A method for dynamic collaborative inspection of urban roads using multiple devices

Abstract

This invention discloses a dynamic collaborative inspection method for multiple devices on urban roads, comprising: S1, identifying road surface damage features and generating event trigger signals through a wide-area perception layer; S2, analyzing the event trigger signals in real time through a data intelligence decision layer to generate scheduling instructions containing location coordinates and damage types; S3, a dynamic execution layer executing vehicle lane detection and driving vibration feature detection, and pedestrian / sidewalk detection, through the scheduling instructions; S4, performing time-domain-spatial correlation analysis on image features and vibration waveforms through the data intelligence decision layer to generate detection results; S5, adjusting the inspection path and operating parameters of each device based on the detection results of each terminal, forming a closed-loop feedback. According to this invention, the efficiency and quality of road inspection are greatly improved, resource allocation is optimized, and road quality problems can be discovered and resolved more promptly and comprehensively.

Description

Technical Field

[0001] This invention relates to the technical field of road surface quality detection and automatic information collection, and in particular to a method for dynamic collaborative inspection of urban roads using multiple devices. Background Technology

[0002] With the rapid development of urban transportation systems, road hazards have become a key factor affecting urban traffic safety and efficiency. Traditional methods for identifying road hazards mainly rely on manual patrols, which are not only time-consuming and labor-intensive but also inefficient, making it difficult to achieve comprehensive monitoring of the vast urban road network. Therefore, the need for efficient and automated methods for identifying road hazards has become an urgent requirement for improving urban road safety management. Moreover, existing technologies suffer from problems such as independent operation of equipment, lack of coordination leading to unreasonable resource allocation, and poor efficiency and effectiveness. This invention designs an event-triggered task collaboration mechanism. Summary of the Invention

[0003] To address the shortcomings of existing technologies, the present invention aims to provide a dynamic collaborative inspection method for urban roads using multiple devices, which significantly improves the efficiency and quality of road inspections, optimizes resource allocation, and enables more timely and comprehensive detection and resolution of road quality issues. To achieve the above-mentioned objectives and other advantages of the present invention, a dynamic collaborative inspection method for urban roads using multiple devices is provided, comprising:

[0004] S1. Identify road surface damage features and generate event triggering signals through the wide-area perception layer;

[0005] S2. The event trigger signal is analyzed in real time by the data intelligence decision layer to generate a scheduling instruction containing the location coordinates and damage type;

[0006] S3. The dynamic execution layer executes the motor vehicle lane detection through the scheduling command, collects and analyzes road surface image data and driving vibration characteristics, and executes the sidewalk detection through the scheduling command, collects and analyzes sidewalk pavement surface image data and motion vibration waveform;

[0007] S4. The intelligent decision layer performs time-space correlation analysis on image features and vibration waveforms to generate detection results.

[0008] S5, the intelligent decision-making layer adjusts the inspection path and working parameters of each device based on the detection results of each terminal, forming a closed-loop feedback.

[0009] Preferably, the wide-area perception layer in step S1 includes a drone, which is equipped with a visible light / infrared composite optical sensor to conduct wide-area patrols, identify road damage features, and generate event trigger signals.

[0010] Preferably, the pavement damage characteristics include damage types and evaluation parameters. The damage types include pavement cracks, potholes, network cracks, repair potholes, repair network cracks, and repair cracks. The evaluation parameters include length / area evaluation parameters.

[0011] Preferably, the dynamic execution layer includes a patrol vehicle and a robotic dog, wherein the patrol vehicle performs lane detection based on the dispatch command, and collects and analyzes road image data and driving vibration characteristics;

[0012] The robotic dog performs sidewalk detection based on the scheduling instructions, and collects and analyzes sidewalk paving surface image data and motion vibration waveforms.

[0013] Preferably, the patrol vehicle is equipped with a multi-axis accelerometer array, and the vibration characteristic analysis includes the detection of the exposure duration and magnitude of abnormal vibrations;

[0014] The mechanical dog's limbs are equipped with vibration sensors that collect vibration waveforms from the contact of paving bricks to determine the looseness of the paving bricks.

[0015] Preferably, the step of the mechanical dog detecting the looseness of paving bricks specifically includes:

[0016] 1) Perform gait traversal in the target area and collect vibration signals along both the X and Y axes;

[0017] 2) Construct a model for the attenuation of vibration energy in the brick joint area;

[0018] 3) Locate the three-dimensional coordinates of the loose brick based on the phase abrupt change point of the attenuation model.

[0019] Preferably, the intelligent decision-making layer includes a cloud control center, which has a built-in data fusion module. The data fusion module performs spatiotemporal correlation analysis on the damage distribution detected by the UAV, the vibration time sequence characteristics of the patrol vehicle, and the location data of the loose bricks of the mechanical dog.

[0020] Preferably, the cloud control center includes a damage and deterioration trend prediction module, which is used to analyze the potential loss of driving / walking comfort caused by the detected defects and, in conjunction with historical data, the damage and deterioration trend at that location, thereby generating a composite inspection report that includes repair priorities.

[0021] Compared with existing technologies, the beneficial effects of this invention are as follows: The event-triggered task collaboration mechanism of this invention brings many significant advantages. First, it greatly improves the efficiency of urban road patrols. Through the collaborative work of various terminals, duplicate detection and resource waste are avoided, enabling the completion of patrol tasks over a larger area in a shorter time. Second, it significantly improves the accuracy and comprehensiveness of detection. The combination of large-scale initial screening by drones and precise detection by patrol vehicles and robotic sensors ensures that no potential road problems are overlooked, providing a reliable basis for timely repair and maintenance. Third, it effectively reduces road maintenance costs. Early detection and resolution of problems reduce further deterioration of road damage, thereby reducing the cost of large-scale repairs later. Furthermore, it enhances the flexibility and adaptability of road patrol work. It can dynamically adjust the tasks and collaboration methods of various terminals according to different road conditions and needs, adapting to various complex urban road environments. Finally, it improves the safety of urban roads. Timely detection and treatment of road defects reduce the risk of traffic accidents caused by road quality issues, ensuring the safety of citizens' travel. Attached Figure Description

[0022] Figure 1 This is a schematic diagram illustrating the system architecture and equipment coordination of the multi-device dynamic collaborative inspection method for urban roads according to the present invention;

[0023] Figure 2 The orthogonal gait X / Y alternating movement path of the mechanical dog in the multi-device dynamic collaborative patrol method for urban roads according to the present invention;

[0024] Figure 3 This is an example of vibration data from a mechanical dog in the multi-device dynamic collaborative patrol method for urban roads according to 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 some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Reference Figure 1 A method for dynamic collaborative inspection of urban roads using multiple devices, comprising:

[0027] S1. Identify road surface damage features and generate event trigger signals through a wide-area perception layer; the wide-area perception layer in step S1 includes a drone, which is equipped with a visible light / infrared composite optical sensor to conduct wide-area patrols, identify road surface damage features, and generate event trigger signals; the road surface damage features include damage types and evaluation parameters, the damage types include road surface cracks, potholes, network cracks, repair potholes, repair network cracks, and repair cracks, and the evaluation parameters include length / area evaluation parameters. The drone performs high-altitude patrols along a planned route, identifies road damage types and evaluation parameters through real-time image analysis, and generates trigger signals that include geographical location and preliminary damage assessment.

[0028] S2. The event trigger signal is analyzed in real time by the data intelligence decision layer to generate a scheduling instruction containing the location coordinates and damage type;

[0029] S3. The dynamic execution layer performs lane detection based on the scheduling instructions, collects and analyzes road surface image data and driving vibration characteristics, and performs sidewalk detection based on the scheduling instructions, collects and analyzes sidewalk pavement surface image data and motion vibration waveforms. The dynamic execution layer includes a patrol vehicle and a robotic dog. The patrol vehicle performs lane detection based on the scheduling instructions, collects and analyzes road surface image data and driving vibration characteristics. The patrol vehicle is equipped with a multi-axis accelerometer array, and the vibration characteristic analysis includes the detection of the exposure duration and magnitude of abnormal vibrations.

[0030] The robotic dog performs sidewalk detection based on the scheduling instructions, collecting and analyzing image data and motion vibration waveforms of the sidewalk paving surface. Vibration sensors mounted on the robotic dog's limbs collect contact vibration waveforms of the paving bricks to determine the looseness of the paving bricks. The specific steps for the robotic dog to detect the looseness of the paving bricks include:

[0031] 1) Perform gait traversal in the target area and collect vibration signals along both the X and Y axes;

[0032] 2) Construct a model for the attenuation of vibration energy in the brick joint area;

[0033] 3) Locate the three-dimensional coordinates of the loose brick based on the phase abrupt change point of the attenuation model.

[0034] Furthermore, the method for detecting pedestrian pavement using the mechanical dog is as follows:

[0035] 1. Gait control and data acquisition:

[0036] Orthogonal gait traversal calculation based on grid map is employed, as follows: Figure 2 This ensures 100% coverage of the testing area;

[0037] By using an embedded vibration detection array at the foot, z-axis vibration waveforms are acquired at 5-second intervals at the moment of ground contact, such as... Figure 3 As shown;

[0038] 2. Vibration signal modeling and localization:

[0039] After wavelet denoising of the original signal, short-time Fourier transform is used to extract frequency domain features;

[0040] Construct a vibration energy decay model:

[0041] 3. Loosening positioning and 3D mapping

[0042] Abnormal displacement of brick joints is identified by a phase change point detection algorithm.

[0043] S4. The intelligent decision layer performs time-space correlation analysis on image features and vibration waveforms to generate detection results.

[0044] S5. The intelligent decision-making layer adjusts the inspection path and operating parameters of each device based on the detection results of each terminal, forming a closed-loop feedback. The intelligent decision-making layer includes a dynamic programming module, which executes the following optimization strategies:

[0045] (1) Density adjustment of the subsequent loitering path of the UAV;

[0046] (2) Matching the patrol vehicle's detection speed with the sensor's operating mode;

[0047] (3) Dynamic balance between mechanical dog gait parameters and detection accuracy.

[0048] Furthermore, the intelligent decision-making layer includes a cloud control center, which has a built-in data fusion module. This module performs spatiotemporal correlation analysis on the damage distribution detected by the UAV, the vibration time-series characteristics of the patrol vehicle, and the location data of the loose bricks from the mechanical dog. The terminal control center has a built-in multi-source data fusion engine.

[0049] Furthermore, the cloud control center includes a damage and deterioration trend prediction module, which is used to analyze the potential loss of driving / walking comfort caused by the detected defects and, in conjunction with historical data, the damage and deterioration trend at that location, thereby generating a composite inspection report that includes repair priorities.

[0050] Furthermore, intelligent collaboration and dynamic optimization are detailed as follows:

[0051] 1. Multi-device task scheduling:

[0052] Detection equipment is allocated according to the type of damage: when a drone triggers a crack of 2cm or more, the patrol vehicle is dispatched first; when it triggers a misalignment of the joint, the robot dog is dispatched.

[0053] Establish an equipment status monitoring dashboard to display parameters such as workload, remaining power, and testing progress in real time;

[0054] 2. Closed-loop feedback control:

[0055] The dynamic programming module trains a detection efficiency optimization model based on historical data and executes three core strategies, which are the optimization strategies executed by the dynamic programming module:

[0056] The drone automatically adjusts the aerial overlap rate based on the damage density (dynamically adjustable from 20% to 80%).

[0057] The patrol vehicle's speed is matched to the complexity of the road conditions, i.e., 5-30km / h, and the sampling frequency is matched to 10-100Hz.

[0058] The mechanical dog switches between high-precision mode and fast mode. The high-precision mode is characterized by low stride speed, while the fast mode is characterized by large stride speed.

[0059] 3. Emergency Response Mechanism:

[0060] Upon receiving an emergency order from the municipality, a priority interruption protocol is activated. This means that when an emergency patrol order is received from the municipal system, the nearest available equipment is prioritized to generate a response team. An optimal dispatch chain is constructed based on GPS location to ensure that a response plan is generated within 30 seconds.

[0061] The number of devices and processing scale described herein are for simplification of the invention. Applications, modifications, and variations of this invention will be readily apparent to those skilled in the art. Although embodiments of the invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. It can be applied to various fields suitable for this invention, and further modifications can be readily implemented by those skilled in the art. Therefore, without departing from the general concept defined by the claims and their equivalents, this invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. A method for dynamic collaborative inspection of urban roads using multiple devices, characterized in that, Includes the following steps: S1. Identify road surface damage features and generate event trigger signals through a wide-area perception layer. The wide-area perception layer includes a drone. The drone is equipped with a visible light / infrared composite optical sensor to conduct wide-area patrols, identify road surface damage features, and generate event trigger signals. S2. The event trigger signal is analyzed in real time by the data intelligence decision layer to generate a scheduling instruction containing the location coordinates and damage type; S3. The dynamic execution layer performs vehicle lane detection through the scheduling instructions, collects and analyzes road surface image data and driving vibration characteristics, and performs sidewalk detection through the scheduling instructions, collects and analyzes sidewalk paving surface image data and motion vibration waveforms; the dynamic execution layer includes a patrol vehicle and a mechanical dog, wherein the patrol vehicle performs vehicle lane detection based on the scheduling instructions, collects and analyzes road surface image data and driving vibration characteristics; The robotic dog performs sidewalk detection based on the scheduling instructions, and collects and analyzes sidewalk paving surface image data and motion vibration waveforms; S4. By performing time-space correlation analysis on the image features and driving vibration features of the motor vehicle lanes in the intelligent decision-making layer, as well as the image data of the pavement surface of the sidewalk and the vibration waveform, detection results are generated. S5, the intelligent decision-making layer adjusts the inspection path and working parameters of each device based on the detection results of each terminal, forming a closed-loop feedback.

2. The method for dynamic collaborative inspection of urban roads using multiple devices as described in claim 1, characterized in that, The pavement damage characteristics include damage types and evaluation parameters. The damage types include pavement cracks, potholes, mesh cracks, repair potholes, repair mesh cracks, and repair cracks. The evaluation parameters include length / area evaluation parameters.

3. The method for dynamic collaborative inspection of urban roads using multiple devices as described in claim 1, characterized in that, The patrol vehicle is equipped with a multi-axis accelerometer array, and the vibration characteristic analysis includes the detection of the exposure duration and magnitude of abnormal vibrations; The mechanical dog's limbs are equipped with vibration sensors that collect vibration waveforms from the contact of paving bricks to determine the looseness of the paving bricks.

4. The method for dynamic collaborative inspection of urban roads using multiple devices as described in claim 3, characterized in that, The specific steps for the mechanical dog to detect the looseness of paving bricks include: 1) Perform gait traversal in the target area and collect vibration signals along both the X and Y axes; 2) Construct a model for the attenuation of vibration energy in the brick joint area; 3) Locate the three-dimensional coordinates of the loose brick based on the phase abrupt change point of the attenuation model.

5. The method for dynamic collaborative inspection of urban roads using multiple devices as described in claim 4, characterized in that, The intelligent decision-making layer includes a cloud control center, which has a built-in data fusion module. The data fusion module performs spatiotemporal correlation analysis on the damage distribution detected by the UAV, the vibration time sequence characteristics of the patrol vehicle, and the location data of the loose bricks of the mechanical dog.

6. The method for dynamic collaborative inspection of urban roads using multiple devices as described in claim 5, characterized in that, The cloud control center includes a damage and deterioration trend prediction module, which is used to analyze the potential loss of driving / walking comfort caused by the detected defects and, in conjunction with historical data, the damage and deterioration trend at that location, thereby generating a composite inspection report that includes repair priorities.

7. The method for dynamic collaborative inspection of urban roads using multiple devices as described in claim 6, characterized in that, In step S5, the intelligent decision-making layer includes a dynamic programming module, which executes the following optimization strategy: (1) Density adjustment of the subsequent loitering path of the UAV; (2) Matching the patrol vehicle's detection speed with the sensor's operating mode; (3) Dynamic balance between mechanical dog gait parameters and detection accuracy.

Images