Mountain area unmanned aerial vehicle autonomous inspection path planning method

Abstract

The invention relates to the technical field of unmanned aerial vehicle flight path planning, and discloses a mountainous area unmanned aerial vehicle autonomous inspection path planning method and system.The method comprises the steps that according to starting point and terminal point coordinates of an unmanned aerial vehicle flight task, a map model of formulated obstacles is built, a world coordinate system and a camera coordinate system are built, and a target guiding path is generated; and node pruning and path smoothing are carried out on the path to complete a set task. According to the method, the dependence on high-end unmanned aerial vehicle equipment and flight operators is reduced, the labor intensity of the operators is reduced, the distribution line inspection efficiency and the automation and intelligence level are improved, the consumed time is shorter, a large amount of space calculation is reduced, and the method has the advantages of being high in flexibility and low in calculation cost.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) trajectory planning, and particularly to a method and system for autonomous inspection path planning of UAVs in mountainous areas. Background Art

[0002] In recent years, as a high-tech inspection tool, UAVs have been rapidly popularized and applied in the power grid industry, and the era of machine inspection for power grid inspection has fully arrived. Due to the wide coverage area of the distribution network, which is widely distributed in rural or mountainous areas with inconvenient transportation and exposed to the natural environment, the operating conditions of overhead distribution lines are extremely harsh. Traditional operation and maintenance modes mainly relying on humans generally have problems such as high safety risks, limited carrying capacity, incomplete inspections, untimely troubleshooting, and maintenance by emergency repair instead of regular maintenance. These problems have always been the weak links in distribution network work. With the continuous improvement of the performance of multi-rotor UAVs, they have shown broad application prospects in many fields. However, in actual applications, the flight environment of UAVs is very complex, and they need to fly in different terrains and environments, such as mountainous areas, forests, cities, etc. These terrains and environments pose high requirements for the flight path planning of UAVs. Summary of the Invention

[0003] The present invention provides a method and system for autonomous inspection path planning of UAVs in mountainous areas to solve the problems of a large number of sampling points, large computational amount, and long time consumption in three-dimensional space path planning in the prior art.

[0004] To achieve the above object, the present invention is realized through the following technical solutions: In the first aspect, the present invention provides a method for autonomous inspection path planning of UAVs in mountainous areas, including: S1: Obtain the starting coordinates, ending coordinates of the flight mission, and the terrain database of the mountainous area where the UAV is located, and build a map model of the terrain database according to the flight mission; S2: Establish the coordinates of the UAV starting point, the UAV's current position, and the map model in the camera coordinate system, and convert the coordinates in the camera coordinate system to the coordinates in the world coordinate system; S3: Generate a target-oriented path based on the ending coordinates of the flight mission, and optimize the target-oriented path based on the world coordinate system and the coordinates in the world coordinate system to obtain an optimized flight path; S4: Execute the flight mission based on the optimized flight path.

[0005] In the second aspect, the present application provides a system for autonomous inspection path planning of UAVs in mountainous areas, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method described in the first aspect above.

[0006] Beneficial effects: This application provides a method for autonomous inspection path planning for UAVs in mountainous areas. Based on the starting and ending coordinates of the UAV flight mission, a formulated obstacle map model is constructed, a world coordinate system and a camera coordinate system are established, a target-oriented path is generated, and node pruning and path smoothing are performed on this path to complete the set mission. This method helps reduce reliance on high-end UAV equipment and flight operators, reduces operator labor intensity, improves the efficiency of distribution line inspections, and enhances automation and intelligence. It also reduces time, eliminates a large amount of spatial calculations, and offers high flexibility and low computational costs.

[0007] In a further solution, under the map model, based on the formulated information of the starting point, the end point and the obstacles, a straight line path from the starting point to the end point is obtained. The straight line path is a path that cannot avoid obstacles and passes through obstacles, indicating the general direction for goal-oriented path planning.

[0008] In a further solution, the new path is backtracked and smoothed using a B-spline curve to make the autonomous inspection path of the drone more consistent with the flight trajectory. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 This is a flow chart of a method for autonomous inspection path planning of a UAV in mountainous areas according to a preferred embodiment of the present invention; Figure 2 Schematic diagram of vision-based goal-oriented path planning. DETAILED DESCRIPTION

[0010] The following is a clear and complete description of the technical solutions of the present invention. It should be understood that the embodiments described are only a portion of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0011] Unless otherwise defined, the technical or scientific terms used in the present invention shall have the usual meanings understood by persons of ordinary skill in the field to which the present invention belongs. The words "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "a" do not indicate a quantity limitation, but rather indicate the existence of at least one. Words such as "connected" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.

[0012] Please refer to Figure 1 , a method for autonomous inspection path planning of an unmanned aerial vehicle (UAV) in mountainous areas provided by this application includes: S1: Obtain the starting coordinates, ending coordinates of a flight mission, and the terrain database of the mountainous area where the UAV is located, and build a map model of the terrain database according to the flight mission; S2: Establish the coordinates of the UAV starting point, the UAV's location, and the map model in the camera coordinate system, and convert the coordinates in the camera coordinate system to the coordinates in the world coordinate system; The body coordinate system and the camera coordinate system are established to achieve UAV flight and information collection. These two coordinate systems ultimately need to be converted to the coordinates in the world coordinate system through matrix operations.

[0013] S3: Generate a target-oriented path based on the ending coordinates of the flight mission, and optimize the target-oriented path based on the world coordinate system and the coordinates in the world coordinate system to obtain an optimized flight path; S4: Execute the flight mission based on the optimized flight path.

[0014] The above-mentioned method for autonomous inspection path planning of an unmanned aerial vehicle in mountainous areas aims at the problems of many sampling points, large computational amount, and long time consumption in three-dimensional space path planning in the prior art, and proposes a method for autonomous inspection path planning of an unmanned aerial vehicle in mountainous areas. This method builds a map model of formulaic obstacles according to the starting and ending coordinates of the UAV flight mission, establishes the world coordinate system and the camera coordinate system, generates a target-oriented path, and prunes nodes and smooths the path to complete the set tasks. It helps to reduce the dependence on high-end UAV equipment and flight control personnel, reduce the labor intensity of operators, improve the inspection efficiency and automation and intelligence level of power distribution lines, takes less time, reduces a large amount of spatial calculations, has the advantages of high flexibility and low calculation cost.

[0015] Next, a complete embodiment is used to describe the steps of the above method in detail.

[0016] 1. Obtain the starting coordinates, ending coordinates of a flight mission, and the terrain database of the mountainous area where the UAV is located, and build a map model of the terrain database according to the flight mission.

[0017] Step1: Determine the starting coordinates and ending coordinates of the UAV according to the flight mission and initial conditions of the UAV; Step2: Generate a digital topographic map of the target area according to the terrain data in the terrain database of the mountainous area where the UAV is located, and determine the boundary of the topographic map, the height information and vertex positions of obstacles in the target area; It is worth explaining that the target area refers to the set of the entire task space (with the relative position from the starting point to the ending point, taking the starting point as the point (0, 0, 0) and the ending point as (xmax, ymax, z), and the maximum value of the z-axis can be zmax according to the position with the highest peak in the terrain database of the mountainous area). Among them, the digital topographic map is not like the topographic map of 3D modeling, but the terrain recorded and organized in a register. The GPS coordinates of the terrain are converted into coordinates in the world coordinate system for establishing the set of non-feasible regions later.

[0018] Step3: Wrap the obstacle in a semi-ellipsoid enclosure manner based on the vertex positions of the obstacle, number the obstacles, and formulate the obstacle to satisfy the following relational expression: ; Among them, is the central coordinate of the ellipse, is the semi-axis length along the axis, is the semi-axis length along the axis, is the semi-axis length along the axis; Determine the set of non-feasible regions according to the digital topographic map , satisfying the following relational expression: ; Among them, is the task space, which is determined by the relative position from the starting point to the ending point, taking the starting point as the point (0, 0, 0) and the ending point as (xmax, ymax, z), and the maximum value of the z-axis is zmax according to the position with the highest peak in the terrain database of the mountainous area, is the central coordinate of the ellipse of the th obstacle wrapped in a semi-ellipsoid enclosure manner, is the th semi-axis length of the ellipse of the th obstacle along the axis, is the th semi-axis length of the ellipse of the th obstacle along the axis, is the

[0019] th semi-axis length of the ellipse of the It is worth explaining that the entire task space (with the relative position from the starting point to the ending point, taking the starting point as the point (0, 0, 0) and the ending point as (xmax, ymax, z), and the maximum value of the z-axis can be zmax according to the position with the highest peak in the terrain database of the mountainous area) is a set, and the complement of the non-feasible region is the feasible region. When the node obtained through calculation is in the non-feasible region, it is judged as a collision.

[0020] Determine the target area based on the non-feasible area and the feasible area, and use the target area as the task space. Among them, the task space is the space established in the world coordinate system.

[0021] In this step, the most important thing is to construct a three-dimensional formulaic topographic map. To solve the path, information about the map is needed to achieve the purpose of obstacle avoidance. In mountainous areas, the terrain is rough, the vegetation is lush, and there are too many unstable factors. If you want to replicate the map, the amount of calculation required is an astronomical figure. Therefore, it is necessary to simplify the mountainous conditions. It should be clear that the purpose of constructing the map is to avoid obstacles and prevent the drone from crashing. Therefore, only a few regular geometric figures are needed to cover the mountainous terrain, and the irregular mountainous environment can be transformed into an environment that can be expressed by mathematical formulas.

[0022] 2. Establish a coordinate system model based on the starting point of the drone, the position of the drone, and the map model. The coordinate system model includes a world coordinate system and a camera coordinate system.

[0023] Step1: Based on the map model, obstacles, end point coordinates, and starting point coordinates, start the drone and hover at a fixed height h. Take the position of the drone as the origin (0, 0, h) of the camera coordinate system, the forward direction as the axis, the vertical direction as the axis, and obtain the axis according to the right-hand screw rule. The direction of the camera focus carried is coincident with the axis direction to create a camera coordinate system.

[0024] It is worth explaining that forward refers to the direction of the drone's nose and also the direction of the camera. Before the drone takes off, the body coordinate system of the drone coincides with the camera coordinate system. During the flight, the body coordinate system will change with the movement (roll, pitch, yaw), while the camera has a gimbal fixed, making its coordinate system always unchanged. The coordinate origin is the coordinate of the camera, so that the camera can identify objects more stably without shaking, and the obtained moving obstacle coordinates are more accurate.

[0025] Step2: Take the current position data of the drone as the origin (0, 0, 0) of the world coordinate system, and the three-axis directions are parallel to the camera coordinate system. Also, convert the position data of the end point of the flight mission and the vertex coordinates of the obstacles into coordinates in the current world coordinate system to create a model of the task space in the world coordinate system.

[0026] Among them, in the case where the obstacle is a dynamic obstacle, the drone identifies the obstacle through a depth binocular camera, generates feedback data according to the recognition result. The feedback data includes the coordinates of the dynamic obstacle in the camera coordinate system and the distance between the dynamic obstacle and the drone, and the drone avoids the obstacle in real time according to the feedback data.

[0027] 3. Generate a target-oriented path based on the end coordinates of the flight mission, and optimize the target-oriented path based on the world coordinate system and the camera coordinate system to obtain an optimized flight path, as Figure 2 shown.

[0028] Step1: Under the map model, based on the starting coordinates, ending coordinates of the flight mission, and the formulated information of the obstacles, obtain a straight-line path from the starting coordinates to the ending coordinates of the flight mission as the target-oriented path. The straight-line path is a path that cannot avoid obstacles and passes through obstacles; Step2: Divide the target-oriented path into n segments on average to obtain n path segments, and remove the starting point and ending point of the target-oriented path to obtain n - 1 nodes, where n = Length / step, Length is the path length, and step is the step size which can also be regarded as the precision; Step3: Among the n - 1 nodes and the non-feasible region there exists a set . Inside the , place the vertex coordinates of both ends of the path segment passing through the obstacle into two sets FP and SP respectively. p represents the number of nodes, and N is a mathematical symbol representing the set of natural numbers; Among them, when a path point appears from the feasible region in the non-feasible region, put the parent node of this path point into the set FP. When a path point appears from the non-feasible region in the feasible region, then put the child node of this path point into the set SP. Traverse the path nodes to obtain the set FP and the set SP.

[0029] It is worth explaining that there are i path points in the section of the road (path segment) that penetrates the obstacle. This section of path points is translated upward as a whole. After reaching a certain height, there will be path points inside the obstacle coming out. The points that come out replace the previous path point outside. The position of the previous point can be determined to remain unchanged, and the other path points continue to loop the previous calculation until all path points come out.

[0030] Step4: The point placed in the set FP is the entry point x p_in . Modify the p_in axis coordinate of the entry point x to to obtain a new node PA; the point placed in the set SP is the exit point x p_out . Modify the p_out axis coordinate of the exit point x to to obtain a new node PB. Connect PA as the parent node and PB as the child node to obtain a new path generated by climbing over the obstacle. After that, only judge the new path segment connected by PB and PA until the sets FP and SP are empty.

[0031] Step 5: Backtrack the new path and perform smoothing processing using B-spline curves to obtain an optimized flight path.

[0032] Among them, the axis coordinates of the starting point and the ending point rise by 5 meters in height, which can reduce interference factors.

[0033] 4. Execute the flight mission based on the optimized flight path.

[0034] In this application, the drone is equipped with a Jetson Orin NX on-board computer, which uses PID to control the flight trajectory of the drone and make the drone reach the end point. The drone is also equipped with a D435i depth binocular camera. The on-board computer needs to be connected to the D435i depth binocular camera to identify obstacles ahead and replan the path. Since the drone is used to collect photos of the wild mountain forest environment and pictures of target power towers, a large number of photos at different angles need to be taken. Therefore, for the required training effect, according to the visual effect when the drone is flying, 6 pictures need to be taken of each power tower, so that the entire circumference of the power tower can be photographed. Some scarce parts need to be photographed independently for model training. Then label the pictures and divide them into a training set, a test set and a validation set, and use YOLOv5 for training. When the model reaches the expected effect after passing through the validation set, it can be put into use. When encountering dynamic obstacles, the drone identifies the obstacles through the D435i, and returns the coordinates in the camera coordinate system and the distance from the drone. According to the transmitted data, it dodges the obstacles in real time. At the same time, the drone is equipped with an infrared obstacle avoidance function to avoid moving obstacles outside the vision.

[0035] The embodiment of the present application also provides a mountain area drone autonomous inspection path planning system, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented. This mountain area drone autonomous inspection path planning system can implement each embodiment of the above method and achieve the same beneficial effects, which will not be elaborated here.

[0036] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations based on the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field of the present invention based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art should be within the protection scope determined by the claims.

Claims

1. An autonomous inspection path planning method for unmanned aerial vehicles in mountainous areas, characterized in that Including: S1: Obtain the starting coordinates, ending coordinates of the flight mission, and the terrain database of the mountain area where the UAV is located, and build a map model of the terrain database according to the flight mission; S2: Establish the coordinates of the UAV starting point, the UAV's location, and the map model in the camera coordinate system, and convert the coordinates in the camera coordinate system into coordinates in the world coordinate system; S3: Generate a target guiding path based on the ending coordinates of the flight mission, and optimize the target guiding path based on the world coordinate system and the coordinates in the world coordinate system to obtain an optimized flight path; S4: Execute the flight mission based on the optimized flight path.

2. The autonomous inspection path planning method for mountain drones according to claim 1, wherein The S1 includes: S11: Determine the starting coordinates and ending coordinates of the UAV according to the flight mission and initial conditions of the UAV; S12: Generate a digital topographic map of the target area based on the terrain data in the terrain database of the mountain area where the UAV is located, and determine the boundary of the topographic map, the height information and vertex positions of the obstacles in the target area; S13: Wrap the obstacles in a semi-ellipsoidal enclosure manner based on the vertex positions of the obstacles, number the obstacles, and formulate the obstacles to satisfy the following relationship: ； Among them, is the central coordinate of the ellipse, is the semi-axis length along the axis, is the semi-axis length along the axis, is the semi-axis length along the axis; Determine the set of non-feasible regions based on the digital topographic map , satisfying the following relational expressions: ； Among them, is the task space, is the center coordinate of the th obstacle ellipse wrapped by the semi-ellipsoid enclosure method, is the th semi-major axis length of the obstacle ellipse along the axis, is the th semi-major axis length of the obstacle ellipse along the axis, is the th semi-major axis length of the obstacle ellipse along the axis; Determine the target area according to the non-feasible region and the feasible region, and use the target area as the mission space.

3. The autonomous inspection path planning method for drones in mountainous areas according to claim 1, wherein The S2 includes: S21: Start the drone based on the map model, obstacles, end coordinates, and start coordinates, and hover at a fixed height h. Take the position of the drone as the origin (0, 0, h) of the camera coordinate system, with the forward direction as axis, the vertical direction as axis, and obtain axis according to the right-hand screw rule. The direction of the camera focus carried is coincident with the axis direction to create a camera coordinate system; S22: Take the current position data of the UAV as the origin (0, 0, 0) of the world coordinate system, with the three-axis directions parallel to the camera coordinate system, and also convert the position data of the ending point of the flight mission and the vertex coordinates of the obstacles into coordinates in the current world coordinate system to create a model of the mission space in the world coordinate system.

4. The autonomous inspection path planning method for drones in mountainous areas according to claim 1, wherein The S3 includes: S31: Under the map model, based on the starting coordinates, ending coordinates of the flight mission, and the formulated information of the obstacles, obtain a straight-line path from the starting coordinates of the flight mission to the ending coordinates as the target guiding path, and the straight-line path is a path that cannot avoid obstacles and passes through obstacles; S32: Divide the target guiding path into n segments on average to obtain n path segments, and remove the starting point and ending point of the target guiding path to obtain n - 1 nodes, where n = Length / step, Length is the path length, and step is the step size; S33: Among the n - 1 nodes and the non - feasible region there exists a set , within the , the vertex coordinates of both ends of the path segment passing through the obstacle are respectively placed in two sets FP and SP, p represents the number of nodes, and N is a mathematical symbol representing the set of natural numbers; S34: The points placed in the set FP are the entry points x p_in , and all the entry points x p_in 's axis coordinates are modified to , obtaining a new node PA; the points placed in the set SP are the exit points, and all the exit points x p_out 's axis coordinates are modified to , obtaining a new node PB. Connect PA as the parent node and PB as the child node to obtain a new path generated by crossing the obstacle. Then, only judge the new path segment connected by PB and PA until the sets FP and SP are empty; S35: Backtrack the new path and perform smoothing processing using a B-spline curve to obtain an optimized flight path.

5. The autonomous inspection path planning method for mountain drones according to claim 4, characterized in that The S33 includes: When a path point appears from the feasible region in the non-feasible region, put the parent node of the path point into the set FP, and when a path point appears from the non-feasible region in the feasible region, put the child node of the path point into the set SP, traverse the path points to obtain the set FP and the set SP; where each path segment includes multiple path points.

6. The method for autonomous inspection path planning of an unmanned aerial vehicle in mountainous areas according to claim 2, wherein In the case where the obstacle is a dynamic obstacle, the UAV identifies the obstacle through a depth binocular camera, generates feedback data according to the recognition result, and the feedback data includes the coordinates of the dynamic obstacle in the camera coordinate system and the distance between the dynamic obstacle and the UAV, and the UAV avoids the obstacle in real time according to the feedback data.

7. An autonomous inspection path planning system for drones in mountainous areas, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6 above.

Images