Method for generating road surface risk map and drive path risk information in unmanned automated port

The method and system for generating an unmanned automated port road surface risk map using AGV sensors address road surface damage issues by integrating stereo and LiDAR images to predict and manage damage progression, reducing maintenance costs and enhancing safety through path distribution and real-time guidance.

KR103044503B1Active Publication Date: 2026-11-06ISM INC
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
KR1020230069898
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2023-05-31
Publication Date
2026-11-06
Estimated Expiration
2043-05-31

AI Technical Summary

Technical Problem

Unmanned automated ports face issues with road surface damage due to concentrated AGV trajectory guidance, leading to subsidence risks, frequent manual inspections, lack of quantified pothole and crack severity, and absence of absolute positional maps for driving avoidance, with limited scanning technology and high maintenance costs.

Method used

A method and system for generating an unmanned automated port road surface risk map using AGV-mounted sensors to detect and analyze road surface damage, integrating stereo and LiDAR images, predicting damage progression, and providing maintenance and driving guidance services.

Benefits of technology

Minimizes road surface damage, reduces maintenance costs, and enhances safety by evenly distributing AGV paths, analyzing damage causes, and providing real-time risk assessment and avoidance guidance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112023059998087-PAT00001_ABST
    Figure 112023059998087-PAT00001_ABST
Patent Text Reader

Abstract

A method for configuring a road surface risk map for an unmanned automated port and generating driving risk section information is provided. The unmanned automated port road surface management method according to an embodiment of the present invention generates road surface information, which is information about the road surface on which an AGV operates in an unmanned automated port, generates a map showing road surface damage information based on the generated road surface information, and provides a service for road surface management based on the road surface damage information. By doing so, the AGV movement path can be evenly distributed to minimize the extent of road surface damage and slow down the progression of damage, and human, material, and uneconomical factors caused by partial repairs and maintenance work can be minimized, and damage to cargo and equipment caused by the intensification of partial damage due to operation on the same trajectory can be minimized.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to road surface management technology for unmanned automated ports, and more specifically, to a method for detecting road surface damage in the driving section of an AGV (Automated Guided Vehicle) in an unmanned automated port and providing related services. Background Technology

[0003] In unmanned automated ports, AGVs are typically guided along the same trajectory, which leads to serious road surface damage. Currently, there is no preventive solution, and the situation is managed through the repair and reinforcement of damaged road surfaces. This is believed to be a problem arising from the fact that AGV position information is concentrated on the transponder ground installation device, relying on a center trajectory guidance method.

[0004] In addition to the problem of road surface damage, the risk caused by extensive ground subsidence is also increasing; furthermore, there is the issue that frequent inspections and maintenance are difficult due to the current reliance on manual inspections based on the manager's visual judgment and unmanned operations.

[0005] Furthermore, since it is impossible to construct a detailed map based on the absolute coordinates provided by the transponder, it is also impossible to provide an absolute positional risk map for driving avoidance. Additionally, dedicated scanning technology and equipment for driving sections are not widely available and consist of expensive devices, making continuous scanning and frequent inspection of damage conditions difficult.

[0006] In addition, there are issues where field application is difficult because the severity of potholes or cracks is not quantified, absolute map display services are not being utilized, and driving avoidance criteria have not been established. The problem to be solved

[0008] The present invention has been devised to solve the aforementioned problems. The objective of the present invention is to provide a method for constructing an unmanned automated port road surface risk map and generating driving risk section information, as a means to slow down the progression of road surface damage to minimize maintenance costs and to analyze the causes of road surface damage and present solutions. means of solving the problem

[0010] A method for managing an unmanned automated port road surface according to an embodiment of the present invention for achieving the above objective comprises: a step of generating road surface information, which is information about a road surface on which an AGV (Automated Guided Vehicle) operates in an unmanned automated port; a step of generating a map showing road surface damage information based on the generated road surface information; and a step of providing a service for road surface management based on the road surface damage information.

[0011] The road surface information generation step may include: a step of generating location information of the road surface; a step of generating a stereo image of the road surface; a step of generating a lidar image of the road surface; and a step of synchronizing the generated stereo image and lidar image based on the generated location information.

[0012] The road surface information generation step can generate road surface information using sensors mounted on the AGV.

[0013] The map generation step may include: a step of detecting the location of road surface damage in a stereo image; a step of generating road surface depth information from a LiDAR image; and a step of fusing the detected damage location and the road surface depth information to generate road surface damage information.

[0014] Road surface damage may include potholes and cracks.

[0015] The map generation step may further include: a step of reconstructing road surface damage information by adding location information to the generated road surface damage information; a step of storing the reconstructed road surface damage information in a DB for training data; and a step of displaying the reconstructed road surface damage information on a map of an unmanned automated port.

[0016] The service provision step may include a step of training a road surface risk prediction model using accumulated road surface damage information; and a step of receiving road surface damage information as input with the trained road surface risk prediction model and predicting the degree of damage progression and risk.

[0017] The service provision step may further include the step of generating a maintenance notification for the road surface and displaying it on a map based on the predicted damage progress and risk level.

[0018] The service provision step may further include a step of generating driving avoidance information based on predicted damage progress and risk level and displaying it on a map.

[0019] According to another aspect of the present invention, an unmanned automated port road surface management system is provided, characterized by comprising: a sensing system that generates road surface information, which is information about a road surface on which an AGV operates in an unmanned automated port; a map generation server that generates a map showing road surface damage information based on the generated road surface information; and a service providing platform that provides a service for road surface management based on the road surface damage information.

[0020] According to another aspect of the present invention, a method for managing a road surface in an unmanned automated port is provided, characterized by comprising: a step of generating road surface information, which is information about a road surface on which an AGV operates in an unmanned automated port; and a step of generating a map showing road surface damage information based on the generated road surface information.

[0021] According to another aspect of the present invention, an unmanned automated port road surface management system is provided, characterized by comprising: a sensing system that generates road surface information, which is information about a road surface on which an AGV operates in an unmanned automated port; and a map generation server that generates a map showing road surface damage information based on the generated road surface information. Effects of the invention

[0023] As explained above, according to the embodiments of the present invention, the AGV movement path can be evenly distributed to minimize the extent of road surface damage and slow down the progression of damage, and human, material, and uneconomical factors caused by partial repairs and maintenance work can be minimized, and the safety of cargo and damage to equipment caused by the intensification of partial damage due to operation on the same trajectory can be minimized.

[0024] In addition, according to embodiments of the present invention, data is analyzed by time and type of operation of the AGV to infer the progress of damage and to analyze the degree of damage according to the change state of the risk map, and long-term data utilization is possible to derive standard guidelines for the design, construction, and operation of unmanned ports. Brief explanation of the drawing

[0026] FIG. 1 is a diagram illustrating the configuration of an unmanned automated port road surface management system according to an embodiment of the present invention. FIG. 2 is a drawing illustrating the detailed configuration of the AGV composite sensing system shown in FIG. 1. FIG. 3 is a flowchart provided for explaining the process of a method for generating road surface information of an unmanned automated port, FIG. 4 is a flowchart provided for the process description of a method for generating a road surface risk map of an unmanned automated port, Figure 5 is a flowchart provided for the process description of a method for providing road surface management services in an unmanned automated port. Specific details for implementing the invention

[0027] The present invention will be described in more detail below with reference to the drawings.

[0028] An embodiment of the present invention presents a method for configuring an unmanned automated port road surface risk map and generating driving risk section information.

[0029] The technology provides the following detailed technologies for detecting road surface damage in the driving section of an AGV (Automated Guided Vehicle) in an unmanned automated port, generating image information synchronized with the absolute position coordinates of the AGV driving section, and transmitting stereo image information and LiDAR sensor data of the scanned section to a server to perform information processing and precise damage recognition.

[0031] 1) Data collection based on complex sensing

[0032] - Recognition of damaged areas and precise crack depths based on high-definition stereo video data

[0033] - Extraction and display of damage depth data via high-speed LiDAR scanning

[0034] - Synchronization of composite sensing data and absolute position information

[0036] 2) Generation of a risk map applying damage progression via periodic scanning

[0037] - Equip the AGV with a scanning device to periodically acquire data.

[0038] - Intelligent analysis applying a combination of AGV cargo load, transport frequency, external environmental change factors, road surface composition data, etc.

[0039] - Uses artificial intelligence to determine the progress of damage by time and driving trajectory

[0041] 3) Identification of driving risk zones and generation of information

[0042] - New driving route guidance capable of identifying damage risk factors and avoiding hazardous areas based on real-time AGV cargo information

[0043] - Extraction of road surface damage causes based on long-term driving trajectory information

[0045] FIG. 1 is a diagram illustrating the configuration of an unmanned automated port road surface management system according to an embodiment of the present invention. As illustrated, the unmanned automated port road surface management system according to an embodiment of the present invention comprises an AGV composite sensing system (100), a road surface risk map generation server (200), and a road surface management service provision platform (300).

[0046] The AGV composite sensing system (100) is a system for sensing information about the road surface of an unmanned automated port, and the road surface risk map generation server (200) generates a map in which risk information about the road surface is displayed from the information sensed by the AGV composite sensing system (100), and the road surface management service provision platform (300) provides a service for road surface management from the risk information generated by the road surface risk map generation server (200).

[0047] FIG. 2 is a diagram illustrating the detailed configuration of the AGV composite sensing system (100) shown in FIG. 1. The AGV composite sensing system (100) is configured to be mounted / installed on an AGV (10) and includes a positioning sensor (111, 112), an image sensor (121, 122), and a LiDAR sensor (131, 132, 133).

[0048] The positioning sensors (111, 112) are sensors for generating position information of the AGV composite sensing system (100), that is, the position information of the AGV (10), the image sensors (121, 122) are sensors for generating high-definition stereo images of the road surface of the unmanned automated port, and the lidar sensors (131, 132, 133) are sensors for generating lidar images of the road surface.

[0049] FIG. 3 is a flowchart provided for explaining the process of a method for generating road surface information of an unmanned automated port by an AGV composite sensing system (100).

[0050] As described above, first, the positioning sensors (111, 112) of the AGV composite sensing system (100) generate real-time location information of the AGV (10) (S410), the image sensors (121, 122) generate high-definition stereo images of the road surface of the unmanned automated port (S420), and the lidar sensors (131, 132, 133) generate lidar images of the road surface (S430).

[0051] The next AGV composite sensing system (100) synchronizes the stereo image generated in step S420 and the lidar image generated in step S430 based on the real-time location information generated in step S410 (S440), then packetizes them (S450) and wirelessly transmits them to the road surface risk map generation server (200) (S460).

[0052] FIG. 4 is a flowchart provided for explaining the process of a method for generating a road surface risk map of an unmanned automated port by a road surface risk map generation server (200).

[0053] When all road surface information of an unmanned automated port is generated by the AGV composite sensing system (100) and stored in the DB (S510), the road surface risk map generation server (200) separates the information stored in the DB into stereo images and LiDAR images (S520).

[0054] Next, the road surface risk map generation server (200) detects pothole and crack areas in the stereo image (S530) and generates road surface depth information from the LiDAR image (S540). Then, the road surface risk map generation server (200) fuses the location of the pothole and crack areas detected in step S530 with the road surface depth information generated in step S540 to generate road surface damage information (S550).

[0055] In the road surface damage information generated in step S550, the type (shape) and location of the damage are identified from stereo images, and the damage depth (degree of gouging) is identified from lidar images, so the content of the information can be accurately labeled.

[0056] Subsequently, the road surface risk map generation server (200) reconstructs the road surface damage information by adding location information stored in the DB to the road surface damage information generated in step S550 (S560). The reconstructed road surface damage information is stored in the DB for training data (S570) and displayed on the map of the unmanned automated port (S580).

[0057] FIG. 5 is a flowchart provided for explaining the process of a method for providing road surface management services for an unmanned automated port by a road surface management service providing platform (300).

[0058] As described above, the road surface management service provision platform (300) trains a road surface risk prediction model using road surface damage information built in a training data DB (S610). The road surface risk prediction model is a machine learning model that takes road surface damage information as input and outputs damage progression and risk, and can be implemented with a structure such as a CNN. The road surface risk prediction model can be supervised learning using a training dataset consisting of road surface damage information and damage progression / risk labels.

[0059] The road surface risk prediction model, once trained, receives road surface damage information generated by the road surface risk map generation server (200) and predicts the degree of damage progression and risk (S620 and S630).

[0060] If the degree of damage progression is above a standard, i.e., damage has progressed to a certain extent (S620-is), and the road surface risk is above a standard, i.e., the road surface risk is determined to be high (S630-high), the service platform (300) updates and displays the map by reflecting the risk level generated by the map generation server (200) (S640). Meanwhile, the service platform (300) may also generate and provide a new map unique to the degree of road surface damage progression (S650).

[0061] The map updated in step S640 or the map generated in step S650 can be used to identify risk factors for damage based on real-time cargo information of the AGV, and can also function as basic data required for a service model that can find the cause of road surface damage through a new driving path guidance that can avoid dangerous sections and a long-term driving trajectory information learning process.

[0062] Afterwards, the service platform (300) can generate a maintenance notification for the road surface (S660), display the generated notification on a map (S670), and provide a maintenance notification service.

[0063] In addition, the service platform (300) can generate driving avoidance information based on the road surface risk level (S680), display the generated avoidance driving path on a map (S690), and provide an avoidance guide service. In generating the avoidance driving path, the AGV's cargo load, transport frequency, external environmental change factors, road surface composition data, etc., can be applied in combination.

[0064] So far, the method for configuring an unmanned automated port road surface risk map and generating driving risk section information has been described in detail with reference to preferred embodiments.

[0065] In the above embodiment, to minimize maintenance costs by slowing down the progression of road surface damage, a technology is presented that minimizes the extent of road surface damage and slows down the progression of damage by evenly distributing the AGV path, minimizes human and material and uneconomical factors caused by partial repairs and maintenance work, and minimizes the safety of cargo and equipment damage caused by the intensification of partial damage through operation on the same trajectory.

[0066] In addition, the above embodiment presents a technology for analyzing the causes of road surface damage and resolving them by analyzing data by time and type of operation of the AGV to infer the progress of damage, analyzing the degree of damage according to the change state of the risk map, and presenting long-term data to derive standard guidelines for the design, construction, and operation of unmanned ports.

[0067] The technical concept of the present invention is applicable to an intelligent road surface scanning system mounted on an AGV, a wireless information transmission device for road surface scanning information and a server image processing / recognition solution, a map configuration and operation information service for an unmanned port platform, a port road surface management and maintenance service solution, etc.

[0068] In this case, there is an additional effect of reducing maintenance costs and equipment damage, and the quality of port services can be improved through the safe transport of cargo.

[0069] Furthermore, although preferred embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above. Various modifications are possible by those skilled in the art without departing from the essence of the invention as claimed in the claims, and such modifications should not be understood individually from the technical spirit or perspective of the present invention. Explanation of the symbols

[0071] 100 : AGV Composite Sensing System 111,112: Positioning sensors 121,122 : Image sensor 131,132,133: LiDAR sensor 200 : Road surface hazard map generation server 300: Road surface management service provision platform

Claims

Claim 1 The method comprises: a step of generating road surface information, which is information about a road surface on which an AGV (Automated Guided Vehicle) operates in an unmanned automated port; a step of generating a map showing road surface damage information based on the generated road surface information; and a step of providing a service for road surface management based on the road surface damage information; wherein the road surface information generation step includes: a step of generating location information of the road surface; a step of generating a stereo image of the road surface; a step of generating a LiDAR image of the road surface; and a step of synchronizing the generated stereo image and the LiDAR image based on the generated location information; and the map generation step includes: a step of detecting the location of road surface damage in the stereo image; and a step of generating road surface damage depth information from the LiDAR image. A method for managing an unmanned automated port road surface, comprising: a step of generating road surface damage information by fusing detected damage location and road surface damage depth information; and a service provision step comprising: a step of training a road surface risk prediction model using accumulated road surface damage information; a step of receiving road surface damage information into the trained road surface risk prediction model and predicting damage progress and risk; a step of updating a map by reflecting the predicted risk; and a step of generating a driving avoidance path based on the predicted damage progress and risk and displaying it on a map; wherein the generation of the driving avoidance path references the cargo load of an AGV, transport frequency, external environment, and road surface configuration data, and the updated map is used in a service model for finding the cause of road surface damage. Claim 2 delete Claim 3 The unmanned automated port road surface management method according to claim 1, wherein the road surface information generation step is characterized by generating road surface information using sensors mounted on an AGV. Claim 4 delete Claim 5 An unmanned automated port road surface management method according to claim 1, characterized in that the road surface damage includes potholes and cracks. Claim 6 A method for managing a road surface in an unmanned automated port, characterized in that, in claim 1, the map generation step further comprises: a step of reconstructing road surface damage information by adding location information to the generated road surface damage information; a step of storing the reconstructed road surface damage information in a DB for training data; and a step of displaying the reconstructed road surface damage information on a map of an unmanned automated port. Claim 7 delete Claim 8 The unmanned automated port road surface management method according to claim 1, characterized in that the service provision step further includes the step of generating a maintenance notification for the road surface and displaying it on a map based on the predicted damage progress and risk level. Claim 9 delete Claim 10 A sensing system that generates road surface information, which is information about the road surface on which an AGV operates in an unmanned automated port; a map generation server that generates a map showing road surface damage information based on the generated road surface information; A service providing platform that provides a service for road surface management based on road surface damage information; wherein the sensing system generates location information of the road surface, generates a stereo image of the road surface, generates a LiDAR image of the road surface, and synchronizes the generated stereo image and LiDAR image based on the generated location information; the map generation server detects the location of road surface damage in the stereo image, generates road surface damage depth information from the LiDAR image, and generates road surface damage information by fusing the detected damage location and road surface damage depth information; the service providing platform trains a road surface risk prediction model using the accumulated road surface damage information, receives road surface damage information as input to the trained road surface risk prediction model to predict the progress of damage and the risk level, updates the map by reflecting the predicted risk level, generates a driving avoidance path based on the predicted progress of damage and the risk level and displays it on the map, wherein the AGV's cargo load, transport frequency, external environment, and road surface configuration data are referenced in the generation of the driving avoidance path, and the updated map is used in a service model for finding the cause of road surface damage. System. Claim 11 delete Claim 12 delete

Citation Information

Patent Citations

  • Road condition recognition system and method thereof for automobile safety

    KR1020140104565A

  • System and method of Automatically Generating High Definition Map Based on Camera Images

    KR1020210081983A

  • Method and device for detecting a road surface using a lidar sensor

    KR1020230001003A

  • Road surface of tunnel monitoring device and road surface of tunnel maintenance system using the same

    KR1020200022726A

  • Autonomous driving device for detecting road condition and operation method thereof

    KR1020210100775A