Unmanned aerial vehicle non-inductive safety supervision and inspection method and system based on three-dimensional point cloud route automatic planning

Abstract

The invention discloses an unmanned aerial vehicle non-inductive safety supervision and inspection method and system based on three-dimensional point cloud route automatic planning. The method comprises the following steps: collecting ledger data; issuing a non-inductive supervision inspection work order; scheduling an optimal airport; marking a no-fly zone and potential obstacles according to the point cloud data of the tower line; constructing an outbound route; according to the inspection type of the inspection work order, constructing an inspection operation area point route, if the inspection type is general inspection, the inspection operation area point route is an S-shaped route, and if the inspection type is key inspection, the inspection operation area point route is a surrounding route; constructing a return route according to the reverse order of the forward route; generating a complete safety supervision route by combining the go route, the supervision operation area point route and the return route; and carrying out three-dimensional point cloud collision detection, if the detection is passed, issuing the detection result to the unmanned aerial vehicle to execute a complete safety supervision route, otherwise, carrying out fine adjustment on the complete safety supervision route until the detection is passed. According to the invention, the efficiency and accuracy of route planning are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) route planning, and particularly to a method and system for non-intrusive safety supervision and inspection of UAVs based on automatic three-dimensional point cloud route planning. Background Art

[0002] In recent years, UAV technology has developed rapidly and has been widely used in many fields. In the field of inspection, the traditional manned inspection method has low efficiency, high cost and certain safety risks. With the maturity of UAV technology, using UAVs for inspection has gradually become a trend. Currently, during the inspection process, UAVs usually require manual route planning. This method relies on the experience of operators, has low efficiency, and is difficult to adapt to complex environments.

[0003] The existing UAV inspection route planning methods mainly include rule-based planning and optimization algorithm-based planning. The rule-based planning method determines the route according to preset rules, such as flying at a fixed distance and angle. This method is simple but lacks flexibility and is difficult to adapt to different operation scenarios. The optimization algorithm-based planning methods, such as genetic algorithms and particle swarm algorithms, although can optimize the route to a certain extent, have high computational complexity, high hardware requirements, and poor performance in dealing with complex obstacle environments. In addition, most of the existing inspection methods cannot achieve non-intrusive safety supervision and are difficult to efficiently complete the inspection task while ensuring safety. Summary of the Invention

[0004] Aiming at the problems of low efficiency, poor adaptability and difficulty in achieving non-intrusive safety supervision in the existing UAV safety supervision and inspection route planning, the present invention provides a method and system for non-intrusive safety supervision and inspection of UAVs based on automatic three-dimensional point cloud route planning, which improves the automation degree of route planning, reduces manual intervention; enables the route planning to better adapt to complex environments, including various obstacles and different operation areas; realizes non-intrusive safety supervision during the inspection process of UAVs, and ensures the efficient and safe progress of the inspection task.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for non-intrusive safety supervision and inspection of UAVs based on automatic three-dimensional point cloud route planning, comprising the following steps:

[0007] Collect ledger data, including tower line point cloud data, substation point cloud data, longitude and latitude coordinates of towers, tower height, tower type, span between towers, and location boundary coordinates of the operation area;

[0008] Dispatch non-intrusive supervision and inspection work orders;

[0009] Dispatch the optimal airport and assign inspection tasks to the UAVs;

[0010] Mark the no-fly zone and potential obstacles based on the point cloud data of the pole and tower lines;

[0011] Construct the outbound route based on the departure airport, the point cloud data of the pole and tower lines, the no-fly zone, and potential obstacles;

[0012] Construct the point route of the inspection operation area according to the inspection type of the inspection work order. If the inspection type is general inspection, the point route of the inspection operation area is an S-shaped route. If the inspection type is key inspection, the point route of the inspection operation area is a circular route;

[0013] Construct the return route in reverse order of the outbound route;

[0014] Generate a complete safety inspection route by combining the outbound route, the point route of the inspection operation area, and the return route;

[0015] Perform 3D point cloud collision detection on the complete safety inspection route. If the detection passes, send it to the drone to execute the complete safety inspection route. Otherwise, fine-tune the complete safety inspection route until the detection passes.

[0016] To optimize the above technical solution, the specific measures taken also include:

[0017] Further, the construction of the outbound route is specifically as follows: Based on the departure airport, the point cloud data of the pole and tower lines, the no-fly zone, and potential obstacles, use the K-nearest neighbor algorithm to calculate the optimal path. The flight height of the drone for the outbound route is the sum of the height of the highest pole in the path and the safety redundancy height.

[0018] Further, the construction method of the circular route is as follows:

[0019] Set the key inspection point of the key operation area as the center, set the circular radius according to the target size to form a circular operation area range, take several equally spaced points on the circular area as the operation inspection photo-taking points, the drone traverses the operation inspection photo-taking points in turn, takes pictures facing the operation area inspection point at each operation inspection photo-taking point, and sets the camera tilt angle.

[0020] Further, the construction method of the S-shaped route is as follows:

[0021] Set basic parameters, including flight height, flight speed, and route spacing;

[0022] Define the area boundary. The drone starts from one end of the area and flies along the long side. After reaching the boundary, it turns in an arc to enter the next route, forming a continuous S-shaped route.

[0023] Further, the flight height is determined according to the ground resolution requirement, and the calculation formula is as follows:

[0024] H = f × sensor width × GSD × image width

[0025] Wherein, H is the flight altitude, f is the camera focal length, and GSD is the ground resolution.

[0026] Furthermore, the flight speed is matched with the camera shutter interval, and the calculation formula is as follows:

[0027] V = image coverage width × (1 - lateral overlap rate) × shutter interval

[0028] Wherein, V is the flight speed.

[0029] Furthermore, the flight line spacing is determined by the lateral overlap rate, and the calculation formula is as follows:

[0030] D = image width × (1 - lateral overlap rate) × GSD × pixel size

[0031] Wherein, D is the flight line spacing, and GSD is the ground resolution.

[0032] Furthermore, the method further includes:

[0033] In the case where the S-shaped flight line or the circular flight line cannot meet the complex on-site operation environment, the UAV switches to remote manual control at the supervision site, controls the forward, backward, left, and right directions of the UAV's travel direction, up and down control, turns the pan-tilt head, and performs manual remote photographing control. After completing the remote manual control, it switches back to the autonomous inspection mode, and the UAV returns to the top safety point position of the supervision site.

[0034] Furthermore, the construction of the outbound flight line is specifically:

[0035] Based on the departure airport, the point cloud data of the pole and tower line, the no-fly zone, and potential obstacles, the Dijkstra algorithm or the A* algorithm is used to calculate the optimal path, and the flight altitude of the UAV on the outbound flight line is the sum of the highest pole height in the path and the safety redundancy height.

[0036] The present invention also proposes a UAV non-intrusive safety supervision and inspection system based on three-dimensional point cloud flight line automatic planning, including:

[0037] A data acquisition module for acquiring ledger data, including point cloud data of pole and tower lines, point cloud data of substations, longitude and latitude coordinates of poles and towers, pole and tower heights, pole and tower types, tower span, and location boundary coordinates of the operation area;

[0038] A work order issuing module for issuing non-intrusive supervision and inspection work orders;

[0039] A work order scheduling module for scheduling the optimal airport and assigning inspection tasks to the UAV;

[0040] An environmental modeling module for marking no-fly zones and potential obstacles based on tower line point cloud data;

[0041] A flight route construction module for constructing an outbound flight route based on the departure airport, tower line point cloud data, no-fly zones, and potential obstacles; constructing a point flight route for the inspection operation area according to the inspection type of the inspection work order. If the inspection type is general inspection, the point flight route for the inspection operation area is an S-shaped route. If the inspection type is key inspection, the point flight route for the inspection operation area is a circular route; constructing a return flight route in reverse order of the outbound flight route; generating a complete safe inspection flight route by combining the outbound flight route, the point flight route for the inspection operation area, and the return flight route;

[0042] A flight route verification module for performing three-dimensional point cloud collision detection on the complete safe inspection flight route. If the detection passes, it issues the complete safe inspection flight route to the UAV for execution. Otherwise, it makes fine adjustments to the complete safe inspection flight route until the detection passes.

[0043] The beneficial effects of the present invention are as follows: (1) Compared with the prior art, the present invention realizes automatic flight route planning based on three-dimensional point cloud data and various algorithms, reduces manual intervention, improves the efficiency and accuracy of flight route planning, and can quickly adapt to different operation scenarios. For example, in the inspection scenario of complex mountain transmission lines, traditional manual flight route planning may take several hours, while the automatic planning algorithm of the present invention only takes a few minutes to complete.

[0044] (2) The multi-angle coverage flight route design and the layered flight strategy can inspect the operation area more comprehensively and meticulously, improving the inspection quality. By setting different camera angles and flight altitudes, more inspection data can be obtained and more potential problems can be discovered. Description of the Drawings

[0045] Figure 1 It is the overall flowchart of the UAV non-intrusive safety inspection and patrol method for automatic flight route planning based on three-dimensional point cloud proposed by the present invention.

[0046] Figure 2 It is a schematic diagram of the safe inspection circular flight route.

[0047] Figure 3 It is the operation effect diagram of the actual safe inspection circular flight route.

[0048] Figure 4 It is a schematic diagram of the safe inspection S-shaped flight route.

[0049] Figure 5 It is the operation effect diagram of the actual safe inspection S-shaped flight route. Detailed Embodiment

[0050] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0051] Embodiment 1

[0052] The present invention proposes a method for non-intrusive safety supervision and inspection of unmanned aerial vehicles based on automatic three-dimensional point cloud flight path planning. The overall process of this method is as Figure 1 shown, including the following steps:

[0053] Collect ledger data, including tower line point cloud data, substation point cloud data, the longitude and latitude coordinates of the towers, tower height, tower type (such as transmission tower / communication tower), span between towers, and the location boundary coordinates of the operation area; in this embodiment, the ledger data is prepared by the holographic digital power grid system.

[0054] The unmanned aerial vehicle inspection business system issues non-intrusive supervision and inspection work orders.

[0055] Dispatch the optimal airport and assign inspection tasks to the unmanned aerial vehicle.

[0056] Mark no-fly zones and potential obstacles based on the tower line point cloud data.

[0057] Construct an outbound flight path based on the departure airport, tower line point cloud data, no-fly zones, and potential obstacles; specifically, based on the departure airport, tower line point cloud data, no-fly zones, and potential obstacles, use the K-nearest neighbor algorithm to calculate the optimal path. The flight height of the unmanned aerial vehicle for the outbound flight path is the sum of the highest tower height in the path and the safety redundancy height. In this embodiment, the safety redundancy height ≥ 10 meters. In addition to the K-nearest neighbor algorithm, the Dijkstra algorithm or A* algorithm can also be used to calculate the optimal path. In three-dimensional space, for point p(x, y, z) (the three-dimensional coordinates of the off-site safety point) and point Q(x0, y0, z0) (the three-dimensional coordinate point of the safety point at the top of the tower), the Euclidean distance formula can directly and accurately measure the straight-line distance between the starting point in space and the safety point of the line tower, ensuring its safety and reliability and being relatively close to the airport.

[0058] Construct a point flight path for the supervision operation area according to the inspection type of the inspection work order. If the inspection type is general supervision, the point flight path for the supervision operation area is an S-shaped flight path. If the inspection type is key supervision, the point flight path for the supervision operation area is a circular flight path; Camera angle adjustment: The camera tilt angle can be adjusted between 50° - 80° according to different operation requirements to adapt to special inspection targets and take into account side view and top view.

[0059] The construction method of the circumferential flight path is as follows:

[0060] Set the supervision point of the key operation area as the center, set the circumferential radius according to the target size to form a circular operation area range, take several equally divided points on the circular area as the operation supervision photographing points, the UAV traverses the operation supervision photographing points in sequence, takes pictures facing the operation area supervision point at each operation supervision photographing point, and sets the camera inclination angle.

[0061] For example, set the longitude and latitude coordinates of a transmission tower as the supervision center point, take 10m above the tower top, and form a supervision area with a radius of half of the cross-arm length of the tower plus 10m. Divide 8 operation supervision photographing points on the circular area. The UAV flies from photographing point 1 to photographing point 8 in sequence, sets the camera inclination angle of 50°-80° at each supervision photographing point to take pictures facing the supervision point. The UAV takes pictures during this process to ensure 360° monitoring without dead angles of the operation area of the supervision point. The schematic diagram of the circumferential flight path for safety supervision is as Figure 2 shown, and the actual operation effect diagram of the circumferential flight path for safety supervision is as Figure 3 shown.

[0062] The construction method of the S-shaped flight path is as follows:

[0063] Set basic parameters, including flight altitude, flight speed, and flight path spacing; the flight altitude is determined according to the ground resolution requirement, and the calculation formula is as follows:

[0064] H = f × sensor width × GSD × image width

[0065] In the formula, H is the flight altitude, f is the camera focal length, and GSD is the ground resolution.

[0066] The flight speed is matched with the camera shutter interval, and the calculation formula is as follows:

[0067] V = image coverage width × (1 - lateral overlap rate) × shutter interval

[0068] In the formula, V is the flight speed.

[0069] The flight path spacing is determined by the lateral overlap rate, and the calculation formula is as follows:

[0070] D = image width × (1 - lateral overlap rate) × GSD × pixel size

[0071] In the formula, D is the flight path spacing, and GSD is the ground resolution. The lateral overlap rate is usually ≥60% to avoid gaps.

[0072] Use GIS tools (such as Google Earth, QGIS) or drone planning software (such as Pix4Dcapture, DJIGSPro) to delimit the area boundary. The drone starts from one end of the area and flies along the long side. After reaching the boundary, it turns in an arc to enter the next flight path, forming a continuous S-shaped flight path. The inspection flight path flies in an S shape above the inspection operation points, sets the camera inclination angle at 75° (not vertical), mainly for looking down, and collects the top view of the operation area. The schematic diagram of the S-shaped flight path for safety inspection is as shown in Figure 4 shown, and the actual effect diagram of the S-shaped flight path operation for safety inspection is as shown in Figure 5 shown.

[0073] Construct the return flight path in reverse order of the outbound flight path;

[0074] Combine the outbound flight path, the point flight path of the inspection operation area, and the return flight path to generate a complete safety inspection flight path;

[0075] Perform three-dimensional point cloud collision detection on the complete safety inspection flight path. If the detection passes, send it to the drone to execute the complete safety inspection flight path. Otherwise, fine-tune the complete safety inspection flight path until the detection passes.

[0076] The drone automatically takes off from the airport according to the automatically generated complete safety inspection flight path, executes the outbound flight path, and executes the S-shaped flight path or the circumferential flight path after reaching the safety inspection point. In the case where the S-shaped flight path or the circumferential flight path cannot meet the complex on-site operation environment, the drone switches to remote manual control at the inspection site, controls the forward, backward, left, and right directions of the drone's travel, controls up and down, turns the pan-tilt head, and controls manual remote photography. After completing the remote manual control, it switches back to the autonomous inspection mode, and the drone returns to the top safety point position of the inspection site. The drone returns to the airport according to the return flight path.

[0077] During the flight of the drone, the heights of the high-altitude layer and the low-altitude layer can be adjusted according to the actual situation. For example, in some areas with open views and fewer obstacles, the high-altitude layer can be increased to 120 meters, and the low-altitude layer can be reduced to 40 meters to better meet the inspection requirements.

[0078] Embodiment 2

[0079] The present invention proposes a drone non-intrusive safety inspection and patrol system based on three-dimensional point cloud flight path automatic planning corresponding to the method of Embodiment 1, including:

[0080] The data acquisition module is used to acquire ledger data, including pole line point cloud data, substation point cloud data, longitude and latitude coordinates of poles, pole heights, pole types, tower spans, and location boundary coordinates of the operation area;

[0081] The work order issuing module is used to issue non-intrusive inspection and patrol work orders;

[0082] The work order scheduling module is used to schedule the optimal airport and assign inspection tasks to the drones.

[0083] The environmental modeling module is used to label no-fly zones and potential obstacles based on the point cloud data of the pole and tower lines.

[0084] The flight route construction module is used to construct the outbound flight route based on the departure airport, the point cloud data of the pole and tower lines, the no-fly zones, and the potential obstacles; construct the point flight route of the supervision operation area according to the inspection type of the inspection work order. If the inspection type is general supervision, the point flight route of the supervision operation area is an S-shaped route. If the inspection type is key supervision, the point flight route of the supervision operation area is a surrounding route; construct the return flight route in reverse order of the outbound flight route; generate a complete safety supervision flight route by combining the outbound flight route, the point flight route of the supervision operation area, and the return flight route.

[0085] The flight route verification module is used to perform 3D point cloud collision detection on the complete safety supervision flight route. If the detection passes, it is sent to the drone to execute the complete safety supervision flight route. Otherwise, the complete safety supervision flight route is fine-tuned until the detection passes.

[0086] The implementation methods of each module and module functions in the system are exactly the same as the steps of the method in Embodiment 1, so they will not be described in detail here.

[0087] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this application can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0088] The above is only the preferred implementation mode of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art of this technology, several improvements and refinements made without departing from the principle of the present invention should be regarded as within the protection scope of the present invention.

Claims

1. A method for drone non-intrusive safety inspection and patrol based on automatic route planning of three-dimensional point clouds, characterized in that, It includes the following steps: Collect ledger data, including tower line point cloud data, substation point cloud data, longitude and latitude coordinates of towers, tower height, tower type, span between towers, and location boundary coordinates of the operation area; Dispatch non-intrusive inspection patrol work orders; Dispatch the optimal airport and assign inspection tasks to the UAV; Mark no-fly zones and potential obstacles based on tower line point cloud data; Construct an outbound route based on the departure airport, tower line point cloud data, no-fly zones, and potential obstacles; Construct an inspection operation area point route according to the inspection type of the inspection work order. If the inspection type is general inspection, the inspection operation area point route is an S-shaped route. If the inspection type is key inspection, the inspection operation area point route is a circular route; Construct a return route in reverse order according to the outbound route; Generate a complete safety inspection route by combining the outbound route, inspection operation area point route, and return route; Perform 3D point cloud collision detection on the complete safety inspection route. If the detection passes, send it to the UAV to execute the complete safety inspection route. Otherwise, fine-tune the complete safety inspection route until the detection passes.

2. The method for non-intrusive safety supervision and inspection of drones based on automatic route planning of three-dimensional point clouds according to claim 1, characterized in that, The specific construction of the outbound route is as follows: Based on the departure airport, tower line point cloud data, no-fly zones, and potential obstacles, use the K-nearest neighbor algorithm to calculate the optimal path. The flight height of the outbound route UAV is the sum of the highest tower height in the path and the safety redundancy height.

3. The method for non-intrusive safety supervision and inspection of an unmanned aerial vehicle based on automatic route planning of three-dimensional point clouds as claimed in claim 1, wherein The construction method of the circular route is as follows: Set the key operation area inspection point as the center, set the circular radius according to the target size to form a circular operation area range. Take several equally spaced points on the circular area as operation inspection photo-taking points. The UAV traverses the operation inspection photo-taking points in turn, takes pictures facing the operation area inspection point at each operation inspection photo-taking point, and sets the camera tilt angle.

4. The method for drone non-intrusive safety inspection and patrol based on automatic route planning of three-dimensional point clouds as claimed in claim 1, wherein, The construction method of the S-shaped route is as follows: Set basic parameters, including flight height, flight speed, and route spacing; Define the area boundary. The UAV starts from one end of the area and flies along the long side. After reaching the boundary, it turns in an arc to enter the next route, forming a continuous S-shaped route.

5. The method for non-intrusive safety inspection and patrol of an unmanned aerial vehicle based on automatic route planning of three-dimensional point clouds according to claim 4, wherein, The flight height is determined according to the ground resolution requirement, and the calculation formula is as follows: H = f × sensor width × GSD × image width In the formula, H is the flight height, f is the camera focal length, and GSD is the ground resolution.

6. The method for drone non-intrusive safety inspection and patrol based on automatic three-dimensional point cloud route planning as claimed in claim 4, wherein, The flight speed is matched with the camera shutter interval, and the calculation formula is as follows: V = image coverage width × (1 - lateral overlap rate) × shutter interval In the formula, V is the flight speed.

7. The method for drone non-intrusive safety supervision and inspection based on automatic route planning of three-dimensional point clouds according to claim 4, wherein The route spacing is determined by the lateral overlap rate, and the calculation formula is as follows: D = image width × (1 - lateral overlap rate) × GSD × pixel size In the formula, D is the route spacing, and GSD is the ground resolution.

8. The method for non-intrusive safety supervision and inspection of drones based on automatic route planning of three-dimensional point clouds according to claim 1, wherein, The method also includes: In the case where the S-shaped route or circular route cannot meet the complex on-site operation environment, the UAV switches to remote manual control at the inspection site, controls the forward, backward, left, and right directions, up and down directions of the UAV, turns the pan-tilt head, and performs manual remote photo-taking control. After completing the remote manual control, it switches back to the autonomous inspection mode, and the UAV returns to the top safety point position of the inspection site.

9. The method for non-intrusive safety supervision and inspection of an unmanned aerial vehicle based on automatic route planning of three-dimensional point clouds according to claim 1, characterized in that, The specific construction of the outbound route is as follows: Based on the departure airport, point cloud data of pole and tower lines, no-fly zones, and potential obstacles, the Dijkstra algorithm or A* algorithm is used to calculate the optimal path. The flight altitude of the UAV for the outbound route is the sum of the height of the highest pole and tower in the path and the safety redundancy height.

10. A drone non-intrusive safety inspection and patrol system based on automatic three-dimensional point cloud flight path planning, characterized in that, Including: A data collection module for collecting ledger data, including point cloud data of pole and tower lines, point cloud data of substations, longitude and latitude coordinates of poles and towers, pole and tower heights, pole and tower types, inter-tower spans, and location boundary coordinates of the operation area; A work order issuing module for issuing non-intrusive supervision and inspection work orders; A work order scheduling module for scheduling the optimal airport and assigning inspection tasks to UAVs; An environmental modeling module for marking no-fly zones and potential obstacles based on the point cloud data of pole and tower lines; A route construction module for constructing an outbound route based on the departure airport, point cloud data of pole and tower lines, no-fly zones, and potential obstacles; constructing a point route for the supervision operation area according to the inspection type of the inspection work order. If the inspection type is general supervision, the point route for the supervision operation area is an S-shaped route. If the inspection type is key supervision, the point route for the supervision operation area is a circumferential route; constructing a return route in reverse order of the outbound route; generating a complete safety supervision route by combining the outbound route, the point route for the supervision operation area, and the return route; A route verification module for performing 3D point cloud collision detection on the complete safety supervision route. If the detection passes, it is sent to the UAV to execute the complete safety supervision route. Otherwise, the complete safety supervision route is fine-tuned until the detection passes.

Images