A hybrid planning method and system for three-dimensional trajectory of unmanned aerial vehicles in distribution network

Abstract

The present invention discloses a method and system for hybrid planning of three-dimensional tracks of unmanned aerial vehicles in distribution network lines. The method includes modeling and constructing a map model; determining a random track point in the track planning space according to the target selection strategy, searching through the set track point set, and finding the track point in the track point set that is closest to the random track point; taking the closest track point as the starting point, determining a new track point according to an adaptive step size; judging whether there is an obstacle between the new track point and the closest track point; if there is no obstacle, adding the new track point to the track point set and generating a new unmanned aerial vehicle track route; if there is an obstacle, randomly generating a random track point in the track planning space again; repeating the above steps until the generated new unmanned aerial vehicle track route reaches the final target point. The solution provided by the present invention greatly improves the track planning capability of unmanned aerial vehicles and can better meet the needs of power inspection of distribution networks in mountainous areas.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution network line inspection, and in particular to a method and system for hybrid planning of three-dimensional flight paths of unmanned aerial vehicles (UAVs) in power distribution network lines. Background Art

[0002] In recent years, drones have been rapidly promoted and applied in the power grid industry as a high-tech inspection tool. The era of drone inspection for power grid inspection has arrived. Because the distribution network covers a wide area and is widely distributed in rural or mountainous areas with inconvenient transportation. Due to exposure to the natural environment, the operating conditions of overhead distribution lines are very poor. The traditional human-based inspection mode generally has high safety risks, limited carrying capacity, inadequate inspections, untimely inspections, and maintenance by rushing. These problems have always been the weak link in the distribution network work. Although with the continuous improvement of the performance of multi-rotor drones, they have shown broad application prospects in many fields, but in actual applications, the flight environment of drones is very complex. Drones need to fly in different terrains and environments, such as mountains, forests, cities, etc. These terrains and environments place high demands on the flight path planning of drones. Summary of the invention

[0003] In order to solve the technical problems of many sampling points, large amount of calculation and long time consumption in the existing three-dimensional spatial path planning of distribution networks, an embodiment of the present invention provides a three-dimensional trajectory hybrid planning method and system for unmanned aerial vehicle lines in distribution networks.

[0004] The technical solution of the embodiment of the present invention is achieved as follows:

[0005] The embodiment of the present invention provides a method for hybrid planning of three-dimensional tracks of unmanned aerial vehicles in a distribution network line, the method comprising: using a three-dimensional grid map to model the mountainous environment of the distribution network and construct a map model; each grid in the three-dimensional grid map represents a volume element in the mountainous environment of the distribution network; based on the map model, determining the starting point, the final target point and the track planning space of the track planning; determining a random track point in the track planning space according to a target selection strategy, searching through a set track point set, and finding the track point in the track point set that is closest to the random track point; taking the track point with the closest distance as the starting point, determining a new track point according to an adaptive step size; judging whether there is an obstacle between the new track point and the track point with the closest distance; if there is no obstacle, adding the new track point to the track point set and generating a new unmanned aerial vehicle track route; if there is an obstacle, randomly generating a random track point in the track planning space again; repeating the above steps until the generated new unmanned aerial vehicle track route reaches the final target point.

[0006] In one embodiment, a random track point is determined in the track planning space according to a target selection strategy, including: randomly generating a preset number of random track points in the track planning space; calculating the probability of each of the random track points based on the target selection strategy; and selecting a final random track point from the preset number of random track points according to the probability of the random track point and the randomly generated random probability.

[0007] In one embodiment, calculating the probability of each of the random track points based on the target selection strategy includes: calculating the weight of each random track point using the following calculation formula;

[0008] Formula (1)

[0009] in, is the weight of each random track point, is the node in the search tree closest to the final target point, represents a sampling point in the sampling point set, is the distance from the nearest node to the sampling point The track cost;

[0010] Calculate the probability of each random track point according to the weight of each random track point;

[0011] Formula (2)

[0012] in, is the probability of a random track point, , is a sampling point in the sampling point set, is the weight of the sampling point.

[0013] In one embodiment, selecting a final random track point from the preset random track points according to the probability of the random track point and the randomly generated random probability includes: selecting the final random track point from the preset random track points using the following calculation formula:

[0014] Formula (3)

[0015] in, is the final random track point selected, is the probability of a random track point, is a random probability.

[0016] In one embodiment, the adaptive step length is calculated by: using the following calculation formula to determine the adaptive step length:

[0017] Formula (4)

[0018] in, is the adaptive step size, To adjust the parameters, used to adjust the step size, is the number of iterations, For the current node and the target node The Euclidean distance between .

[0019] In one embodiment, determining a new track point includes: randomly generating a preset number of new track points near the nearest track point; calculating a direction vector between the nearest track point and the new track point and a direction vector between the nearest track point and a final target point, and an angle between the two direction vectors; adjusting a probability of the new track point based on the direction vector between the nearest track point and the final target point, and an angle between the two direction vectors; and selecting a new track point from the preset number of new track points based on the adjusted probability of the new track point.

[0020] In one embodiment, calculating the direction vector between the closest track point and the new track point, the direction vector between the closest track point and the final target point, and the angle between the two direction vectors includes: calculating the direction vector between the closest track point and the new track point, the direction vector between the closest track point and the final target point, and the angle between the two direction vectors using the following calculation formula:

[0021] Formula (5)

[0022] Formula (6)

[0023] Formula (7)

[0024] in, is the direction vector between the nearest track point and the new track point, is the coordinate of the nearest track point, is the coordinates of the track point; is the direction vector between the nearest track point and the final target point, is the coordinate of the final target point, is the angle between the two direction vectors.

[0025] In one embodiment, the probability of the new track point is adjusted according to the direction vector between the closest track point and the final target point, and the angle between the two direction vectors, including: adjusting the probability of the new track point using the following calculation formula:

[0026] Formula (8)

[0027] in, is the adjusted probability, is the probability before adjustment, is the target bias parameter.

[0028] An embodiment of the present invention also provides a three-dimensional trajectory hybrid planning system for UAVs in a distribution network, comprising: a processor and a memory for storing a computer program that can be run on the processor; wherein, when the processor is used to run the computer program, the steps of any one of the above methods are executed.

[0029] An embodiment of the present invention further provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0030] The method and system for hybrid planning of three-dimensional trajectory of unmanned aerial vehicle in distribution network provided by the embodiment of the present invention can improve The algorithm improves the efficiency and feasibility of trajectory planning in a long-distance global environment, avoiding falling into the dead zone of local minimum. The algorithm greatly improves the trajectory planning capability of drones and can better meet the needs of power distribution network inspection in mountainous areas.

[0031] Compared with the prior art, the present invention has the following advantages:

[0032] 1. The present invention is directed to The algorithm has low efficiency in three-dimensional environment, track quality needs to be optimized, slow convergence speed and high computational cost. The target selection strategy, target bias, adaptive step size, smoothing processing and other improvements have been made. The algorithm improves the efficiency of trajectory generation.

[0033] 2. In view of the fact that traditional drones are not capable of handling complex environments and unexpected obstacles during distribution network inspections, we have improved The algorithm enables autonomous route planning during distribution network inspection, reducing the difficulty of distribution network inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of a method for hybrid planning of three-dimensional trajectory of unmanned aerial vehicles in a power distribution network according to an embodiment of the present invention;

[0035] Figure 2 Improvement of the embodiment of the present invention Schematic diagram of the algorithm flow;

[0036] Figure 3 For the embodiment of the present invention Algorithms and improvements after optimization The algorithm generates a track comparison diagram; among them, Figure 3(a) Schematic diagram of the algorithm search process. Figure 3 (b) Schematic diagram of trajectory planning after the algorithm performs optimization processing such as target selection and target bias;

[0037] Figure 4 The embodiment of the present invention performs adaptive step size improvement The algorithm generates a track comparison diagram; among them, Figure 4 (a) Schematic diagram of trajectory planning after the algorithm performs optimization processing such as target selection and target bias. Figure 4 (b) Schematic diagram of the trajectory planning effect after the algorithm's adaptive step size optimization;

[0038] Figure 5 The embodiment of the present invention and the conventional The algorithm generates a comparative diagram of track lengths under various map environments;

[0039] Figure 6 The embodiment of the present invention and the conventional The algorithm generates a schematic diagram comparing the time of track planning in various map environments;

[0040] Figure 7 1 is a diagram showing the internal structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

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

[0042] The embodiment of the present invention provides a hybrid planning method for three-dimensional trajectory of unmanned aerial vehicles in a distribution network. Figure 1 As shown, the method includes:

[0043] Step 101: Modeling the mountainous environment of the distribution network using a three-dimensional grid map to construct a map model; each grid in the three-dimensional grid map represents a volume element in the mountainous environment of the distribution network;

[0044] Step 102: Based on the map model, determine the starting point, final destination point and track planning space of the track planning;

[0045] Step 103: Determine a random track point in the track planning space according to the target selection strategy, search through the set track point set, and find the track point in the track point set that is closest to the random track point; take the track point with the closest distance as the starting point, and determine a new track point according to the adaptive step size; determine whether there is an obstacle between the new track point and the track point with the closest distance; if there is no obstacle, add the new track point to the track point set and generate a new UAV track route; if there is an obstacle, randomly generate a random track point in the track planning space again; repeat the above steps until the generated new UAV track route reaches the final target point.

[0046] This embodiment is a traditional (Rapidly-exploring Random Trees Star, ) algorithm is improved. Algorithm-based planning of drone flight routes can reduce reliance on high-end drone equipment and flight operators, reduce the labor intensity of operators, and improve the efficiency of distribution line inspections and the level of automation and intelligence.

[0047] Specifically, traditional The algorithm searches for an optimal path from the starting point to the end point in space by random sampling and incremental method. The algorithm locates the starting point as the starting node. , and initialize the search tree ; Then set the number of iterations and randomly sample a point in the configuration space In the tree Find the The nearest node ; Then, at a preset step length, move from arrive Generate new nodes in the direction of ; Check from arrive Is the trajectory collision-free? If so, perform the following operations: Search all nodes in the neighborhood of and select the node with the lowest cost As The parent node of Add to Tree and connect and ; then through Reconnect Other nodes in the neighborhood of the node are selected to reduce the cost of trajectory planning. This algorithm has the disadvantages of low efficiency and poor trajectory quality in complex obstacle environments.

[0048] Based on this, the present invention aims to improve The algorithm's target selection strategy introduces random probability This method will make it more likely that the algorithm will conduct a comprehensive exploration in the search space and avoid the algorithm from falling into the local optimal solution to a certain extent, thus making it more likely to find the global optimal solution. Introducing target bias will improve The algorithm converges to the target area faster, reduces algorithm complexity, and improves track quality. The specific improvements are as follows:

[0049] improve The algorithm performs global three-dimensional trajectory planning. The algorithm may have a local optimal solution, which can be improved by The algorithm's target selection strategy introduces random probability , which can make the algorithm more likely to conduct a comprehensive exploration in the search space and to a certain extent avoid the algorithm from falling into the local optimal solution, thus making it more likely to find the global optimal solution. All potential target locations are covered. According to random probability to select the target point.

[0050] (1) Target selection strategy

[0051] Calculate the weight of each target point :

[0052]

[0053] in, is the node in the tree closest to the final target point, represents a sampling point in the sampling point set, is the distance from the nearest node to the sampling point The track cost.

[0054] Then, the probability of selecting each target point is calculated based on the weight of the target point :

[0055]

[0056] in, is the probability of a random track point, , is a sampling point in the sampling point set, is the weight of the sampling point.

[0057] Its cumulative probability :

[0058]

[0059] is the sampling point of the kth iteration.

[0060] Random Generation To make roulette selection, according to the cumulative probability Select the next sampling point. Introduce random probability It can make the algorithm explore the space more flexibly and comprehensively, and improve the quality of trajectory planning.

[0061] Finally, according to random probability To select the final sampling point :

[0062]

[0063] (2) Introducing target bias

[0064] This method can improve the sampling The convergence speed of the algorithm to the target area reduces the complexity of the algorithm and improves the track quality.

[0065] for The algorithm introduces a target bias parameter , giving the target a certain bias.

[0066] First, calculate the existing nodes To the next target node The direction vector and the existing nodes To the end The angle between the direction vectors of , then the direction vector from the current node to the next target node for:

[0067]

[0068] The direction vector from the current node to the end point for:

[0069]

[0070] The angle between two vectors for:

[0071]

[0072] In this way, the probability of the next target node is adjusted according to the size of the angle, so that it grows towards the target node with a greater probability at a smaller angle. is to select the next target node The probability of Is an existing node To the next target node and to the end The angle between the direction vectors. Then the adjusted probability for:

[0073]

[0074] According to the angle, the probability of selecting the next target node can be adjusted through the formula. This affects the growth direction of the tree. By adjusting the target bias parameter The size of the control target bias is achieved, so that the algorithm can reach the end point faster. .

[0075] The introduction of target bias parameters improves The convergence speed of the algorithm is improved and the complexity of the algorithm is reduced.

[0076] (3) Track smoothing

[0077] The track generated by the algorithm is composed of straight lines connecting the start and end points. The track contains sharp corners and undesirable bends, which require track smoothing. Smoothing can reduce sharp corners and sharp bends in the track, making the track smoother and more continuous.

[0078] Use cubic B-spline interpolation to generate a smoother track. Assume that a line consisting of multiple nodes has been determined. The composed track.

[0079] First, calculate the distance between nodes Assumptions yes The Euclidean distance between It can be expressed as the sum of the distances of all previous nodes:

[0080]

[0081] Then calculate the cubic B-spline interpolation coefficients, using the node and distance To calculate the coefficients of the cubic B-spline interpolation.

[0082] Then, based on the interpolation polynomial and parameterized distance, interpolation operations are performed between nodes to generate smooth track points. Representation Node 、 The interpolation polynomial between At each distance The value of is used to obtain smooth points on the track, and a curve in the form of a parametric equation is obtained:

[0083]

[0084] is a parameter, is a cubic B-spline basis function. By changing the curve, the position of any point can be obtained. By connecting the points on the smooth track, a smoother track can be obtained.

[0085] (4) Adaptive step size

[0086] The computational complexity of the algorithm in a complex environment grows geometrically, and can be improved by introducing an adaptive step size. The purpose is to achieve faster convergence speed; by adjusting the step size according to the distance between the current node and the target node and the number of iterations, the algorithm can have a larger step size when far away from the target area and quickly explore the search space; and a smaller step size when close to the target area, so as to perform a refined search and accelerate the convergence speed of the trajectory planning.

[0087] The adaptive step size depends on the number of iterations and the distance between the current node and the target node.

[0088] Set Node and the target node , hoping to calculate the distance between them and the number of iterations Calculation step length :

[0089]

[0090] in: For the current node and the target node The Euclidean distance between . To adjust the parameters, used to adjust the step size. Larger values ​​will increase the step size and improve the search speed; Smaller values ​​result in smaller step sizes and more precise search of the space.

[0091] Based on the above, see Figure 2 , the trajectory planning process of this embodiment is as follows: model the obstacles, use a three-dimensional grid map to model the mountainous environment of the distribution network, build a map model, and set the starting and ending positions of the trajectory planning. Randomly generate a track point in the trajectory planning space; find the nearest random track point in the set of known track points; intercept a new track point from the nearest track point with a step length of t in the straight line direction between the two points. Determine whether there are any obstacles between the nearest track point and the new track point. If so, discard them; otherwise, add the new track point to the set of UAV tracks. Repeat the above process until the target point is reached, and optimize the generated UAV track.

[0092] Specifically, the execution process is as follows:

[0093] 1. Initialization:

[0094] Initialize the search tree , containing only the initial node .

[0095] 2. Loop iteration:

[0096] To the maximum number of iterations , iterate from step 3 to step 9.

[0097] 3. Sampling: In the configuration space In the example, we use the target selection strategy to sample a point .

[0098] The sampling point selection strategy is as follows:

[0099] Calculate the weight of each sampling point :

[0100]

[0101] in is the node in the tree closest to the final target point, represents a sampling point in the sampling point set, is the distance from the nearest node to the sampling point The track cost.

[0102] Then, the probability of selecting each sampling point is calculated according to the weight of the sampling point :

[0103]

[0104] Random Generation To perform roulette selection, introduce random probability It can make the algorithm explore the space more flexibly and comprehensively, and improve the quality of trajectory planning.

[0105] Finally, according to random probability To select the final sampling point :

[0106]

[0107] 4. Nearest neighbor query: in the tree Find the The nearest node .

[0108] 5. Expand and generate new nodes: Generate new nodes with adaptive step size and target bias strategy .

[0109] The specific measures of adaptive step size are as follows:

[0110] because The computational complexity of the algorithm grows geometrically in complex environments, so an adaptive step size is introduced to improve it. The purpose is to achieve faster convergence speed; by adjusting the step size according to the distance between the current node and the target node and the number of iterations, the algorithm can have a larger step size when far away from the target area and quickly explore the search space; when close to the target area, the step size is smaller, so as to perform a refined search and accelerate the convergence speed of trajectory planning.

[0111] The adaptive step size depends on the number of iterations and the distance between the current node and the target node.

[0112] Set Node and the target node , hoping to calculate the distance between them and the number of iterations Calculation step length :

[0113]

[0114] in: For the current node and the target node The Euclidean distance between . To adjust the parameters, used to adjust the step size. Larger values ​​will increase the step size and improve the search speed; Smaller values ​​result in smaller step sizes and more precise search of the space.

[0115] The target bias strategy is as follows:

[0116] This method can improve the sampling The convergence speed of the algorithm to the target area reduces the complexity of the algorithm and improves the track quality.

[0117] for The algorithm introduces a target bias parameter , for the new node The generation direction has a certain bias.

[0118] First, compute the nodes To the next node The direction vector and node To the end The angle between the direction vectors of , then the direction vector from the node to the next node for:

[0119]

[0120] node To the end Direction vector for:

[0121]

[0122] The angle between two vectors for:

[0123]

[0124] In this way, the probability of the next node is adjusted according to the size of the angle, so that it grows towards the target node with a greater probability at a smaller angle. is to select the next node The probability of Is an existing node To the next destination node and to the end The angle between the direction vectors. Then the adjusted probability for:

[0125]

[0126] According to the size of the angle, the new node can be adjusted by the formula The probability of . This affects the growth direction of the tree. By adjusting the target bias parameter The size of the control target bias is achieved, so that the algorithm can reach the end point faster. .

[0127] The introduction of target bias parameters improves The convergence speed of the algorithm is improved and the complexity of the algorithm is reduced.

[0128] 6. Collision detection: Check from arrive Is the trajectory collision-free?

[0129] If the track is collision free, do the following:

[0130] 7. Cost minimization: Search all nodes in the neighborhood of and select the node with the lowest cost As The parent node of .

[0131] 8. Add a new node: Add to Tree and connect and .

[0132] 9. Reconnect to neighboring nodes: Try to Reconnect other nodes in the neighborhood of to reduce their costs.

[0133] 10. Track extraction:

[0134] If from the initial node To the target node If a track is found, it is extracted and returned.

[0135] 11. Track optimization.

[0136] The track generated by the algorithm is composed of straight lines connecting the start and end points. The track contains sharp corners and undesirable bends, which require track smoothing. Smoothing can reduce sharp corners and sharp bends in the track, making the track smoother and more continuous.

[0137] Use cubic B-spline interpolation to generate a smoother track. Assume that a line consisting of multiple nodes has been determined. The composed track.

[0138] First, calculate the distance between nodes Assumptions yes The Euclidean distance between It can be expressed as the sum of the distances of all previous nodes:

[0139]

[0140] Then calculate the cubic B-spline interpolation coefficients, using the node and distance To calculate the coefficients of the cubic B-spline interpolation.

[0141] Then, based on the interpolation polynomial and parameterized distance, interpolation operations are performed between nodes to generate smooth track points. Representation Node 、 The interpolation polynomial between At each distance The value of is used to obtain smooth points on the track, and a curve in the form of a parametric equation is obtained:

[0142]

[0143] is a parameter, is a cubic B-spline basis function. By changing the curve, the position of any point can be obtained. By connecting the points on the smooth track, a smoother track can be obtained.

[0144] See also Figure 3-Figure 6 , see Figure 3-Figure 6 For the comparison between the improved algorithm and the traditional algorithm. It can be seen that the method of this embodiment can better reduce the computing cost, improve the efficiency and quality of track planning, and is more suitable for the UAV mountain distribution network power inspection task.

[0145] The method and system for hybrid planning of three-dimensional trajectory of unmanned aerial vehicle in distribution network provided by the embodiment of the present invention can improve The algorithm improves the efficiency and feasibility of trajectory planning in a long-distance global environment, avoiding falling into the dead zone of local minimum. The algorithm greatly improves the trajectory planning capability of drones and can better meet the needs of power distribution network inspection in mountainous areas.

[0146] In order to implement the method of an embodiment of the present invention, an embodiment of the present invention also provides a distribution network line UAV three-dimensional trajectory hybrid planning system, including: a processor and a memory for storing a computer program that can be run on the processor; wherein, when the processor is used to run the computer program, the steps of any one of the above methods are executed.

[0147] The above-mentioned system provided in this embodiment belongs to the same concept as the above-mentioned method embodiment. The specific implementation process thereof is detailed in the method embodiment and will not be repeated here.

[0148] In order to implement the method of the embodiment of the present invention, the embodiment of the present invention also provides a computer program product, which includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of the above method.

[0149] Based on the hardware implementation of the above program modules, and in order to implement the method of the embodiment of the present invention, the embodiment of the present invention further provides an electronic device (computer device). Specifically, in one embodiment, the computer device may be a terminal, and its internal structure diagram may be as follows: Figure 7As shown. The computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05 and a memory (not shown in the figure) connected through a system bus. Among them, the processor A01 of the computer device is used to provide computing and control capabilities. The memory of the computer device includes an internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 in the non-volatile storage medium A06. The network interface A02 of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor A01, the method of any one of the above embodiments is implemented. The display screen A04 of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device A05 of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0150] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0151] The device provided by the embodiment of the present invention includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the method of any one of the above embodiments is implemented.

[0152] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0153] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0154] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0155] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0156] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0157] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0158] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0159] It can be understood that the memory of the embodiment of the present invention can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAMbus random access memory (DRRAM, Direct Rambus Random Access Memory).The memories described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memories.

[0160] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0161] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A hybrid planning method for three-dimensional trajectory of unmanned aerial vehicles in distribution network, characterized in that: The method comprises: A three-dimensional grid map is used to model the mountainous environment of the distribution network and construct a map model; each grid in the three-dimensional grid map represents a volume element in the mountainous environment of the distribution network; Based on the map model, determine the starting point, final destination point and track planning space of the track planning; According to the target selection strategy, a random track point is determined in the track planning space, and the set track point set is searched to find the track point in the track point set that is closest to the random track point; taking the track point with the closest distance as the starting point, a new track point is determined according to the adaptive step size; judging whether there is an obstacle between the new track point and the track point with the closest distance; if there is no obstacle, the new track point is added to the track point set, and a new UAV track route is generated; if there is an obstacle, a random track point is randomly generated in the track planning space again; repeating the above steps until the generated new UAV track route reaches the final target point; Wherein, determining a random track point in the track planning space according to the target selection strategy includes: Randomly generating a preset number of random track points in the track planning space; Calculate the probability of each of the random track points based on the target selection strategy; Selecting a final random track point from the preset random track points according to the probability of the random track point and the randomly generated random probability; Wherein, selecting a final random track point from the preset random track points according to the probability of the random track point and the randomly generated random probability includes: The final random track point is selected from the preset random track points using the following calculation formula: Formula (3) in, is the final random track point selected, is the probability of a random track point, is a random probability.

2. The method for hybrid planning of three-dimensional trajectory of unmanned aerial vehicle in distribution network according to claim 1 is characterized in that: Calculating the probability of each of the random track points based on the target selection strategy includes: The weight of each random track point is calculated using the following formula; Formula (1) in, is the weight of each random track point, is the node in the search tree closest to the final target point, represents a sampling point in the sampling point set, is the distance from the nearest node to the sampling point The track cost; Calculate the probability of each random track point according to the weight of each random track point; Formula (2) in, is the probability of a random track point, , is a sampling point in the sampling point set, is the weight of the sampling point.

3. The hybrid planning method for three-dimensional trajectory of unmanned aerial vehicle in distribution network according to claim 1 is characterized in that: The calculation method of the adaptive step size includes: The adaptive step size is calculated using the following formula: Formula (4) in, is the adaptive step size, To adjust the parameters, used to adjust the step size, is the number of iterations, For the current node and the target node The Euclidean distance between .

4. The method for hybrid planning of three-dimensional trajectory of unmanned aerial vehicle in distribution network according to claim 1 is characterized in that: Determine new track points, including: Randomly generate a preset number of new track points near the track point closest to the distance; Calculate the direction vector between the closest track point and the new track point, the direction vector between the closest track point and the final target point, and the angle between the two direction vectors; Adjust the probability of the new track point according to the direction vector between the track point closest to the final target point and the angle between the two direction vectors; A new track point is selected from the preset new track points according to the adjusted probability of the new track point.

5. The method for hybrid planning of three-dimensional trajectory of unmanned aerial vehicle in distribution network according to claim 4 is characterized in that: Calculating the direction vector between the closest track point and the new track point, the direction vector between the closest track point and the final target point, and the angle between the two direction vectors, including: The direction vector between the closest track point and the new track point, the direction vector between the closest track point and the final target point, and the angle between the two direction vectors are calculated using the following formula: Formula (5) Formula (6) Formula (7) in, is the direction vector between the nearest track point and the new track point, is the coordinate of the nearest track point, is the coordinates of the track point; is the direction vector between the nearest track point and the final target point, is the coordinate of the final target point, is the angle between the two direction vectors.

6. The method for hybrid planning of three-dimensional trajectory of unmanned aerial vehicle in distribution network according to claim 4 is characterized in that: According to the direction vector between the closest track point and the final target point, and the angle between the two direction vectors, the probability of the new track point is adjusted, including: The probability of the new track point is adjusted using the following calculation: Formula (8) in, is the adjusted probability, is the probability before adjustment, is the target bias parameter.

7. A three-dimensional trajectory hybrid planning system for UAVs in distribution network lines, characterized in that: include: A processor and a memory for storing a computer program that can be run on the processor; wherein, when the processor is used to run the computer program, the steps of the method according to any one of claims 1 to 6 are executed.

8. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images