A Path Planning and Obstacle Avoidance Method for Emergency Unmanned Aerial Vehicles in Watersheds

Abstract

This invention provides a path planning and obstacle avoidance method for watershed emergency unmanned aerial vehicles (UAVs), relating to the field of UAV technology. The method includes: constructing a task map centered on the UAV platform and constrained by obstacle coverage as a radius; randomly selecting a task node from a local task region and scanning the local task region from the randomly selected node as the starting point until all task nodes in the local task region have been scanned, thus obtaining a first path; obtaining obstacle circles on the first path; determining an offset path set based on the obstacle circles according to a preset obstacle avoidance strategy; correcting the first path based on the offset path set to obtain a second path set; and determining a target path from the second path set, with the goal of maximizing the total number of task nodes and constrained by energy consumption and payload. The method of this invention plans a more optimized path, effectively improving the utilization rate of UAVs to meet the needs of emergency missions.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a path planning and obstacle avoidance method for a watershed emergency UAV. Background Technology

[0002] my country has a vast territory and numerous rivers, whose abundant water resources are of great significance to our economic development. These water resources can be used for hydropower generation, agricultural production, transportation, and more. However, the terrain of river basins is complex, with many high mountains and deep valleys on both sides of river channels and reservoirs, making them prone to geological disasters such as mudslides. Once a large-scale landslide occurs on the slopes of a river or reservoir, the resulting blockage of the river channel by earth and rocks can easily form a barrier lake, which can have a significant impact on the power generation of hydropower stations and the safety of nearby residents.

[0003] In recent years, drone technology has developed rapidly. Due to its flexibility, speed and lack of constraints from terrain and other conditions, it has been gradually applied to river basin surveying and 3D modeling. However, existing drone path planning methods for river basins cannot meet the needs of emergency rescue in river basins.

[0004] Therefore, developing a path planning and obstacle avoidance method for watershed emergency drones has become an urgent need. Summary of the Invention

[0005] To address the aforementioned problems in the prior art, this invention provides a path planning and obstacle avoidance method for a watershed emergency drone.

[0006] The technical solution adopted in this invention is as follows:

[0007] On the one hand, a path planning and obstacle avoidance method for a watershed emergency drone is provided, including: constructing a task map with the drone platform as the center and the obstacle coverage rate as the radius constraint, wherein the task map includes multiple local task regions, and obstacles in the local task regions are marked by obstacle circles;

[0008] Randomly select a task node from the local task area, and scan the local task area from the randomly selected task node as the starting point until all task nodes in the local task area have been scanned to obtain the first path, wherein the first path is a straight path with the UAV platform as the starting and ending point and the task node as the intermediate point.

[0009] Obtain obstacle circles on the first path, determine an offset path set based on the obstacle circles according to a preset obstacle avoidance strategy, and correct the first path based on the offset path set to obtain a second path set;

[0010] With the goal of maximizing the total number of task nodes and constraints on energy consumption and load capacity, the target path is determined from the second path set.

[0011] Preferably, a task graph is constructed with the UAV platform as the center and obstacle coverage as the radius constraint, including:

[0012] Acquire geographic information, weather information, and mission information, and input the geographic information, weather information, and mission information into a pre-trained altitude prediction model to obtain the optimal flight altitude;

[0013] Based on the optimal height, extract two-dimensional images of each obstacle from the three-dimensional watershed map, and fit an obstacle circle to the two-dimensional image of each obstacle;

[0014] Add the obstacle circles to the task node graph to obtain the initial task graph.

[0015] Preferably, the task graph is constructed with the UAV platform as the center and obstacle coverage as the radius constraint, and further includes:

[0016] With the UAV platform as the center and obstacle coverage and task node coverage as the radius constraints, the initial task map is divided into a circular task region and multiple annular task regions to obtain the task map.

[0017] Preferably, the radius constraint based on obstacle coverage and task node coverage includes: the difference between obstacle coverage and task node coverage between each task area is within a preset range.

[0018] Preferably, scanning the local task graph starting from a randomly selected task node includes:

[0019] When the local task area is a circular area, the first task node is used as the starting point, and the scanned task nodes are added to the initial path one by one in a ray-shaped scanning manner to obtain the first path;

[0020] When the local area is a ring-shaped area, the first task node is used as the starting point, and the scanned task nodes are added to the initial path one by one in a wave-shaped scanning manner to obtain the first path.

[0021] Preferably, the initial path includes a drone platform as the start and end point and a first task node as the starting point.

[0022] Preferably, obtaining obstacle circles on the first path and determining an offset path set based on the obstacle circles according to a preset obstacle avoidance strategy includes:

[0023] Outside the barrier circle, several concentric circles of the barrier circle are constructed, each concentric circle does not intersect with other barrier circles, and the radius difference between adjacent concentric circles is a preset value;

[0024] When the tangent of the concentric circle passes through the tangent of the next task node and the concentric circle, the point of tangency between the tangent and the concentric circle is taken as the offset point, and the offset path is determined based on the offset point. The offset path includes an arc path.

[0025] Repeat the above steps until you have obtained the offset paths corresponding to all concentric circles, thus obtaining the offset path set.

[0026] Preferably, the second path set is obtained by correcting the first path according to the offset path set, including:

[0027] The first path is corrected based on the set of offset paths;

[0028] Repeat the above steps until there are no obstacles between two adjacent task nodes, thus obtaining the second path set.

[0029] Preferably, with the goal of maximizing the total number of task nodes and constraints on energy consumption and load capacity, the target path determined from the second path set includes:

[0030] The objective function is defined as follows:

[0031]

[0032] In the formula, H is the set of all task nodes in the second path, and F γ If task node γ exists in the target path, then F... γ =1, if task node γ does not exist in the target path, then F γ =0;

[0033] The energy consumption and load-bearing constraints are determined as follows:

[0034]

[0035]

[0036]

[0037] In the formula, F represents the set of task nodes in the target path, and N... γ E represents the load requirement value of task node γ. γ L represents the energy consumption value from task node γ to task node γ+1, n represents the number of straight paths in the path corresponding to task node γ to task node γ+1, and L represents the energy consumption value. γi L represents the length of the γi-th straight path, m represents the number of arc paths in the path from task node γ to task node γ+1, and L represents the length of the straight path. γj R represents the length of the path of the γj-th arc. γjLet represent the radius of the γj-th arc path, ΔN represent the reduction in rated load, and represent the total load requirement from the first task node to task node γ. a, b, c, and d are constants.

[0038] This facilitates the use of a pre-defined algorithm to determine the target path from the second path set.

[0039] On the other hand, a computer storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the method as described in any of the preceding claims.

[0040] The beneficial effects of this invention are as follows: This invention provides a path planning and obstacle avoidance method for a watershed emergency drone. It constructs a task map centered on the drone platform and constrained by obstacle coverage rate as a radius. The task map includes multiple local task regions, where obstacles are marked with obstacle circles. A task node is randomly selected from each local task region, and the scan continues from this starting point until all task nodes in the local task region have been scanned, obtaining a first path. This first path is a straight line with the drone platform as the starting and ending point and the task nodes as intermediate points. Obstacle circles are obtained along the first path. An offset path set is determined based on the obstacle circles according to a preset obstacle avoidance strategy. The first path is then corrected based on the offset path set to obtain a second path set. With the goal of maximizing the total number of task nodes and constraints on energy consumption and payload, a target path is determined from the second path set. This invention provides a more optimized path plan, effectively improving the utilization rate of the drone to meet emergency mission requirements. Attached Figure Description

[0041] Figure 1 The present invention provides a flowchart of a path planning and obstacle avoidance method for a watershed emergency drone. Detailed Implementation

[0042] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0043] Example

[0044] like Figure 1 As shown, this embodiment of the invention provides a path planning and obstacle avoidance method for a watershed emergency drone, including:

[0045] Step 1: Construct a mission map with the UAV platform as the center and the obstacle coverage rate as the radius constraint. The mission map includes multiple local mission regions, and obstacles in the local mission regions are marked with obstacle circles.

[0046] In this embodiment, a mission map is constructed with the UAV platform as the center and obstacle coverage as the radius constraint. This includes: acquiring geographic information, weather information, and mission information; inputting the geographic information, weather information, and mission information into a pre-trained altitude prediction model to obtain the optimal flight altitude; extracting two-dimensional maps of each obstacle from the three-dimensional watershed map based on the optimal altitude; fitting an obstacle circle to the two-dimensional map of each obstacle; and adding the obstacle circles to the mission node map to obtain the initial mission map.

[0047] In this embodiment, a task map is constructed with the UAV platform as the center and obstacle coverage as the radius constraint. The process also includes dividing the initial task map into a circular task region and multiple annular task regions with the UAV platform as the center and obstacle coverage and task node coverage as the radius constraints, thereby obtaining the task map. The difference between obstacle coverage and task node coverage between each task region is within a preset range.

[0048] This embodiment ensures task balance by dividing different regions.

[0049] Step 2: Randomly select a task node from the local task area, and scan the local task area with the randomly selected task node as the starting point until all task nodes in the local task area have been scanned to obtain the first path, wherein the first path is a straight path with the UAV platform as the starting and ending point and the task node as the intermediate point.

[0050] In this embodiment, scanning the local task map with a randomly selected task node as the starting point includes: when the local task area is a circular area, adding the scanned task nodes one by one to the initial path in a ray-like scanning manner, starting from the first task node, to obtain the first path; when the local area is an annular area, adding the scanned task nodes one by one to the initial path in a wave-like scanning manner, starting from the first task node, to obtain the first path.

[0051] In this embodiment, the initial path includes a drone platform as the start and end point and a first task node as the starting point.

[0052] By setting different scanning methods for different regions, we can better meet the characteristics of each region and improve the efficiency of task node discovery and the accuracy of path planning. Selecting the appropriate scanning method based on the specific task scenario helps improve the efficiency of path planning and the success rate of task execution.

[0053] In this embodiment, correcting the first path according to the offset path set to obtain the second path set includes: correcting the first path according to the offset path set; repeating the above steps until there are no obstacles between two adjacent task nodes to obtain the second path set.

[0054] By correcting based on the offset path set, automatic obstacle avoidance can be achieved. In each correction, according to the preset obstacle avoidance strategy, the first path is offset to bypass the obstacle, thereby ensuring the safety and passability of the path, ensuring good continuity of the second path between adjacent task nodes, and improving the flight comfort and stability of the UAV.

[0055] Step 3: Obtain the obstacle circles on the first path, determine the offset path set according to the obstacle circles and a preset obstacle avoidance strategy, and correct the first path according to the offset path set to obtain the second path set;

[0056] In this embodiment, obtaining the obstacle circles on the first path and determining the offset path set according to the obstacle circles based on a preset obstacle avoidance strategy includes: constructing several concentric circles around the obstacle circles outside the obstacle circles, where each concentric circle does not intersect with other obstacle circles and the radius difference between adjacent concentric circles is a preset value; when the tangent of the concentric circle passes through the tangent of the next task node and the concentric circle, the point of tangency between the tangent and the concentric circle is taken as the offset point, and the offset path is determined based on the offset point, the offset path including an arc path; repeating the above steps until the offset paths corresponding to all concentric circles are obtained, thus obtaining the offset path set.

[0057] Traditional obstacle avoidance methods involve setting a certain offset angle and direction, and obtaining multiple arc paths by iterating the offset angle and direction until the obstacle is avoided. However, if the offset angle is too large, the flight radius will be too large, which may lead to energy waste or flight malfunctions. If the offset angle is too small, the turning resistance will increase, which will also lead to energy waste. Therefore, this embodiment provides a new obstacle avoidance method, which constructs concentric circles with reference to the radius of each obstacle circle, and determines the corresponding offset path based on the concentric circles.

[0058] The method in this embodiment requires only one arc path to avoid an obstacle. It should be noted that the larger the radius, the lower the energy consumption, but if the radius is too large, the flight path will increase. Therefore, it is necessary to optimize the path corresponding to each concentric circle by considering other paths, energy consumption, and other factors to determine the target path.

[0059] It should also be noted that if two obstacle circles intersect or the straight-line distance between the nearest points is less than a preset value, the traditional method will still be used to determine the path for that segment.

[0060] Step 4: With the goal of maximizing the total number of task nodes and with energy consumption and load as constraints, determine the target path from the second path set.

[0061] In this embodiment, with the goal of maximizing the total number of task nodes and constraints on energy consumption and load capacity, the target path determined from the second path set includes:

[0062] The objective function is defined as follows:

[0063]

[0064] In the formula, H is the set of all task nodes in the second path, and F γ If task node γ exists in the target path, then F... γ =1, if task node γ does not exist in the target path, then F γ =0;

[0065] The energy consumption and load-bearing constraints are determined as follows:

[0066]

[0067]

[0068]

[0069] In the formula, F represents the set of task nodes in the target path, and N... γ E represents the load requirement value of task node γ. γ L represents the energy consumption value from task node γ to task node γ+1, n represents the number of straight paths in the path corresponding to task node γ to task node γ+1, and L represents the energy consumption value. γi L represents the length of the γi-th straight path, m represents the number of arc paths in the path from task node γ to task node γ+1, and L represents the length of the straight path. γj R represents the length of the path of the γj-th arc. γj Let represent the radius of the γj-th arc path, ΔN represent the reduction in rated load, and represent the total load requirement from the first task node to task node γ. a, b, c, and d are constants.

[0070] This facilitates the use of a pre-defined algorithm to determine the target path from the second path set.

[0071] The tasks of drones may include material delivery and transportation. When performing such tasks, the payload requirement for each task node is different, which is determined according to the actual situation of the task node. As the task progresses, the payload of the drone will gradually decrease, and the unit energy consumption required for the drone's flight will also gradually decrease. It can be seen that the energy consumption of the drone is closely related to its payload. In addition, since drones need to maintain good attitude stability during flight, the lateral force and lift distribution of the aircraft will change when flying in an arc, which increases the difficulty of control and adjustment and requires more energy to maintain stable flight. Furthermore, the arc path will introduce more curved flight and turning maneuvers, which will increase convective drag. The drone must exert more thrust on these additional drags, which will also increase the drone's energy consumption. The smaller the radius, the greater the increase in energy consumption.

[0072] Therefore, when constructing energy consumption constraints, this embodiment considers not only the length of the arc path and the straight path, but also the radius of the arc path and the load requirement value of each task node. By using the objective function and constraints constructed in this embodiment, the energy consumption of the UAV during the execution of the task can be calculated more accurately, thereby better planning the path and avoiding situations such as the task not being completed or the task taking a long time.

[0073] It should also be noted that different drones have different maximum energy consumption and maximum payload. The method of this embodiment can realize the combination of flight paths of different drones. After the target path is determined according to the method of the above embodiment, the remaining path can be used as a new second path to continue to determine the corresponding target path until all task nodes in the second path have been completed.

[0074] When there are a large number of available drones, different drones can be dispatched to execute different target paths in the same local task area. When there are fewer available drones, different drones can be dispatched to execute different target paths in different local task areas. It should be noted that if the energy consumption and payload requirements for determining the target path in two adjacent local areas are much less than the maximum energy consumption and maximum payload of the drones, the target paths in the two adjacent local task areas can be combined to obtain a new second path. From the second path, a new target path can be determined, thereby achieving joint flight.

[0075] Those skilled in the art will understand that although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and modifications of the invention fall within the scope of the machine equivalents of the claims, the invention also intends to include these modifications and modifications.

Claims

1. A path planning and obstacle avoidance method for a watershed emergency unmanned aerial vehicle (UAV), characterized in that, include: A mission map is constructed with the UAV platform as the center and the obstacle coverage rate as the radius constraint. The mission map includes multiple local mission regions, and obstacles in the local mission regions are marked by obstacle circles. Randomly select a task node from the local task area, and scan the local task area from the randomly selected task node as the starting point until all task nodes in the local task area have been scanned to obtain the first path, wherein the first path is a straight path with the UAV platform as the starting and ending point and the task node as the intermediate point. Obtain obstacle circles on the first path, determine an offset path set based on the obstacle circles according to a preset obstacle avoidance strategy, and correct the first path based on the offset path set to obtain a second path set; With the goal of maximizing the total number of task nodes and energy consumption and load as constraints, the target path is determined from the second path set; The task graph is constructed with the drone platform at the center and obstacle coverage as the radius constraint, including: Using the UAV platform as the center and obstacle coverage and task node coverage as the radius constraints, the initial task map is divided into a circular task area and multiple annular task areas to obtain the task map. The differences in obstacle coverage and task node coverage between each task area are all within the preset range.

2. The method according to claim 1, characterized in that, Construct a task graph with the drone platform as the center and obstacle coverage as the radius constraint, including: Acquire geographic information, weather information, and mission information, and input the geographic information, weather information, and mission information into a pre-trained altitude prediction model to obtain the optimal flight altitude; Based on the optimal flight altitude, two-dimensional images of each obstacle are extracted from the three-dimensional watershed map, and an obstacle circle is fitted to the two-dimensional image of each obstacle. Add the obstacle circles to the task node graph to obtain the initial task graph.

3. The method according to claim 1, characterized in that, The scan in the local task graph, starting from a randomly selected task node, includes: When the local task area is a circular area, the first task node is used as the starting point, and the scanned task nodes are added to the initial path one by one in a ray-shaped scanning manner to obtain the first path; When the local task area is a ring-shaped area, the first task node is used as the starting point, and the scanned task nodes are added to the initial path one by one in a wave-shaped scanning manner to obtain the first path.

4. The method according to claim 3, characterized in that, The initial path includes the drone platform as the start and end point and the first task node as the starting point.

5. The method according to claim 1, characterized in that, Obtain the obstacle circles on the first path, and determine the offset path set according to the obstacle circles and a preset obstacle avoidance strategy, including: Outside the barrier circle, several concentric circles of the barrier circle are constructed, each concentric circle does not intersect with other barrier circles, and the radius difference between adjacent concentric circles is a preset value; When the tangent between the current task node and the concentric circle passes through the tangent between the next task node and the concentric circle, the point of tangency between the two tangents and the concentric circle is taken as the offset point, and the offset path is determined based on the offset point. The offset path includes an arc path. Repeat the above steps until you have obtained the offset paths corresponding to all concentric circles, thus obtaining the offset path set.

6. The method according to claim 1, characterized in that, The first path is corrected based on the offset path set to obtain the second path set, which includes: The first path is corrected based on the set of offset paths; Repeat the above steps until there are no obstacles between two adjacent task nodes, thus obtaining the second path set.

7. The method according to claim 1, characterized in that, With the goal of maximizing the total number of task nodes, and constrained by energy consumption and load capacity, the target paths determined from the second path set include: The objective function is defined as follows: In the formula, This is the set of all task nodes in the second path. If a task node has an identifier, then the task node... If it exists in the target path, then If task node If it does not exist in the target path, then ; The energy consumption and load-bearing constraints are determined as follows: In the formula, This represents the set of task nodes in the target path. Represents task node The load requirement value, Represents task node To the task node Energy consumption value, Represents task node To the task node The number of straight paths in the corresponding path. Indicates the first The length of a straight path, Represents task node To the task node The number of arc paths in the corresponding path. Indicates the first The length of the arc path, Indicates the first The radius of the arc path, This indicates the reduction in rated load, and represents the distance from the first task node to the next task node. The total load-bearing requirement, It is a constant; This facilitates the use of a pre-defined algorithm to determine the target path from the second path set.

8. A computer storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 6.

Images