Intelligent inspection robot system and intelligent inspection method for underground comprehensive pipe gallery

Abstract

The invention discloses an intelligent inspection robot system for an underground comprehensive pipe gallery and an intelligent inspection method, relates to the technical field of underground comprehensive pipe gallery monitoring, and solves the problems that the detection capability of internal defects of a pipeline is limited, the communication stability is poor and the like due to difficult positioning and navigation in the existing underground pipe gallery monitoring technology. The robot end comprises a multi-dimensional sensor, a laser radar, a central processing unit, an ARM processor, a motor assembly and a wheel type assembly. And a wheel type assembly moving design is adopted, and different from a traditional fixed track inspection mode, the working efficiency and the flexibility are remarkably improved. A multi-dimensional sensor fusion technology is integrated, comprehensive monitoring and early warning of a pipe gallery environment are realized, the accuracy of a routing inspection path is ensured through real-time mapping, path planning and intelligent obstacle avoidance in combination with a mileage positioning system, the high precision and reliability of routing inspection are ensured, and the safety and efficiency of underground pipe gallery routing inspection are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of underground comprehensive pipe gallery monitoring, in particular to a patrol robot system based on a Nezha development platform. BACKGROUND

[0002] With the rapid advancement of urbanization and efficient development and utilization of underground space resources, as a core component of urban infrastructure, the monitoring and maintenance requirements of the internal facilities of the underground comprehensive pipe gallery are increasing, and the development and application of underground pipe gallery patrol robots have become an important way to solve this problem. Although the existing intelligent patrol robots use visual recognition, intelligent navigation, etc. to improve the patrol efficiency and accuracy to a certain extent, they still face some technical challenges. Positioning and navigation in complex environments are difficult, the detection capability of subtle defects in the pipeline is limited, and the communication stability is insufficient, etc. The key problems to be solved by the current intelligent patrol system are to realize comprehensive evaluation and predictive maintenance of the pipe gallery environment.

[0003] Based on this background, the present application uses artificial intelligence and sensor theory, relies on a high-performance Nezha development platform, and integrates multi-dimensional environmental perception, precise modeling and path planning algorithms to build an integrated pipe gallery patrol system that integrates comprehensive environmental detection, autonomous navigation, intelligent obstacle avoidance and real-time monitoring. The purpose is to provide a complete solution for the intelligent management of various key pipelines in underground pipe galleries and to realize normal intelligent monitoring and early warning of target space. The system robot end, server end and operation and maintenance end work together to monitor the cracking, deformation, pipeline leakage and damage, temperature and humidity, gas, etc. of the underground comprehensive pipe gallery structure in real time, and the environmental parameters are transmitted back to the operation and maintenance end in real time. Form a complete intelligent patrol system to effectively solve the safety problems existing in the patrol. SUMMARY

[0004] The present application provides an intelligent patrol robot system for underground comprehensive pipe galleries and an intelligent patrol method to solve the problems of existing underground pipe gallery monitoring technology, such as difficult positioning and navigation, limited detection capability of internal defects in the pipeline, and poor communication stability. The system is based on the Nezha development platform, combines AI and robot-related technologies, uses multi-dimensional environmental sensors, high-precision mapping and path planning algorithms, and designs a comprehensive pipe gallery patrol system with multi-dimensional environmental detection, autonomous path planning and navigation, intelligent obstacle avoidance, and real-time monitoring to realize intelligent management of various practical pipelines in underground pipe galleries.

[0005] The intelligent patrol robot system for underground comprehensive pipe galleries includes a robot end, a server end and a network operation and maintenance end. The robot end includes multi-dimensional sensors, laser radars, central processors, ARM processors, motor assemblies and wheel assemblies.

[0006] The central processor receives the environmental parameters detected by the multi-dimensional sensor, generates an underground pipe gallery environment map according to the point cloud data scanned by the laser radar in the multi-dimensional sensor, and obtains the driving track of the robot by using a path planning algorithm, and transmits the driving track to the ARM processor;

[0007] The ARM processor controls the rotation of the motor assembly, thereby driving the wheeled assembly to move in the underground pipe gallery according to the driving track.

[0008] The central processor sends the environmental parameters detected by the multi-dimensional sensor and the position information of the robot to the server side in real time, and the server side generates an inspection detection report and real-time data after receiving the detected environmental parameters and transmits them to the network operation side.

[0009] The network operation side runs the data transmitted by the server side and performs visual display.

[0010] The application also provides an intelligent inspection method for an underground comprehensive pipe gallery, which is realized by the following steps:

[0011] Step one, initialize the intelligent inspection robot system;

[0012] Step two, the robot side moves in the underground pipe gallery according to the set path, the multi-dimensional sensor collects the environmental parameters of the underground pipe gallery and transmits them to the central processor;

[0013] Step three, the central processor integrates the environmental parameters sent by the multi-dimensional sensor and generates an environment map, and realizes the positioning of the robot through the odometer counting realized by the motor assembly, and transmits the environmental parameters and positioning information to the server side through the communication module.

[0014] Step four, the server side generates a detection report after processing the received data, and transmits the processed data to the network operation side.

[0015] Step five, remote monitoring, management and visual display are performed through the network operation side.

[0016] The application has the following beneficial effects:

[0017] First, the application uses an Intel N97 processor as the central processor, integrates a network interface controller, realizes parallel computing of data, optimizes the data processing flow, accelerates the data flow speed, and enhances the system real-time response and overall performance.

[0018] Secondly, the multi-dimensional sensor is adopted in the application, the environment of the whole underground pipe gallery is detected, early warning and comprehensive evaluation are carried out through measuring parameters such as methane, temperature and humidity in the environment, the safety of the later relevant staff is ensured to enter, the infrared temperature sensor is used to monitor the heating, power and water supply systems in the underground pipe gallery in real time, the working state of the pipeline is monitored, automatic monitoring and early warning are carried out, the three-dimensional point cloud data of the pipe gallery is obtained by using the laser radar, and favorable data support is provided for subsequent robot automatic tracking and obstacle avoidance.

[0019] Thirdly, the automatic inspection system based on the Nezha development platform is designed, autonomous navigation and intelligent obstacle avoidance are carried out by using high-precision mapping and path planning algorithm. The high-precision underground pipe gallery system map is generated by using the Hector SLAM algorithm, the robot driving track is obtained by combining the LSTM-DDPG path planning algorithm, the real-time detection of the forward path is carried out by using the laser radar, and the real-time positioning is realized by using the laser radar and the encoder in the direct-current speed reducer, which replaces the inaccurate underground positioning of GPS, ensures that the robot can still maintain high-precision positioning when long-distance inspection, and can timely adjust the route when encountering obstacles or abnormal conditions, and mark the position of the obstacle, so as to clean the obstacle in the later period. Thus, the path planning, autonomous navigation and intelligent obstacle avoidance functions of the inspection robot in the underground pipe gallery are realized.

[0020] Fourthly, the laser radar and the mileage calculation are used to realize accurate positioning, since the GPS signal needs to be directly connected with the satellite, and the underground environment can block or seriously weaken these signals, the three-dimensional point cloud data obtained by the laser radar is used to realize real-time mapping, the encoder in the direct-current speed reducer calculates the number of rotations, and the mileage is calculated according to the wheel diameter and the number of rotations. The combination of the two can realize that the robot can still maintain high-precision positioning when long-distance inspection.

[0021] Fifthly, the robot low-power automatic return charging is designed, the power management module of the robot end monitors the power in real time, and when the power is lower than the preset threshold, the automatic return program is started. The space position description of the underground pipe gallery where the mobile robot is located and the standard charging station position are established, the robot automatically plans the shortest return path and returns to the charging station for charging.

[0022] Sixthly, the service system based on the Nezha development platform is designed, due to the influence of environmental factors such as high network delay and poor stability in the pipe gallery, the CoAP lightweight network communication protocol is adopted, the measurement data can be uploaded to the server end in real time, the communication between multiple devices is realized, and the communication delay is low. The CoAP protocol has retransmission mechanism and confirmation mechanism, which effectively ensures the reliability of communication.

[0023] VII. This invention designs an operation and maintenance system based on the Nezha development platform, which receives, processes, and visualizes environmental information. It adopts a B / S architecture to design a browser interface and displays it on an LCD screen in real time. Users can interact with the network operation and maintenance terminal through the display interface, which facilitates remote monitoring and auxiliary management and improves the convenience and flexibility of operation.

[0024] 8. The intelligent inspection robot system for underground utility tunnels described in this invention integrates multi-dimensional sensors onto the Nezha development platform, enabling real-time detection and visualization of various integrated lines within the utility tunnel. Attached Figure Description

[0025] Figure 1 This is a block diagram of the intelligent inspection robot system for underground integrated utility tunnels as described in this invention;

[0026] Figure 2 This is a flowchart illustrating the path planning and automatic obstacle avoidance process of the inspection robot described in this invention. Detailed Implementation

[0027] Specific Implementation Method 1: Combination Figure 1 and Figure 2 This embodiment describes a high-precision intelligent inspection robot system for underground utility tunnels. The system includes a robot terminal, a server terminal, and a network maintenance terminal.

[0028] The robot end includes multi-dimensional sensors, a central processing unit, an ARM processor, a motor assembly, a wheel assembly, and a power management module;

[0029] The robot integrates Intel. The N97 processor, serving as both a central processing unit and an ARM processor (F407VET6 microcontroller), is used for complex calculations and real-time control, ensuring efficient and stable system operation.

[0030] Intel The N97 processor integrates a network interface controller, enabling it to connect to a network and perform data transmission via network devices. (Intel) The N97 processor sends the environmental parameters measured by the multi-dimensional sensors and the robot's position information to the server. After receiving the detection data, the server generates an inspection report and real-time data, which is then transmitted to the network maintenance terminal. The network maintenance terminal connects to the network through network devices to receive data and runs data processing and visualization programs. The visualization interface is then displayed on an LCD screen to support decision support and emergency response.

[0031] In this embodiment, the multidimensional sensor includes a methane sensor, an infrared temperature sensor, a temperature and humidity sensor, a camera, and a lidar. The methane sensor is used to monitor the concentration of combustible gas in the underground utility tunnel in real time, the infrared temperature sensor detects the heating of the cables, the temperature and humidity sensor detects temperature and humidity data, the camera captures the situation inside the utility tunnel in real time, and the lidar performs spatial scanning to generate three-dimensional point cloud data.

[0032] The environmental parameter data detected by the multi-dimensional sensor is preprocessed and then processed by Intel. The N97 processor performs data fusion, utilizing data fusion algorithms to ensure the comprehensiveness and accuracy of environmental monitoring. (Intel) The N97 processor generates a high-precision map of the underground utility tunnel environment based on point cloud data scanned by LiDAR; it calculates mileage using motor speed to achieve precise positioning; it obtains the robot's trajectory using path planning algorithms and sends the trajectory to the ARM processor. The ARM processor controls the rotation of the motor assembly (multiple MD DC geared motors). Through the drive of the integrated wheel assembly, the robot can move flexibly in the complex underground utility tunnel environment.

[0033] The robot's power management module monitors the battery level in real time and automatically returns to recharge when the battery level falls below a preset threshold. Simultaneously, the system matches mileage data returned by the DC geared motor with real-time radar scan data and map modeling. This data is then processed by the central processing unit using a corresponding algorithm to achieve positioning. The mileage counting positioning within the DC geared motor ensures the accuracy of long-distance inspections.

[0034] The server-side primarily acts as a bridge in the overall system architecture, providing interfaces for robot-side application data access and business logic processing upwards, and interacting with the file system, local data, and IoT cloud platform downwards. A remote server is set up to connect the robot-side and the network maintenance terminal, receiving real-time video and location data from the robot-side and responding to interaction requests from the network maintenance terminal to display robot status, real-time environmental monitoring, video, and location information. Furthermore, after inspection completion, an inspection monitoring report is generated and uploaded to the network maintenance terminal.

[0035] In this embodiment, the robot and the server communicate via an NB-IoT module, Intel. The N97 processor connects to the NB-IoT module, transmitting measurement data in real-time via the NB-IoT module to the CoAP server software running on the server side in CoAP message format. This enables communication between multiple devices with low latency, facilitating remote real-time monitoring and data analysis. Upon receiving a data request, the CoAP server decodes, verifies, and stores the data. The server then analyzes the received data to generate a real-time status report. This embodiment proposes a combination of CoAP and NB-IoT, providing not only wide coverage and low power consumption communication but also ensuring reliable and secure data transmission, meeting the needs of various underground utility tunnel application scenarios.

[0036] In this embodiment, the network operation and maintenance terminal includes a computer, network equipment, and an LCD screen. The computer of the network operation and maintenance terminal is connected to the network through the network equipment to receive and process data. It uses B / S architecture technology to design a browser, which includes a system background control interface, environmental parameters, robot position information, and robot perspective video data. The network operation and maintenance terminal can be used for auxiliary monitoring of artificial underground utility tunnels.

[0037] Specific Implementation Method Two: Combination Figure 2 This embodiment describes an intelligent detection method implemented using the high-precision intelligent inspection robot system for underground utility tunnels as described in Specific Embodiment 1. The specific implementation process of this method is as follows:

[0038] Step 1, System Initialization:

[0039] Robot side: Start Intel The N97 processor and ARM Cortex-M3 processor are connected to the network to initialize multi-dimensional sensors, including a methane sensor, an infrared temperature sensor, a temperature and humidity sensor, and a lidar.

[0040] Server-side: Connects to the network and prepares to receive and process data.

[0041] Network operations and maintenance terminal: Connect to the network, prepare to receive and process data, start the B / S architecture software, and prepare for remote monitoring and management.

[0042] Step 2, Environmental Data Collection:

[0043] The robot's multi-dimensional sensors begin collecting environmental parameters from the underground utility tunnel, such as methane concentration, temperature, and humidity. These parameters are then transmitted in real-time to the server via the CoAP communication protocol.

[0044] Step 3, Data Processing and Fusion:

[0045] The robot's central processing unit receives multi-dimensional sensor parameters, processes and fuses the data, and uses HectorSLAM to generate a real-time environmental map. Simultaneously, the encoders on the motor assembly enable odometry, ensuring the robot's accurate perception of its environment.

[0046] Step 4, Path Planning and Obstacle Avoidance:

[0047] Based on the environmental map and inspection task requirements, the LSTM-DDPG algorithm is used for path planning to achieve autonomous navigation and intelligent obstacle avoidance. The path planning strategy is optimized by combining the Long Short-Term Memory network and deep reinforcement learning within the LSTM-DDPG algorithm. The robot monitors obstacles ahead in real time and can dynamically adjust its path based on real-time environmental data to avoid obstacles and ensure the smooth progress of the inspection task.

[0048] Step 5: Inspection task execution:

[0049] The robot performs inspections along a planned path, employing a wheeled mobility system for stable, fast, and flexible movement. It utilizes LiDAR for real-time obstacle avoidance and odometry to ensure accuracy over long distances, guaranteeing a smooth and safe inspection process. The robot transmits inspection data and real-time video to a server for network maintenance monitoring and management. The CoAP communication protocol ensures stable and low-latency data transmission.

[0050] like Figure 2 As shown, based on the environmental map and inspection task requirements, a high-precision grid map is generated using the Hector SLAM algorithm, leveraging LiDAR data to achieve environmental depiction and self-localization. Simultaneously, the LSTM-DDPG algorithm is employed for path planning, integrating global and local strategies and utilizing a self-attention mechanism to enhance dynamic obstacle avoidance capabilities in complex environments.

[0051] The robot performs inspections according to a planned path, controls the movement of wheeled components, and uses multi-dimensional sensors to collect data, which is then processed by Intel. N97 processor receives;

[0052] Determine if an obstacle is present. If so, adjust the path and return to local path planning; otherwise, end or wait for new instructions. Ensure that the robot can both effectively plan paths and intelligently avoid obstacles.

[0053] Step 6, Battery Monitoring and Automatic Return:

[0054] The robot's power management module monitors the battery level in real time. When the battery level falls below a preset threshold, it initiates an automatic return-to-home procedure. The robot plans the shortest return path and safely returns to the charging station for recharging.

[0055] Step 7: Report Generation and Data Analysis

[0056] The server utilizes the CoAP lightweight network communication protocol to receive and process requests from the front end, enabling rapid data exchange and report generation, ensuring the timeliness and transparency of information transmission. Reports are generated based on inspection data to analyze potential security risks and environmental changes. The network operations and maintenance team views reports and real-time data through B / S architecture software, providing decision support and emergency response.

[0057] Step 8, System Maintenance and Upgrade:

[0058] Based on the inspection results and system operation, perform necessary maintenance and upgrades on the robot system. Update the software and algorithms to improve system stability and efficiency.

[0059] The intelligent inspection method described in this embodiment significantly improves inspection efficiency and reduces labor costs and safety risks in the practical application of underground utility tunnels. Through high-precision environmental perception and autonomous decision-making capabilities, it can promptly identify potential fault points, reduce accidents caused by human negligence, and provide reliable assurance for the intelligent operation and maintenance of urban infrastructure.

[0060] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0061] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A high-precision intelligent inspection robot system for underground utility tunnels, comprising a robot terminal, a server terminal, and a network maintenance terminal; its characteristics are: The robot end includes multi-dimensional sensors, lidar, central processing unit, ARM processor, motor assembly, and wheel assembly; The central processing unit receives environmental parameters detected by multi-dimensional sensors, generates an underground utility tunnel environmental map based on the point cloud data scanned by the lidar in the multi-dimensional sensors, obtains the robot's driving trajectory using a path planning algorithm, and transmits the driving trajectory to the ARM processor. The ARM processor controls the rotation of the motor assembly, thereby driving the wheel assembly to move in the underground utility tunnel according to the travel trajectory; The central processing unit sends the environmental parameters detected by the multi-dimensional sensors and the robot's position information to the server in real time. After receiving the detected environmental parameters, the server generates an inspection report and real-time data and transmits them to the network maintenance terminal. The network operation and maintenance terminal runs the data transmitted by the server and displays it visually.

2. The high-precision intelligent inspection robot system for underground utility tunnels according to claim 1, characterized in that: The robot also includes a power management module, which monitors the robot's battery level in real time and automatically returns to its home base to recharge when the battery level falls below a preset threshold.

3. The high-precision intelligent inspection robot system for underground utility tunnels according to claim 1, characterized in that: The multidimensional sensor includes a methane sensor, an infrared temperature sensor, a temperature and humidity sensor, a camera, and a lidar. The methane sensor monitors the concentration of combustible gas in the underground utility tunnel in real time, the infrared temperature sensor detects the heating of the cables, the temperature and humidity sensor detects the temperature and humidity data in the underground utility tunnel, the camera captures the situation in the utility tunnel in real time, and the lidar performs spatial scanning to generate three-dimensional point cloud data.

4. The high-precision intelligent inspection robot system for underground utility tunnels according to claim 1, characterized in that: The robot's position information is obtained by counting the mileage returned by the encoder in the motor assembly, thus enabling the robot to be located.

5. The high-precision intelligent inspection robot system for underground utility tunnels according to claim 1, characterized in that: The robot and the server exchange data via an NB-IoT module using the CoAP network communication protocol, and the server generates a real-time status report.

6. The high-precision intelligent inspection robot system for underground utility tunnels according to claim 1, characterized in that: The network operation and maintenance terminal adopts a B / S architecture browser design, which includes the display of control interface, environmental parameters, robot position information and robot perspective video data, realizing auxiliary monitoring of artificial underground utility tunnels through the network operation and maintenance terminal.

7. A high-precision intelligent inspection method for underground utility tunnels, characterized by: This method is implemented using the high-precision intelligent inspection robot system for underground utility tunnels as described in any one of claims 1-6. The specific implementation process of this method is as follows: Step 1: Initialize the intelligent inspection robot system; Step 2: The robot moves along the set path in the underground utility tunnel, and the multi-dimensional sensors collect environmental parameters of the underground utility tunnel and transmit them to the central processing unit. Step 3: The central processing unit integrates the environmental parameters sent by the multi-dimensional sensors and generates an environmental map. It also uses the mileage counting implemented by the motor component to locate the robot. The environmental parameters and positioning information are then transmitted to the server through the communication module. Step 4: The server processes the received data and generates a detection report, and then transmits the processed data to the network operation and maintenance terminal. Step 5: Perform remote monitoring, management, and visualization through the network operation and maintenance terminal.

8. The intelligent inspection method according to claim 7, characterized in that: In step three, the central processing unit processes and integrates the received multi-dimensional sensor-detected underground utility tunnel environmental parameters, and generates a high-precision underground utility tunnel environmental map using the Hector SLAM algorithm based on the point cloud data scanned by the lidar. It also obtains the robot's driving trajectory by combining the LSTM-DDPG path planning algorithm. Real-time positioning is achieved through real-time detection of the forward path by the lidar and the mileage count returned by the motor components, ensuring that the robot maintains high-precision positioning during long-distance inspections.

9. The intelligent inspection method according to claim 7, characterized in that: In step three, the robot determines in real time whether obstacles appear during its movement. When obstacles appear, it avoids obstacles in real time using LiDAR and adjusts the local path planning by combining the LSTM-DDPG path planning algorithm.

10. The intelligent inspection method according to claim 7, characterized in that: The communication module is an NB-IoT module.

Images