A method for generating dynamic return routes for substation drones

By building safe areas and dynamic path planning, the problem of the intersection of the drone return path and the equipment area is solved, and safe and efficient return within the substation is achieved, reducing the risk of collision.

CN120385350BActive Publication Date: 2025-08-22TIANJIN RICHSOFT ELECTRIC POWER INFORMATION TECH +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202510873038.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-06-27
Publication Date
2025-08-22
Estimated Expiration
2045-06-27

AI Technical Summary

Technical Problem

The existing drone return method fails to fully consider the terrain and obstacle distribution in the substation, resulting in the return path crossing the equipment area, posing a collision risk, especially when the power is insufficient.

Method used

By building a double-layer safety area based on voltage levels, combining point cloud data and two-dimensional maps, a return path that avoids obstacles is dynamically generated, and the A* algorithm is used to calculate the shortest safety path to ensure that the drone can return safely at low power.

Benefits of technology

It improves the ability of drones to respond in complex environments, reduces the risk of equipment damage, ensures safe return, and improves the flexibility and adaptability of path planning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120385350B_ABST
    Figure CN120385350B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for generating a dynamic return route for a UAV in a substation, comprising: S1, constructing a double-layer safety area based on voltage level: n A safe rectangular area is constructed as the center. The bottom of the safe rectangular area is a square with a length of 2r and a height of h. safe ; Then construct an inscribed safety cylinder in the safety cuboid, the radius of the cylinder is r and the height is h safe S2. Dual dynamic filtering of the point cloud: Selecting a safe return point S. S3. Road inflection point-driven obstacle avoidance path generation: Using the A* algorithm, the shortest path sequence from S to the aircraft nest T is calculated as Path = [S, p1, p2, ..., T], which serves as the return waypoint sequence. This method can quickly determine a waypoint that allows the drone to ascend to a safe return altitude during inspection operations if it detects that the drone's battery level is insufficient to complete the route, ensuring the safety of the subsequent return process.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of UAV power inspection, and in particular to a method for generating a dynamic return route for a UAV in a substation. Background Art

[0002] With the rapid development of drone technology, its application in power systems is becoming increasingly widespread, particularly in substation equipment inspections. Drones, with their efficiency and flexibility, are becoming an important auxiliary tool. Currently, drone inspections within substations typically rely on drones equipped with RTK high-precision positioning to execute pre-set routes to inspect mid- and high-level equipment within the station. These routes typically include a safe flight path from the drone airport to the equipment inspection point, ensuring that the drone can fly at a safe altitude above the inspection route, thereby avoiding the risk of collision with equipment.

[0003] However, in actual operation, the low-battery return-to-home (RTH) issue is particularly critical due to limited battery life. Traditional RTH methods for drones rely primarily on simple straight-line or pre-set routes. While simple to implement, these methods pose significant safety risks in complex environments such as substations. Substations typically house a large number of high-precision electrical equipment, often with strict no-fly zones within these areas. Traditional RTH methods fail to fully account for terrain and obstacle distribution, easily causing the return path to intersect with the equipment area, creating a potential collision risk.

[0004] Therefore, how to generate a safe and efficient return route during drone power inspection operations has become a technical problem that needs to be urgently solved in the current substation drone inspection technology. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, one object of the present invention is to provide a method for generating a dynamic return route for a substation drone. This method can rapidly determine a waypoint that allows the drone to ascend to a safe return altitude when it detects that the drone's battery level is insufficient to complete the route during an inspection operation, thereby ensuring the safety of the subsequent return process.

[0006] In order to solve the above problems, the present invention provides a method for generating a dynamic return route for a substation UAV, comprising the following steps:

[0007] S1. Construct a double-layer safety zone based on voltage level: In response to the low battery signal of the drone, obtain the current waypoint P n lon of longitude and latitude coordinates n ,lat n ,h n Substation voltage level, and dynamically determine the safety distance r according to the voltage level; then use Pn A safe rectangular area is constructed as the center. The bottom of the safe rectangular area is a square with a length of 2r and a height of h. safe ; Then construct an inscribed safety cylinder in the safety cuboid, the radius of the cylinder is r and the height is h safe ;

[0008] S2. Point cloud dual dynamic filtering:

[0009] Extract the point cloud dataset G inside the safe cuboid pc , when G pc If the number of points is less than 1000, select P n The safe return point S;

[0010] When G pc When the number of points is ≥ 1000, filter the point cloud dataset G inside the cylinder pc ';When G pc 'Points < 1000, then select P n The safe return point S is reached; otherwise, steps S1-S2 are iterated bidirectionally along the inspection route;

[0011] S3. Obstacle avoidance path generation driven by road turning points:

[0012] Based on the pre-marked equipment no-fly polygon set and patrol road turning point set G={g1,g2,...,g n}, the safe return point S obtained by S2 to each g i Perform ray intersection detection on the connecting lines, and only retain line segments that do not intersect with any device polygons; then construct the retained line segments into an unweighted graph G', and use the A* algorithm to calculate the shortest path sequence Path = [S, p1, p2, ..., T] from S to the machine nest T as the return waypoint sequence.

[0013] Preferably, in S1, when constructing the safety cylinder, the axis of the cylinder coincides with the central axis of the cuboid, and the bottom circle of the cylinder is inscribed in the bottom square of the cuboid.

[0014] Preferably, in S2, when extracting points within the safe cuboid area from the point cloud, it is necessary to calculate the longitude span δlon in the east-west direction and r, and the longitude span δlat in the north-south direction with Pn as the center using the distance conversion coordinate formula;

[0015]

[0016] Where R is the radius of the earth, select the point Pc (lon c ,lat c , h c ) forms a point set G Pc :

[0017]

[0018] Preferably, in S2, when the point cloud is double-dynamically filtered, the spherical distance formula is used to calculate P n Horizontal distance from the point cloud:

[0019] L=R×arccos(sin(lat n )×sin(lat c )+cos(lat n )×cos(lat c )×cos(lon n -lon c );

[0020] Where: R is the radius of the earth, and the earth is regarded as a standard sphere;

[0021] If L≤r, the point is considered to be within the safety cylinder;

[0022] Traverse G Pc All points inside the safety cylinder are put into the new point set G Pc ', if G Pc If the number of points in the area is less than 1000, it is considered that there are no obstructions in this area, and the drone meets the direct ascent condition, and the current waypoint is used as the safe return point S.

[0023] Preferably, S3 includes the following sub-steps:

[0024] S3.1. Constructing a 2D map device area model:

[0025] On a two-dimensional electronic map of the substation, polygons are used to mark all key equipment areas, such as transformers, lightning arresters, and circuit breakers. Each polygon is represented by a set of ordered coordinate points, which are used to define the boundaries of the equipment's safe no-fly zone.

[0026] S3.2. Marking road turning point information:

[0027] Mark all the turning points on the patrol route within the substation, i.e., the key points where the route turns or branches; store them as a point set G = {g1, g2, ..., g n}, where g i =(x i ,y i );

[0028] S3.3. Initial path generation:

[0029] Based on the selected drone breakpoint safety return point S, it moves from the starting point S to each turning point g iConnect the lines to form candidate path segments; for each line segment L(S, g i ), determine whether it intersects with any device area polygon; use ray method or intersection detection algorithm to perform geometric judgment; if the line segment intersects with any device area, discard the line segment;

[0030] S3.4. Construct a feasible path network and find the shortest path:

[0031] All the unremoved path segments are grouped into a graph G', where the nodes are S and the inflection points, and the edges are the safe path segments. Use the Dijkstra algorithm or the A* algorithm to find the shortest path from the starting point S to the nest position T in the graph G'. Output the path point sequence Path = [S, p1, p2, ..., T] as the return waypoint sequence.

[0032] The advantages of the present invention compared with the prior art are:

[0033] (1) The present invention combines three-dimensional point cloud data in real time during flight to construct a safe ascending cylinder model, accurately identifying whether there are obstacles blocking the path, and ensuring that the safe return waypoint selected by the drone when the battery is low has sufficient vertical safety space. This method not only improves the drone's ability to respond to emergencies in a complex substation environment, but also ensures that the drone can safely return to the nest, significantly reducing the risk of equipment damage due to collision.

[0034] (2) The present invention can dynamically generate the shortest return path that avoids high-voltage equipment by marking the equipment area and patrol road turning points on a two-dimensional map and combining it with a path intersection detection algorithm. This method can not only effectively avoid no-fly zones such as high-voltage equipment, but also dynamically adjust the return path according to the actual terrain, thereby improving the flexibility and adaptability of path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 Schematic diagram of the rising safety cuboid and safety cylinder in the present invention.

[0037] Figure 2 This is a flow chart of the method for selecting safe return waypoints for substation UAV inspection breakpoints based on point cloud data of the present invention.

[0038] Figure 3The figure is a flow chart of the method for dynamically planning a safe return route according to the present invention. DETAILED DESCRIPTION

[0039] The following describes in detail embodiments of the present application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application and are not to be construed as limiting the present application.

[0040] In the description of this application, it should be noted that, unless otherwise expressly specified or limited, the terms "installed," "connected," and "connected" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; they can refer to internal communication between two components or the interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0041] The present invention will be described in further detail below with reference to the accompanying drawings.

[0042] Combine Figures 1 to 3 , a method for generating a dynamic return route for a substation UAV of the present invention,

[0043] The technical solution for achieving the first object of the present invention is:

[0044] (1) Determine the scope of safety inspection for the return route

[0045] When using drones for inspection inside a substation, it is necessary to follow the substation inspection safety distance r, which is as follows:

[0046] Voltage level (kV) 110 / 220 330 500 750 1000 Safety distance (m) 1 1.6 1.8 2.9 4.4

[0047] (2) Establish a safe rectangular area for ascent

[0048] Get the coordinates P of the current execution waypoint n (lon n ,lat n , h n ), where h n As the bottom height of the safety cuboid, obtain the return safety height h safe , as the top height of the safety cuboid, the bottom of the cuboid is a square, with P n As the center, the distance r is extended in the four directions of east, west, south and north to form a square area.

[0049] (3) Extract points within the safe cuboid area from the point cloud

[0050] By using the distance conversion coordinate formula, the longitude span δlon when extending from east to west and from a distance r, and the longitude span δlat when extending from north to south and from a distance r are calculated with Pn as the center.

[0051]

[0052] Where R is the radius of the Earth. Since the calculation range is small, the Earth can be approximately regarded as a standard sphere.

[0053] Filter the points Pc(lon) that meet all the following conditions in the point cloud data c ,lat c , h c ) forms a point set G Pc :

[0054]

[0055] If G Pc If the number of points in the waypoint is less than 1000, it is considered that no equipment can ascend directly in this area. Then the current waypoint can be used as the safe return waypoint S.

[0056] (4) Extract points within the safe cylindrical area from the safe rectangular area

[0057] If G Pc If the number of points in the image exceeds 1000, further screening is required in this area, that is, the bottom square area is used as the bottom surface to construct an inscribed safety cylinder. The specific method is to calculate the horizontal distance L from the Pc to Pn coordinates.

[0058] L=R×arccos(sin(lat n )×sin(lat c )+cos(lat n )×cos(lat c )×cos(lon n -lon c );

[0059] Where R is the radius of the Earth. Since the calculation range is small, the Earth can be approximately regarded as a standard sphere.

[0060] If L is less than or equal to r, the point is considered to be inside the safety cylinder.

[0061] Traverse G Pc All points inside the safety cylinder are put into the new point set G Pc '. If G PcIf the number of points in the waypoint is less than 1000, it is considered that no equipment can ascend directly in this area. Then the current waypoint can be used as the safe return waypoint S.

[0062] (5) Repeat until a safe return point is found.

[0063] If G Pc If the number of points within the ' is still greater than 1000, then execute (1)-(4) in sequence upstream and downstream waypoints with the current route as the center until a waypoint that can be safely ascended is found, which will be used as a safe return waypoint. During the return process, the route from the current waypoint to the return waypoint is first executed along the route and then ascended to a safe altitude for return.

[0064] If no waypoint is found to return to after looping to the beginning or end of the route, the previous or subsequent route of the route will be executed to return to the destination.

[0065] A second objective of the present invention is to provide a method for dynamic return route planning. This method involves marking the equipment area on a two-dimensional map of the substation, using a polygonal box. All road turning points—that is, points where patrol routes turn or branch—are then marked. When the drone's battery is low and returning to home, the turning points are connected sequentially, starting from the drone's current location. Any lines intersecting the equipment area are removed, and then the shortest path among all remaining lines is found as a safe return route.

[0066] The technical solution for achieving the second object of the present invention is:

[0067] (1) Constructing a 2D map device area model

[0068] On a two-dimensional electronic map of the substation, use polygons to mark all critical equipment areas, such as transformers, lightning arresters, and circuit breakers. Each polygon, represented by a set of ordered coordinate points, defines the boundaries of the equipment's safe no-fly zone. For example, the boundary of equipment A is a quadrilateral with vertex coordinates: P1(x1, y1), P2(x2, y2), P3(x3, y3), and P4(x4, y4).

[0069] (2) Marking road turning point information

[0070] Mark all the turning points on the patrol route within the substation, i.e., the key points where the route turns or branches; store them as a point set G = {g1, g2, ..., g n}, where g i =(x i ,y i ).

[0071] (3) Initial path generation

[0072] Based on the breakpoint safety return point S of the drone selected in the invention point 1, the drone moves from the starting point S to each turning point g in turn.i Connect the lines to form candidate path segments; for each line segment L(S, g i ) to determine whether it intersects with any device region polygon. Use the ray method or intersection detection algorithm to perform geometric judgment; if the line segment intersects any device region, it is discarded.

[0073] (4) Construct a feasible path network and find the shortest path

[0074] All the unremoved path segments are grouped into a graph G', where the nodes are S and the inflection points, and the edges are the safe path segments. Use the Dijkstra algorithm or the A* algorithm to find the shortest path from the starting point S to the nest position T in the graph G'. Output the path point sequence Path = [S, p1, p2, ..., T] as the return waypoint sequence.

[0075] It should be noted that:

[0076] 1. The method for obtaining point cloud data in the present invention adopts mature means in the existing technology, and preferably includes the following two methods: 1. At present, most substations have completed the construction of point cloud models, and the point cloud data of the substations that have been built can be directly used; 2. Through the airborne laser scanning device of the DJI L2 drone, a panoramic three-dimensional point cloud model of the substation is constructed.

[0077] 2. Current Point P in the Present Invention n Longitude and latitude coordinates lon n ,lat n , h n The acquisition method adopts the mature means in the existing technology, in which P n It is a set of waypoints in the preset route. The waypoint information stores information such as coordinates and altitude, and the longitude and latitude coordinates of the current location of the drone can also be collected by the positioning unit carried by the existing drone; Positioning P n Then use P n Used as a reference to determine the position and range of the safety rectangle and safety cylinder.

[0078] 3. In this invention, 1000 is defined as G pc and G pc The reason for the judgment threshold of points is:

[0079] First, during point cloud acquisition, noise is generated due to factors such as dust and other impurities in the environment and equipment accuracy. However, the number of noise points is relatively small. Furthermore, in conventional point clouds, such as those collected by DJI's L2, the minimum length of power transmission and distribution line conductors within a set safety rectangle or cylinder can still reach tens of thousands of points. Therefore, setting 1,000 points is a reasonable setting based on experience, fully considering the characteristics of point clouds, and actual conditions.

[0080] Finally, all parts not fully described in the present invention adopt mature products and mature technical means in the existing technology.

[0081] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. A method for generating a dynamic return route for a substation drone, characterized in that: The following steps are involved: S1. Construct a double-layer safety zone based on voltage level: In response to the low battery signal of the drone, obtain the current waypoint P n Longitude and latitude coordinates lon n ,lat n , h n The substation voltage level is determined dynamically according to the voltage level; then the safety distance r is determined dynamically according to the voltage level; n A safe rectangular area is constructed as the center. The bottom of the safe rectangular area is a square with a length of 2r and a height of h. safe ; Then construct an inscribed safety cylinder in the safety cuboid, the radius of the cylinder is r and the height is h safe ; S2. Point cloud dual dynamic filtering: Extract the point cloud dataset G inside the safe cuboid pc , when G pc If the number of points is less than 1000, select P n The safe return point S; When G pc When the number of points is ≥ 1000, filter the point cloud dataset G inside the cylinder pc ';When G pc 'Points < 1000, then select P n The safe return point S; Otherwise, iterate and execute steps S1-S2 bidirectionally along the inspection route; S3. Obstacle avoidance path generation driven by road turning points: Based on the pre-marked equipment no-fly polygon set and patrol road turning point set G = {g1, g2, ..., g n }, the safe return point S obtained by S2 to each g i Perform ray intersection detection on the lines and only keep the line segments that do not intersect with any device polygon; The retained line segments are then constructed into an unweighted graph G', and the A* algorithm is used to calculate the shortest path sequence Path = [S, p1, p2, ..., T] from S to the machine nest T as the return waypoint sequence.

2. The method for generating a dynamic return route for a substation drone according to claim 1, characterized in that: In the above S1, when constructing the safety cylinder, the axis of the cylinder coincides with the central axis of the cuboid, and the bottom circle of the cylinder is inscribed in the bottom square of the cuboid.

3. The method for generating a dynamic return route for a substation UAV according to claim 1, characterized in that: In S2, when extracting points within the safe rectangular area from the point cloud, it is necessary to calculate the longitude span δlon in the east-west direction and r, and the longitude span δlat in the north-south direction with Pn as the center through the distance conversion coordinate formula; Where R is the radius of the earth, select the point Pc (lon c ,lat c , h c ) forms a point set G Pc :

4. The method for generating a dynamic return route for a substation UAV according to claim 3, characterized in that: In S2, when the point cloud is double-dynamically filtered, the spherical distance formula needs to be used to calculate P n Horizontal distance from the point cloud: L=R×arccos(sin(lat n )×sin(lat c )+cos(lat n )×cos(lat c )×cos(lon n -lon c ); Where: R is the radius of the earth, and the earth is regarded as a standard sphere; If L≤r, the point is considered to be within the safety cylinder; Traverse G Pc All points inside the safety cylinder are put into the new point set G Pc ', if G Pc If the number of points in the area is less than 1000, it is considered that there are no obstructions in this area, and the drone meets the direct ascent condition, and the current waypoint is used as the safe return point S.

5. The method for generating a dynamic return route for a substation UAV according to claim 3, characterized in that: The S3 includes the following sub-steps: S3.

1. Constructing a 2D map device area model: On a two-dimensional electronic map of the substation, polygons are used to mark all critical equipment areas, including transformers, lightning arresters, and circuit breakers. Each polygon is represented by a set of ordered coordinate points, which are used to define the boundaries of the equipment's safe no-fly zone. S3.

2. Marking road turning point information: Mark all the turning points on the patrol route within the substation, i.e., the key points where the route turns or branches; store them as a point set G = {g1, g2, ..., g n }, where g i =(x i ,y i ); S3.

3. Initial path generation: Based on the selected drone breakpoint safety return point S, it moves from the starting point S to each turning point g i Connect the lines to form candidate path segments; for each line segment L(S, g i ), determine whether it intersects with any device area polygon; Use raycasting or intersection detection algorithms for geometric judgment; if a line segment intersects any device area, the line segment is discarded; S3.

4. Construct a feasible path network and find the shortest path: All the unremoved path segments are grouped into a graph G', where the nodes are S and the inflection points, and the edges are safe path segments. Use the Dijkstra algorithm or the A* algorithm to find the shortest path from the starting point S to the machine nest position T in the graph G'. Output the path point sequence Path = [S, p1, p2, ..., T] as the return waypoint sequence.

Citation Information

Patent Citations

  • Power transmission line unmanned aerial vehicle nest inspection route planning method and system

    CN116225074A

  • Method and device for planning flight path for unmanned aerial vehicle

    WO2017211029A1