminary decision level excellent PCI, good PCI and depth decision levels II, III and IV; mode three corresponds to the preliminary decision level intermediate PCI and depth decision level I; mode four corresponds to the preliminary decision level intermediate PCI and depth decision levels II, III and IV; mode five corresponds to the preliminary decision level secondary PCI, poor PCI and depth decision level I; mode six corresponds to the preliminary decision level secondary PCI, poor PCI and depth decision levels II, III and IV; mode one is daily inspection; mode two is daily inspection and special spot check; mode three is repair; mode four is repair and special rectification; mode five is cover sealing and milling; mode six is ​​cover sealing, milling, expert diagnosis and special rectification.

[0065] The target road maintenance decision model table is shown in Table 3.

[0066] Table 3 Target road maintenance decision model

[0067] The target road maintenance decision model scheme is shown in Table 4.

[0068] Table 4 Target road maintenance decision-making model plan

[0069] Among them, the daily inspection decision library includes: Road surface flatness: Check whether the road surface is flat and whether there are any bumps or potholes. Cracks and damage: Check whether the road surface has cracks, cracks or damage. Markings and signs: Check whether the markings and signs on the road surface are clear and visible, and whether they are worn or missing. Drainage system: Check whether the road surface drainage system is unobstructed and whether there is any water accumulation or leakage. Shoulders and guardrails: Check whether the shoulders and guardrails are intact and whether there is any damage or looseness. Vegetation and debris: Check whether the vegetation and debris around the road surface are cleared and whether there is any impact on traffic safety.

[0070] The special inspection decision database includes: Traffic volume: Traffic volume monitoring; Traffic pressure: Monitoring large and medium-sized trucks and buses, and overweight monitoring; Roadbed conditions: Testing roadbed bearing capacity; Environmental factors: Investigation of environmental factors such as chemicals, salt, and pollutants; Repair material types and quality: Investigation of repair material types and quality; Repair measures and methods: Evaluation of repair measures; Repair construction quality: Investigation of construction quality.

[0071] The repair decision library includes: Hole Filling: Repairing potholes in the road, typically using asphalt concrete or other pavement repair materials. Crack Sealing: Sealing cracks in the road, typically using specialized crack sealants or other sealing materials. Pothole Filling: Repairing depressions or damaged areas in the road, typically requiring cleaning the pothole and filling it with appropriate pavement repair materials. Asphalt Leak Repairing: Repairing asphalt leaks in the road, typically requiring cleaning the leaked area and resurfacing.

[0072] The special rectification decision database includes: Traffic flow: Traffic flow control; Traffic pressure: Truck and bus overload monitoring; Roadbed conditions: Roadbed bearing capacity repair; Environmental factors: Road surface site deep cleaning; Repair material type and quality: Repair material improvement; Repair measures and methods: Repair measures rectification; Repair construction quality: Construction quality supervision.

[0073] It is important to note that the grading standards can be adjusted and customized according to specific circumstances. Different regions, different road types, and different traffic volumes may require different standards to adjust and optimize the evaluation of pavement repair effects.

[0074] In addition, the indicators can be combined using the weight distribution method to comprehensively evaluate the road repair effect, and the weighted average method can be used to combine the two indicators of road benign repair rate and road disease new increase rate.

[0075] First, we need to determine the weights for the road benign repair rate and the road deterioration rate in the comprehensive evaluation. These weights can be assigned based on actual needs and importance. For example, if the road benign repair rate is more critical to the effectiveness of pavement maintenance and repair work, it can be given a higher weight.

[0076] Then, the road benign repair rate and road disease new rate are normalized and converted into proportional values ​​between 0 and 1. This ensures that the two indicators have equal importance in the comprehensive evaluation.

[0077] Then, based on the determined weights, a weighted average of the normalized road benign repair rate and road disease new rate is calculated. The weighted average can be obtained by multiplying the value of each indicator by its corresponding weight and adding the results.

[0078] Finally, a comprehensive evaluation value of the pavement repair effect can be obtained based on the weighted average results. According to specific evaluation criteria, the evaluation value can be divided into different levels, such as excellent, good, fair, or poor, to more intuitively understand the pavement repair effect.

[0079] Of course, seasonal preventive maintenance can also be adopted for road maintenance. For example, the preventive maintenance measures in spring are to deal with temperature shrinkage cracks and other cracks, and use low-temperature spring rain maintenance materials to deal with slurry overflow, looseness and other diseases; the preventive maintenance measures in summer are to deal with oil overflow problems, eliminate waves, bulges and other problems, and repair damage temporarily repaired in winter and spring; the preventive maintenance measures in autumn are winter preventive maintenance treatments, such as emulsified asphalt suction seal, frost heave prevention, crack sealing, etc.; the preventive maintenance measures in winter are road snow and ice prevention treatment, and road maintenance material procurement.

[0080] The road maintenance decision system provided by the present disclosure is described below. The road maintenance decision system described below and the road maintenance method described above can be referenced to each other.

[0081] Please refer to FIG2 , which is a schematic diagram of the structure of a road maintenance decision-making system provided by the present disclosure.

[0082] The present disclosure also provides a road maintenance decision system, including: a PCI determination module 201, configured to determine the PCI of a target road; a preliminary decision level determination module 202, configured to determine a preliminary decision level for target road maintenance based on the PCI; a detection equipment control module 203, configured to control a road detection device to obtain detection data of the target road; a workstation control module 204, configured to control a workstation to determine road damage data based on the detection data; the workstation is a road damage identification and processing platform based on a road damage model; a depth decision level determination module 205, configured to determine a depth decision level for target road maintenance based on the road damage data; and a decision mode determination module 206, configured to determine a target road maintenance decision mode based on the preliminary decision level and the depth decision level of the target road maintenance.

[0083] In the embodiment of the present disclosure, by collecting road images and performing road damage analysis on the road images, the road maintenance decision level can be automatically obtained and the corresponding maintenance decision measures can be output.

[0084] As an optional embodiment, a storage control module is further included, and the storage control module is configured to control the temporary memory to store the detection data so as to import the detection data into the workstation.

[0085] The PCI determination module 201 is configured to determine the PCI of the target road. The PCI of the target road can be determined by performing road surface disease inspections with road detection equipment, or by other methods.

[0086] The preliminary decision level determination module 202 is configured to determine the preliminary decision level of the target road maintenance according to the PCI. The preliminary decision level is divided into five levels, namely, excellent PCI ≥ 92, good PCI 80-92, intermediate PCI 70-80, inferior PCI 60-70, and poor PCI < 60. PCI is an important indicator for identifying the damage condition of the road surface and is closely related to the safety and comfort of vehicle driving. Among them, PCI is a standardized indicator for evaluating and describing the condition of the road surface, which is configured to measure the quality, degree of damage and maintenance requirements of the road surface. PCI is usually expressed as a score from 0 to 100, with a higher value representing a better road condition and a lower value representing a worse road condition. The preliminary decision level for determining the target road maintenance according to PCI is shown in Table 1.

[0087] The detection equipment control module 203 can acquire continuous road surface images. This module is equipped with appropriate road detection equipment, such as cameras or sensors, which are mobile and mounted on inspection vehicles. To ensure high-quality road image data, this module includes image quality control functions, including automatic exposure adjustment, white balance calibration, and image clarity assessment, ensuring sufficient clarity and accuracy of captured images.

[0088] The disclosed road maintenance decision-making system also includes a storage control module. The collected image data is stored on a local storage device. This module also provides data management functions, such as data indexing, querying, and backup, to effectively manage and retrieve image data. A built-in flash drive or solid-state memory, optionally an embedded MultiMediaCard (eMMC), Universal Flash Storage (UFS), or solid-state drive (SSD), is included. In particular, road image data stored in the built-in memory should be promptly transmitted to the server for processing to prevent it from losing its reference value due to long-term inactivity.

[0089] The workstation control module 204 can perform image analysis on the collected road damage image data. First, the collected road damage images undergo preprocessing, such as image noise removal, contrast enhancement, and brightness and color balance adjustment, to improve image quality and clarity. Image analysis then extracts feature information related to road damage, including damage classification and location.

[0090] The workstation control module 204 can also classify the damage images by comparing and matching them with pre-trained damage recognition models. These images can be identified and classified into: standard repairs with no secondary damage progression, continued damage progression after repair, missed repairs, and newly added damages. The workstation control module 204 also counts the number of each type of damage and generates a damage report.

[0091] The workstation control module 204 can also calculate the road benign repair rate and the road disease new addition rate. The road benign repair rate is obtained according to a first preset formula, and the road disease new addition rate is obtained according to a second preset formula.

[0092] The depth decision level determination module 205 is configured to determine the depth decision level of the target road maintenance according to the road disease data. The depth decision level of the target road maintenance determined according to the road disease data is shown in Table 2.

[0093] The road benign repair rate is divided into four levels, the excellent level δ I Road benign repair rate ≥ 90%, good grade δ II Road benign repair rate 80%-90%, intermediate δ III Road benign repair rate 60%-80%, poor grade δ IV The road benign repair rate is less than 60%;

[0094] The rate of new road damage is divided into four levels, the excellent level is ε I The new rate of road damage is less than 10%, and the road is good II New rate of road damage 10%-20%, intermediate ε III New rate of road damage 20%-50%, poor level ε IV The rate of new road damage is greater than 50%;

[0095] The depth decision level is divided into four levels. The depth decision level I includes δ I and ε I , δ I and ε II , δ II and ε I , δ II and ε II ; Depth decision level II includes δ III and ε I , δ III and ε II , δ IV and ε I , δ IV and ε II ; Depth decision level III includes δ I and ε III , δ I and ε IV , δ II and ε III , δ II and ε IV ; Depth decision level IV includes δ III and ε III , δ III and ε IV , δ IV and εIV , and ε III .

[0096] The decision mode determination module 206 is configured to determine the target road maintenance decision mode according to the preliminary decision level of the target road maintenance and the in-depth decision level of the target road maintenance.

[0097] The target road maintenance decision-making model is divided into six modes: Mode 1 corresponds to the preliminary decision levels of Excellent PCI, Good PCI, and Deep Decision Level I; Mode 2 corresponds to the preliminary decision levels of Excellent PCI, Good PCI, and Deep Decision Levels II, III, and IV; Mode 3 corresponds to the preliminary decision level of Intermediate PCI and Deep Decision Level I; Mode 4 corresponds to the preliminary decision levels of Intermediate PCI and Deep Decision Levels II, III, and IV; Mode 5 corresponds to the preliminary decision levels of Substandard PCI, Poor PCI, and Deep Decision Level I; and Mode 6 corresponds to the preliminary decision levels of Substandard PCI, Poor PCI, and Deep Decision Levels II, III, and IV. Mode 1 is routine inspection; Mode 2 is routine inspection and special spot checks; Mode 3 is repair; Mode 4 is repair and special rectification; Mode 5 is overcoating and milling; Mode 6 is overcoating, milling, expert diagnosis, and special rectification. The target road maintenance decision-making model table is shown in Table 3. The target road maintenance decision-making model scheme is shown in Table 4.

[0098] Among them, the daily inspection decision library includes: Road surface flatness: Check whether the road surface is flat and whether there are any bumps or potholes. Cracks and damage: Check whether the road surface has cracks, cracks or damage. Markings and signs: Check whether the markings and signs on the road surface are clear and visible, and whether they are worn or missing. Drainage system: Check whether the road surface drainage system is unobstructed and whether there is any water accumulation or leakage. Shoulders and guardrails: Check whether the shoulders and guardrails are intact and whether there is any damage or looseness. Vegetation and debris: Check whether the vegetation and debris around the road surface are cleared and whether there is any impact on traffic safety.

[0099] The special inspection decision database includes: Traffic volume: Traffic volume monitoring; Traffic pressure: Monitoring large and medium-sized trucks and buses, and overweight monitoring; Roadbed conditions: Testing roadbed bearing capacity; Environmental factors: Investigation of environmental factors such as chemicals, salt, and pollutants; Repair material types and quality: Investigation of repair material types and quality; Repair measures and methods: Evaluation of repair measures; Repair construction quality: Investigation of construction quality.

[0100] The repair decision library includes: Hole Filling: Repairing potholes in the road, typically using asphalt concrete or other pavement repair materials. Crack Sealing: Sealing cracks in the road, typically using specialized crack sealants or other sealing materials. Pothole Filling: Repairing depressions or damaged areas in the road, typically requiring cleaning the pothole and filling it with appropriate pavement repair materials. Asphalt Leak Repairing: Repairing asphalt leaks in the road, typically requiring cleaning the leaked area and resurfacing.

[0101] The special rectification decision database includes: Traffic flow: Traffic flow control; Traffic pressure: Truck and bus overload monitoring; Roadbed conditions: Roadbed bearing capacity repair; Environmental factors: Road surface site deep cleaning; Repair material type and quality: Repair material improvement; Repair measures and methods: Repair measures rectification; Repair construction quality: Construction quality supervision.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present disclosure. Industrial Applicability

[0103] The above solution can be used to automatically detect road defects based on road detection equipment and workstations, with higher efficiency and accuracy, saving human resources and costs. It can also determine the target road maintenance decision model based on the preliminary decision level and the deep decision level, and make reasonable decisions on road maintenance methods to maintain road safety and comfort.

Claims

1. A road maintenance decision-making method, characterized in that, Including: Determine the PCI of the target road; Determine the preliminary decision level for the maintenance of the target road according to the PCI; Control the road detection equipment to obtain the detection data of the target road; Control the workstation to determine the road disease data according to the detection data; the workstation is a road disease identification and processing platform based on a road disease model; Determine the in-depth decision level for the maintenance of the target road according to the road disease data; Determine the maintenance decision mode of the target road according to the preliminary decision level for the maintenance of the target road and the in-depth decision level for the maintenance of the target road.

2. The road maintenance decision-making method according to claim 1, characterized in that The controlling the road detection equipment to obtain the detection data of the target road includes: Control the road detection equipment to detect the target road in a full inspection mode to obtain all the image data of the target road.

3. The road maintenance decision-making method according to claim 2, wherein, Before the controlling the workstation to determine the road disease data according to the detection data, it further includes: Control the temporary memory to store the detection data so as to import the detection data into the workstation.

4. The road maintenance decision-making method according to claim 3, wherein The controlling the temporary memory to store the detection data includes: Control the temporary memory to store all the image data; the temporary memory is a flash drive or a solid-state memory.

5. The road maintenance decision-making method according to any one of claims 2 to 4, characterized in that The controlling the workstation to determine the road disease data according to the detection data includes: Control the workstation to analyze and identify all the image data based on a pre-trained deep learning neural network model to obtain the characteristic information of the road disease; Determine the road good repair rate and the road disease new increase rate according to the characteristic information.

6. The road maintenance decision-making method according to claim 5, characterized in that The characteristic information of the road disease is the information of the disease type and the disease location marked on the image data based on a road disease data set; The disease types include irregular repair, missed repair, secondary development of diseases after repair, and new diseases.

7. The road maintenance decision-making method according to claim 6, characterized in that, The determining the road good repair rate and the road disease new increase rate according to the characteristic information includes: Determine the road good repair rate based on a first preset formula according to the characteristic information; The first preset formula is: α = A / C Where α is the road good repair rate, A is the number of repairs that are standardized and have no secondary development of diseases after repair, and C is the total number of diseases detected in the previous inspection; Determine the road disease new increase rate based on a second preset formula according to the characteristic information; The second preset formula is: β = B / C Where β is the road disease new increase rate, B is the number of new diseases, and C is the total number of diseases detected in the previous inspection.

8. The road maintenance decision-making method according to claim 5, wherein, The determining the preliminary decision level for the maintenance of the target road according to the PCI includes: The preliminary decision level is divided into five levels, excellent level PCI ≥ 92, good level PCI 80 - 92, medium level PCI 70 - 80, secondary level PCI 60 - 70, poor level PCI < 60.

9. The road maintenance decision-making method according to claim 8, characterized in that The determining the in-depth decision level for the maintenance of the target road according to the road disease data includes: The road's good repair rate is divided into four grades, excellent grade δ I The road's good repair rate ≥ 90%, good grade δ II The road's good repair rate 80% - 90%, medium grade δ III The road's good repair rate 60% - 80%, poor grade δ IV The road's good repair rate < 60%; The newly increased rate of road diseases is divided into four grades, excellent grade ε I The newly increased rate of road diseases is less than 10%, good grade ε II The newly increased rate of road diseases is 10% - 20%, medium grade ε III The newly increased rate of road diseases is 20% - 50%, poor grade ε IV The newly increased rate of road diseases > 50%; The described depth decision levels are divided into four levels. Depth decision level I includes δ I and ε I 、δ I and ε II 、δ II and ε I 、δ II and ε II ; Depth decision level II includes δ III and ε I 、δ III and ε II 、δ IV and ε I 、δ IV and ε II ; Depth decision level III includes δ I and ε III 、δ I and ε IV 、δ II and ε III 、δ II and ε IV ; Depth decision level IV includes δ III and ε III 、δ III and ε IV 、δ IV and ε IV 、δ IV and ε III .

10. The road maintenance decision-making method according to claim 9, characterized in that, The determining the maintenance decision mode of the target road according to the preliminary decision level for the maintenance of the target road and the in-depth decision level for the maintenance of the target road includes: The described target road maintenance decision-making mode is divided into six modes. Mode 1 corresponds to the preliminary decision-making level of excellent PCI, good PCI, and the in-depth decision-making level of level I; Mode 2 corresponds to the preliminary decision-making level of excellent PCI, good PCI, and the in-depth decision-making levels of level II, level III, and level IV; Mode 3 corresponds to the preliminary decision-making level of medium PCI and the in-depth decision-making level of level I; Mode 4 corresponds to the preliminary decision-making level of medium PCI and the in-depth decision-making levels of level II, level III, and level IV; Mode 5 corresponds to the preliminary decision-making level of secondary PCI, poor PCI, and the in-depth decision-making level of level I; Mode 6 corresponds to the preliminary decision-making level of secondary PCI, poor PCI, and the in-depth decision-making levels of level II, level III, and level IV; The described Mode 1 is daily inspection; Mode 2 is daily inspection and special sampling inspection; Mode 3 is patching; Mode 4 is patching and special rectification; Mode 5 is surface sealing and milling; Mode 6 is surface sealing, milling, expert diagnosis, and special rectification.

11. A road maintenance decision-making system, characterized in that, It includes: A PCI determination module configured to determine the PCI of the target road; A preliminary decision-making level determination module configured to determine the preliminary decision-making level of target road maintenance according to the PCI; A detection equipment control module configured to control road detection equipment to obtain detection data of the target road; A workstation control module configured to control the workstation to determine road disease data according to the detection data; the workstation is a road disease identification and processing platform based on a road disease model; An in-depth decision-making level determination module configured to determine the in-depth decision-making level of target road maintenance according to the road disease data; A decision-making mode determination module configured to determine the target road maintenance decision-making mode according to the preliminary decision-making level of target road maintenance and the in-depth decision-making level of target road maintenance.

12. The road maintenance decision-making system according to claim 11, wherein It further includes a storage control module configured to control a temporary memory to store the detection data so as to import the detection data into the workstation.

Images