Autonomous inspection system of UAV based on laser point cloud

Abstract

The present invention discloses an autonomous inspection system for unmanned aerial vehicles (UAVs) based on laser point clouds, which relates to the technical field of electric power operation and inspection, and solves the technical problem that in the prior art, during the inspection of power transmission lines, UAVs need to manually perform laser modeling during flight, and cannot automatically plan a reasonable inspection route, resulting in low inspection efficiency of the UAVs. The present invention establishes an inspection three-dimensional model based on point cloud data, and plans an autonomous inspection route based on the inspection three-dimensional model. An edge inspection module starts to control the UAV to complete a corresponding inspection task based on the autonomous inspection route, and then obtains autonomous inspection data. The inspection route can be automatically planned based on the laser point cloud data, and the inspection efficiency of the UAVs is improved. The edge inspection module of the present invention pre-analyzes the acquired exclusive inspection data, reduces the pressure of data processing, and can timely identify abnormal power transmission equipment. At the same time, each edge inspection module communicates information with each other during the inspection process, and reasonably sets supplementary flight tasks to ensure inspection efficiency.

Description

Technical Field

[0001] The present invention belongs to the field of electric power operation and inspection, and relates to an unmanned aerial vehicle autonomous inspection technology based on laser point cloud, and specifically to an unmanned aerial vehicle autonomous inspection system based on laser point cloud. Background Art

[0002] Using drones to inspect power transmission lines or equipment is a very efficient operation mode, but most power companies complete inspection tasks by manually controlling drones, which has low operation efficiency and poor inspection results.

[0003] The prior art (invention patent with publication number CN109062233A) discloses a transmission line UAV automatic driving inspection method, which realizes the flight route planning, automatic driving and refined inspection of the UAV, and can improve the inspection quality. In the transmission line inspection process, the prior art requires the UAV to manually fly laser modeling, and cannot automatically plan a reasonable inspection route, resulting in low UAV inspection efficiency; therefore, there is an urgent need for an autonomous UAV inspection system based on laser point cloud. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes an autonomous UAV inspection system based on laser point cloud, which is used to solve the technical problem that in the prior art, during the inspection of transmission lines, UAVs need to manually fly laser modeling, and reasonable inspection routes cannot be automatically planned, resulting in low UAV inspection efficiency.

[0005] The present invention obtains point cloud data through screening by the central control module, establishes a three-dimensional inspection model based on the point cloud data, and plans an autonomous inspection route based on the three-dimensional inspection model. The edge inspection module starts to control the UAV to complete the corresponding inspection task according to the autonomous inspection route, and then obtains autonomous inspection data. It can automatically plan the inspection route according to the laser point cloud data, thereby improving the UAV inspection efficiency.

[0006] To achieve the above-mentioned object, a first aspect of the present invention provides an autonomous inspection system for unmanned aerial vehicles based on laser point cloud, comprising a central control module, and a plurality of edge inspection modules connected thereto, wherein the edge inspection modules are configured in association with the unmanned aerial vehicles;

[0007] Edge inspection module: starting a drone according to an autonomous inspection route, and flying a dedicated inspection route through the drone; wherein the dedicated inspection route is allocated from the autonomous inspection route; and

[0008] Acquire exclusive inspection data when flying the exclusive inspection route, perform pre-analysis on the exclusive inspection data, and exchange information with the central control module according to the analysis results;

[0009] Central control module: filtering and acquiring point cloud data according to the target position, establishing a three-dimensional inspection model according to the point cloud data, and planning the autonomous inspection route according to the three-dimensional inspection model; and

[0010] The analysis results are verified, and a warning signal is generated according to the verification results; a number of the exclusive inspection data are integrated and analyzed to generate autonomous inspection data.

[0011] Preferably, the central control module is in communication and / or electrically connected with a plurality of the edge inspection modules;

[0012] The edge inspection module is associated with the drone one by one and is built into the associated drone; wherein the edge inspection module is connected to a data acquisition terminal carried by the drone.

[0013] Preferably, the central control module obtains the point cloud data by screening in combination with the target position, including:

[0014] Obtaining original point cloud data of the power transmission lines in the monitoring area;

[0015] Obtain the geographical location corresponding to the power transmission equipment and mark it as the target location;

[0016] The original point cloud data is screened according to the target position to obtain the point cloud data.

[0017] Preferably, the central control module obtains the autonomous inspection route according to the inspection three-dimensional model, including:

[0018] Establishing and acquiring the inspection three-dimensional model according to the point cloud data;

[0019] Determine the number of drones according to the number of power transmission equipment or the area covered in the inspection three-dimensional model;

[0020] The autonomous inspection route is obtained in combination with the number of drones planned, and the autonomous inspection route is sent to several edge inspection modules.

[0021] Preferably, after receiving the autonomous inspection route, a plurality of the edge inspection modules start the corresponding associated drones;

[0022] The plurality of edge inspection modules cooperate with each other to divide the autonomous inspection route into a plurality of the exclusive inspection routes and implement the allocation of the exclusive inspection routes.

[0023] Preferably, the edge inspection module controls the associated drone to fly the exclusive inspection route to obtain the exclusive inspection data, including:

[0024] Starting the associated UAV and the data acquisition terminal carried by the UAV;

[0025] Flying along the exclusive inspection route to collect route inspection data in real time, and performing data processing on the route inspection data to obtain the exclusive inspection data;

[0026] The inspection process of the drone is recorded, an exclusive inspection record is obtained, and the exclusive inspection record is associated with the exclusive inspection data.

[0027] Preferably, the edge inspection module pre-analyzes the exclusive inspection data, including:

[0028] Extracting the exclusive inspection data according to the target location and marking it as target inspection data;

[0029] The target inspection data is identified and analyzed in combination with the built-in intelligent analysis model, a pre-analysis tag is generated according to the identification and analysis result, and the pre-analysis tag is sent to the central control module in real time.

[0030] Preferably, the central control module integrates a number of the exclusive inspection data to obtain the autonomous inspection data, including:

[0031] fusing and matching a number of the exclusive inspection data into the inspection three-dimensional model;

[0032] According to the set requirements, some of the exclusive inspection data are analyzed and extracted, and the inspection three-dimensional model is rendered according to the analysis and extraction results to obtain the autonomous inspection data.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. The present invention obtains point cloud data through screening by the central control module, establishes a three-dimensional inspection model based on the point cloud data, and plans an autonomous inspection route based on the three-dimensional inspection model. The edge inspection module starts to control the drone to complete the corresponding inspection task according to the autonomous inspection route, and then obtains autonomous inspection data. It can automatically plan the inspection route according to the laser point cloud data, thereby improving the drone inspection efficiency.

[0035] 2. The edge inspection module of the present invention pre-analyzes the acquired exclusive inspection data, reduces the data processing pressure, and can identify abnormal power transmission equipment in a timely manner; at the same time, each edge inspection module communicates information with each other during the inspection process, and reasonably sets the supplementary flight tasks to ensure the inspection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 It is a schematic diagram of the working steps of the present invention. DETAILED DESCRIPTION

[0038] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] The prior art (invention patent with publication number CN109062233A) discloses a transmission line UAV automatic driving inspection method, which realizes the flight route planning, automatic driving and refined inspection of the UAV, and can improve the inspection quality. In the transmission line inspection process, the prior art requires the UAV to manually fly laser modeling, and cannot automatically plan a reasonable inspection route, resulting in low UAV inspection efficiency.

[0040] The present invention obtains point cloud data through screening by the central control module, establishes a three-dimensional inspection model based on the point cloud data, and plans an autonomous inspection route based on the three-dimensional inspection model. The edge inspection module starts to control the UAV to complete the corresponding inspection task according to the autonomous inspection route, and then obtains autonomous inspection data. It can automatically plan the inspection route according to the laser point cloud data, thereby improving the UAV inspection efficiency.

[0041] See also Figure 1 , the first aspect of the present application provides an autonomous inspection system for unmanned aerial vehicles based on laser point cloud, including a central control module, and a plurality of edge inspection modules connected thereto, and the edge inspection modules are configured in association with the unmanned aerial vehicles;

[0042] Edge inspection module: Launch drones based on autonomous inspection routes, and fly dedicated inspection routes through drones; and

[0043] Acquire exclusive inspection data when flying exclusive inspection routes, pre-analyze the exclusive inspection data, and exchange information with the central control module based on the analysis results;

[0044] Central control module: filter and obtain point cloud data according to the target location, establish a 3D inspection model based on the point cloud data, and plan an autonomous inspection route based on the 3D inspection model; and

[0045] Verify the analysis results and generate early warning signals based on the verification results; integrate and analyze several exclusive inspection data to generate autonomous inspection data.

[0046] The central control module in this application is equivalent to the central server, and its functions include analyzing the point cloud data to obtain the autonomous inspection route, and obtaining the inspection results of the monitoring area based on the inspection results sent by the edge inspection module, that is, the autonomous inspection data. The edge inspection module is equivalent to the edge server, and its main functions include data interaction with the central control module and controlling the associated drone to complete the inspection task.

[0047] The exclusive inspection routes in this application are allocated from the autonomous inspection routes, that is, each edge inspection module parses the exclusive inspection routes according to the autonomous inspection routes. It can be understood that the autonomous inspection routes include several exclusive inspection routes, and each exclusive inspection route should be equipped with a drone.

[0048] The autonomous inspection data in this application is composed of several exclusive inspection data and matched to the inspection three-dimensional model. The autonomous inspection data of the monitoring area can be used to quickly locate the power transmission equipment and analyze the status of the corresponding power transmission equipment.

[0049] The central control module in the present application communicates and / or is electrically connected to a number of edge inspection modules; the edge inspection modules are associated with drones one by one and are built into the associated drones, and at the same time the edge inspection modules are connected to the data acquisition terminals carried by the drones.

[0050] The edge inspection module is installed inside the drone and communicates with the central control module and the drone through the communication module. The edge inspection module collects data through the data acquisition terminal carried by the drone. The data acquisition terminal includes an image acquisition device and various sensors; the image acquisition device is specifically a camera.

[0051] The central control module in this application obtains point cloud data in combination with target location screening, including:

[0052] Obtaining original point cloud data of the power transmission lines in the monitoring area;

[0053] Obtain the geographical location corresponding to the power transmission equipment and mark it as the target location;

[0054] The original point cloud data is filtered according to the target position to obtain the point cloud data.

[0055] The central control module acquires the original point cloud data in the monitoring area through lidar collection. At that time, the original power data contained a lot of redundant data, which would increase the difficulty of 3D modeling. Therefore, it was necessary to screen the original point cloud data.

[0056] Power grid inspection mainly inspects the power transmission equipment in the monitoring area, so the geographical location of the power transmission equipment is used as the target location, and the target location and its related point cloud data are retained to complete the screening. In the process of screening the original point cloud data, the shape and size of the power transmission equipment can also be combined for fine screening.

[0057] It should be noted that if it is a regular inspection, all power transmission equipment in the monitoring area are the inspection objects; if it is a key inspection, the key power transmission equipment in the monitoring area is the inspection object. The location of the inspection object is the above-mentioned target location.

[0058] The central control module in this application obtains the autonomous inspection route based on the inspection three-dimensional model, including:

[0059] Establish and obtain inspection 3D model based on point cloud data;

[0060] Determine the number of drones based on the number of power transmission equipment or the area covered in the inspection three-dimensional model;

[0061] Combined with the number of drones, autonomous inspection routes are obtained and sent to several edge inspection modules.

[0062] After obtaining the point cloud data, the central control module uses the 3D modeling software to obtain the inspection 3D model. After determining the inspection 3D model, the number of drones also needs to be determined.

[0063] In a preferred embodiment, determining the number of drones according to the number of power transmission equipment in the inspection three-dimensional model includes:

[0064] Obtain the number of power transmission equipment in the inspection three-dimensional model and mark it as SS;

[0065] The required number of drones is determined by the formula WS=α×SS; where α is a proportional coefficient obtained based on actual experience and is a real number greater than 0.

[0066] Alternatively, the monitoring area can be divided into several sector-shaped areas based on the center of the monitoring area. Each sector-shaped area can be equipped with a drone, which can ensure both the inspection quality and the inspection efficiency.

[0067] In another preferred embodiment, the number of drones can also be determined based on the area covered by the monitoring region, that is, the most suitable inspection area for a drone is obtained, and the number of drones can be determined by dividing the monitoring area according to the most suitable inspection area.

[0068] In the present application, after receiving the autonomous inspection route, several edge inspection modules start the corresponding associated drones; several edge inspection modules cooperate with each other to divide the autonomous inspection route into several exclusive inspection routes, and realize the allocation of exclusive inspection routes.

[0069] Several edge inspection modules communicate and cooperate with each other. Each edge inspection module is assigned a dedicated inspection route, and controls the associated drone to inspect the power transmission equipment on the dedicated inspection route.

[0070] The edge inspection module in this application controls the associated drone to fly a dedicated inspection route and obtains dedicated inspection data, including:

[0071] Start the associated UAV and the data collection terminal carried by the UAV;

[0072] Fly along the dedicated inspection route to collect route inspection data in real time, and process the route inspection data to obtain dedicated inspection data;

[0073] The inspection process of the drone is recorded, exclusive inspection records are obtained, and the exclusive inspection records are associated with exclusive inspection data.

[0074] The data processing of route inspection data mainly includes image preprocessing, that is, image segmentation, image denoising and other processing on the collected image data to remove useless image data to ensure data storage and transmission efficiency.

[0075] The exclusive inspection record should at least record the inspection progress of the drone and whether the inspection route deviates from the exclusive inspection route; at the same time, the inspection progress should be compared with the preset progress to balance the inspection speed of each drone.

[0076] The edge inspection module in this application performs pre-analysis on the exclusive inspection data, including:

[0077] Extract exclusive inspection data according to the target location and mark it as target inspection data;

[0078] Combined with the built-in intelligent analysis model, the target inspection data is identified and analyzed, and pre-analysis tags are generated according to the identification and analysis results, and the pre-analysis tags are sent to the central control module in real time.

[0079] The intelligent analysis model is obtained through neural network model training, and is mainly used to identify whether the power transmission equipment is normal based on target inspection data, such as an image recognition model based on a neural network.

[0080] During the inspection process, the drone will continuously collect and obtain exclusive inspection data. If the exclusive inspection data is analyzed and identified after the inspection task is completed, abnormal power transmission equipment may not be processed in time. If the exclusive inspection data is analyzed and identified immediately, the inspection efficiency will be reduced due to the high data processing pressure. Therefore, the inspection data corresponding to the power transmission equipment corresponding to the target location is analyzed to ensure the normal operation of key power transmission equipment.

[0081] When the central control module receives the analysis results corresponding to the pre-analysis, the analysis results should be verified by the staff. If there is indeed an abnormality after verification, an early warning signal will be generated and maintenance personnel will be arranged to perform on-site inspection and maintenance.

[0082] In some other preferred embodiments, additional key power transmission equipment may be defined, and when the UAV obtains the inspection data of the defined power transmission equipment, it immediately analyzes and provides feedback.

[0083] The central control module in this application integrates several exclusive inspection data to obtain autonomous inspection data, including:

[0084] Fusion and matching of several exclusive inspection data into the inspection 3D model;

[0085] Analyze and extract a number of exclusive inspection data according to the set requirements, render the inspection 3D model based on the analysis and extraction results, and obtain autonomous inspection data.

[0086] By combining several exclusive inspection data and matching them to the inspection 3D model, the inspection results can be displayed intuitively. You can also extract and analyze the exclusive inspection data according to the set requirements, and render the analysis results into the inspection 3D model.

[0087] The requirements are set for targeted analysis of several exclusive inspection data. For example, if the cause of abnormal power transmission equipment needs to be analyzed, the exclusive inspection data corresponding to the abnormal power transmission equipment is first extracted and analyzed to determine the cause of the abnormality.

[0088] In this application, each edge inspection module communicates data with each other during the inspection process. If a drone completes the inspection task early, it analyzes the inspection progress of other drones and assists other drones to complete the inspection task.

[0089] Working principle of the present invention:

[0090] The central control module obtains point cloud data based on target location screening, builds a three-dimensional inspection model based on the point cloud data, and plans an autonomous inspection route based on the three-dimensional inspection model.

[0091] The edge inspection module starts the UAV according to the autonomous inspection route, flies the exclusive inspection route by the UAV, and obtains exclusive inspection data when flying the exclusive inspection route.

[0092] The central control module integrates and analyzes a number of exclusive inspection data and generates autonomous inspection data based on the inspection 3D model.

[0093] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An autonomous inspection system for unmanned aerial vehicles based on laser point cloud, comprising a central control module and a plurality of edge inspection modules connected thereto, wherein the edge inspection modules are associated with the unmanned aerial vehicles, and characterized in that: Edge inspection module: starting a drone according to an autonomous inspection route, and flying a dedicated inspection route through the drone; wherein the dedicated inspection route is allocated from the autonomous inspection route; and Acquire exclusive inspection data when flying the exclusive inspection route, perform pre-analysis on the exclusive inspection data, and exchange information with the central control module according to the analysis results; Central control module: filtering and acquiring point cloud data according to the target position, establishing a three-dimensional inspection model according to the point cloud data, and planning the autonomous inspection route according to the three-dimensional inspection model; and Verify the analysis results and generate an early warning signal according to the verification results; integrate and analyze a number of the exclusive inspection data to generate autonomous inspection data; The central control module filters and obtains the point cloud data in combination with the target position, including: Obtaining original point cloud data of the power transmission lines in the monitoring area; Obtain the geographical location corresponding to the power transmission equipment and mark it as the target location; Filtering the original point cloud data according to the target position to obtain the point cloud data; The central control module obtains the autonomous inspection route according to the inspection three-dimensional model, including: Establishing and acquiring the inspection three-dimensional model according to the point cloud data; Determine the number of drones according to the number of power transmission equipment or the area covered in the inspection three-dimensional model; Acquire the autonomous inspection route in combination with the number of drones planned, and send the autonomous inspection route to a number of edge inspection modules; After receiving the autonomous inspection route, the plurality of edge inspection modules start the corresponding associated drones; The plurality of edge inspection modules cooperate with each other to divide the autonomous inspection route into a plurality of exclusive inspection routes, and realize the allocation of the exclusive inspection routes; The central control module is in communication and / or electrically connected with a plurality of the edge inspection modules; The edge inspection module is associated with the drone one by one and is built into the associated drone; wherein the edge inspection module is connected to a data acquisition terminal carried by the drone.

2. The UAV autonomous inspection system based on laser point cloud according to claim 1 is characterized in that: The edge inspection module controls the associated UAV to fly the exclusive inspection route and obtains the exclusive inspection data, including: Starting the associated UAV and the data acquisition terminal carried by the UAV; Flying along the exclusive inspection route to collect route inspection data in real time, and performing data processing on the route inspection data to obtain the exclusive inspection data; The inspection process of the drone is recorded, an exclusive inspection record is obtained, and the exclusive inspection record is associated with the exclusive inspection data.

3. The laser point cloud-based UAV autonomous inspection system according to claim 2 is characterized in that: The edge inspection module pre-analyzes the exclusive inspection data, including: Extracting the exclusive inspection data according to the target location and marking it as target inspection data; The target inspection data is identified and analyzed in combination with the built-in intelligent analysis model, a pre-analysis tag is generated according to the identification and analysis result, and the pre-analysis tag is sent to the central control module in real time.

4. The laser point cloud-based UAV autonomous inspection system according to claim 3 is characterized in that: The central control module integrates a number of the exclusive inspection data to obtain the autonomous inspection data, including: fusing and matching a number of the exclusive inspection data into the inspection three-dimensional model; According to the set requirements, some of the exclusive inspection data are analyzed and extracted, and the inspection three-dimensional model is rendered according to the analysis and extraction results to obtain the autonomous inspection data.

Images