Multi-uav task path planning method and device, equipment and medium

Abstract

The present application relates to the technical field of unmanned aerial vehicle path planning, and discloses a multi-unmanned aerial vehicle task path planning method, device, equipment and medium, the method comprising: establishing a space coordinate system in a target space, and determining a task point space coordinate of each task point; dividing the target space into multiple fan-shaped space regions, selecting any fan-shaped space region as a target region, and constructing a task point feature representation of the task point and a region feature representation of the target region; dividing an unmanned aerial vehicle sub-cluster of the target region and selecting a target unmanned aerial vehicle; when the target unmanned aerial vehicle accesses any task point in the target region, calculating a next task point of the target unmanned aerial vehicle according to a machine space coordinate of the target unmanned aerial vehicle, the task point feature representation of the currently accessed task point and the region feature representation; and until all the task points in the target region are accessed by the unmanned aerial vehicle. The present application can improve the efficiency of path planning of the multi-unmanned aerial vehicle.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) path planning technology, and in particular to a method, apparatus, device, and medium for multi-UAV mission path planning. Background Technology

[0002] L-MUMPPP (Large-Scale Multi-UAV Mission Path Planning) technology has significant practical implications in numerous application scenarios such as environmental monitoring, disaster assessment, and intelligence reconnaissance, and has received widespread attention in recent years. This technology is typically used to enable multiple UAVs to access high-value mission points distributed throughout space, maximizing overall system benefits, within limited range and energy constraints, starting from a shared hub. However, with the continuous expansion of mission scale and the potential for high heterogeneity in the spatial distribution and value structure of mission points, this technology exhibits strong coupling characteristics between mission allocation, path planning, and resource constraints, leading to a significant increase in solution difficulty.

[0003] Existing solutions employ two approaches. First, evolutionary algorithms and their variants are used to directly solve the global path planning problem in space. However, evolutionary algorithms have slow convergence speeds, especially when dealing with high-dimensional problems, potentially requiring significant computational resources and easily getting trapped in local optima, leading to inefficiency in path planning. Second, deep reinforcement learning is used from an end-to-end perspective to directly solve the global path planning problem in space. Through the interaction between the UAV and its environment, optimal strategies can be automatically derived without a definite model. However, due to the presence of multiple tasks, the UAV itself, and its related attributes, a high-dimensional state and action space is typically encountered, resulting in extremely high computational resource demands and further inefficiency in path planning.

[0004] Therefore, the low efficiency of path planning for multiple drones in large-scale scenarios has become an urgent problem to be solved. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and medium for multi-UAV mission path planning, the main purpose of which is to solve the problem of low efficiency when multiple UAVs perform path planning.

[0006] Firstly, to achieve the above objectives, the present invention provides a multi-UAV mission path planning method, comprising: S1. Determine the spatial coordinates of each preset task point within the preset target space; S2. Obtain the preset task feature value for each task point, and calculate the task point weight for each task point based on the preset weight parameter, the spatial coordinates of the task point, and the task feature value. S3. Divide the target space into a preset number of sector-shaped spatial regions according to the weight of the task points; S4. Select any sector-shaped spatial region as the target region, and construct the task point feature representation of each task point based on the task point spatial coordinates and task feature values ​​of each task point in the target region. S5. Perform feature fusion on the task point feature representations of all task points within the target area to obtain the regional feature representation of the target area; S6. Based on the sum of the task point weights of all task points within the target area, select a drone sub-cluster from the drone cluster within the target area, and select any drone within the drone sub-cluster as the target drone. S7. Analyze and determine the next task point corresponding to the current access task point of the target drone according to the preset algorithm, determine whether the next task point can be accessed effectively, and if it can be accessed effectively, control the target drone to access the next task point after the current access task point is completed. S8. When the target UAV visits the next task point, the next task point is taken as the new current task point and the above step S7 is repeated until there is no corresponding next task point that can be effectively visited, or until all task points in the target area have been visited.

[0007] Secondly, the present invention also provides a multi-UAV mission path planning device, comprising: The coordinate construction module is used to determine the spatial coordinates of each preset task point within a preset target space. The spatial partitioning module is used to obtain the preset task feature value of each task point, calculate the task point weight of each task point according to the preset weight parameters, the spatial coordinates of the task point and the task feature value, and divide the target space into a preset number of fan-shaped spatial regions according to the task point weight. The difference between the sum of the task point weights of all task points in each fan-shaped spatial region and the sum of the task point weights of all task points in any other fan-shaped spatial region is less than a preset first threshold. The data processing module is used to select any sector-shaped spatial region as the target region, construct the task point feature representation of each task point based on the task point spatial coordinates and task feature values ​​of each task point in the target region, perform feature fusion on the task point feature representations of all task points in the target region to obtain the regional feature representation of the target region, select a drone sub-cluster from the drone cluster of the target region based on the sum of the task point weights of all task points in the target region, and select any drone in the drone sub-cluster as the target drone. The path planning module is used to analyze and determine the next task point corresponding to the current access task point of the target drone according to a preset algorithm, determine whether the next task point can be accessed effectively, and if it can be accessed effectively, then after the current access task point is accessed, control the target drone to access the next task point. When the target drone accesses the next task point, the next task point is used as the new current access task point, until there is no corresponding next task point that can be accessed effectively, or until all task points in the target area have been accessed.

[0008] Thirdly, the present invention also provides an electronic device, the electronic device comprising: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the multi-UAV mission path planning method described above.

[0009] Fourthly, the present invention also provides a computer-readable storage medium storing at least one computer program, which is executed by a processor in an electronic device to implement the multi-UAV mission path planning method described above.

[0010] In this embodiment of the invention, on the one hand, by integrating information such as task point weights and spatial coordinates to adaptively generate sector-shaped task partitions that satisfy UAV range constraints, the UAV path planning problem for large spaces is decomposed into path planning problems within several controllable-scale sector-shaped task partitions. This allows for fragmented decomposition of the overall planning of large-scale UAV swarms, reducing the amount of data processing and improving the efficiency of path planning. On the other hand, weight-based UAV allocation and path analysis using the regional feature representation of sector-shaped task partitions ensure that path planning within each sector-shaped task partition is based on the characteristics of the task points within that sector, reducing human intervention and route redundancy or decision confusion, thereby further improving the efficiency of path planning. Therefore, this solution can improve the efficiency of path planning for multiple UAVs. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of an application environment for a multi-UAV mission path planning method according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating a multi-UAV mission path planning method according to an embodiment of the present invention. Figure 3 This is a functional block diagram of a multi-UAV mission path planning device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device for implementing a multi-UAV mission path planning method according to an embodiment of the present invention; Figure 5 This is another schematic diagram of an electronic device for implementing a multi-UAV mission path planning method according to an embodiment of the present invention.

[0013] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0014] It should be noted that in the technical solutions disclosed in this invention, the acquisition of user information (personal image data (e.g., facial videos or pictures, facial feature videos or pictures, etc.) and personal privacy information (e.g., name, ID number, occupation, address, etc.)) is all completed with the user's knowledge and consent, and the acquisition of the relevant user information is legal and compliant.

[0015] To enable those skilled in the art to better understand the technical solutions of this disclosure, and to fully understand and implement the process of how this disclosure applies technical means to solve technical problems and achieve corresponding technical effects, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, not all embodiments. The embodiments of this disclosure and the various features within them can be combined with each other without conflict, and the resulting technical solutions are all within the protection scope of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort should fall within the protection scope of this disclosure.

[0016] It should be noted that the terms "first," "second," etc., used in this disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0017] This application provides a multi-UAV mission path planning method. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the device provided in this application: a server, a terminal, etc. In other words, the multi-UAV mission path planning method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0018] The multi-UAV mission path planning method of this invention can be applied to, for example... Figure 1 In this application environment, the client can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. The invention will now be described in detail through specific embodiments.

[0019] Reference Figure 2 The diagram shown is a flowchart illustrating a multi-UAV mission path planning method according to an embodiment of the present invention. In this embodiment, the multi-UAV mission path planning method includes: S1. Determine the spatial coordinates of each preset task point within the preset target space.

[0020] In this embodiment of the invention, the target space can be any space where UAV flight missions can be performed. Typically, in order to facilitate the division of the target space, the center point of the space is often used as the take-off point of the UAV swarm.

[0021] In practical applications of this invention, edge points, corner points, etc., within the target space can also be determined as the takeoff points of the drone swarm, and can be freely set according to the actual situation.

[0022] In this embodiment of the invention, in order to achieve accurate positioning of numerous character nodes within the target space, a spatial coordinate system can be established with the take-off point as the origin, and then the spatial coordinates within the target space can be determined based on the established coordinate system.

[0023] In detail, when any point in the target space is selected as the takeoff point of the UAV swarm, the takeoff point is taken as the origin, the horizontal direction in the target space is taken as the X-axis, the vertical direction in the target space is taken as the Z-axis, and the direction perpendicular to the plane containing the X-axis and Z-axis is taken as the Y-axis to construct a spatial coordinate system, thereby determining the spatial coordinates of any human point in the target space, wherein the task point is the spatial point that the UAV needs to explore by flying.

[0024] For example, the takeoff point of a drone swarm can be marked as For any task point within the target space The coordinates can be represented as The set of coordinates of all task points within the target space can be represented as Where N is a natural number representing the number of task points in the target space.

[0025] S2. Obtain the preset task feature value for each task point, and calculate the task point weight for each task point based on the preset weight parameter, the spatial coordinates of the task point, and the task feature value.

[0026] In this embodiment of the invention, the task feature value is a pre-set constant parameter used to represent the task benefit, task cost, or task priority of each task point in the target space.

[0027] In detail, to improve the efficiency and accuracy of subsequent path planning for each drone in the drone swarm, the task point weight of each task point can be analyzed by combining preset weight parameters, spatial coordinates of task points, and task feature values.

[0028] In this embodiment of the invention, calculating the task point weight of each task point based on preset weight parameters, the spatial coordinates of the task point, and the task feature value includes: Select any task point within the target space as a target task point, and calculate the radial distance between the target task point and the origin based on the task point spatial coordinates of the target task point; Select task points within a preset neighborhood of the target task point, and calculate the sum of radial distances between the target task point and each task point within the preset neighborhood. The radial distance between the target task point and the origin, the sum of the radial distances, and the spatial coordinates of the target task point are normalized to obtain multiple normalized parameters. The task point weight of the target task point is obtained by weighting and summing multiple normalized parameters using preset weight parameters.

[0029] In detail, the radial distance between different points in the target space can be calculated using a preset distance algorithm (such as Euclidean distance algorithm, cosine distance algorithm, etc.).

[0030] For example, for the target task point within the target space The radial distance between it and the origin can be calculated using the following formula. : in, The radial distance between the target task point and the origin. Let these be the coordinate positions of the target task point within the spatial coordinate system. The coordinates are the origin.

[0031] In practical applications of this invention, the distribution density of task points within a local space significantly affects the feasibility and efficiency of the UAV's flight path. Therefore, this solution selects task points within a preset neighborhood of the target task point, such as using the radius of the target task point... The target task point is selected as its neighborhood, and all task points within it are selected. The sum of the radial distances between the target task point and each task point within the preset neighborhood range is calculated. The sum of the radial distances can be used to represent the distribution density of all task points in the local space of the neighborhood of the target task point.

[0032] For example, for the target task point Set neighborhood radius Therefore, the sum of radial distances between the target task point and every task point within the preset neighborhood range can be calculated using the following formula: in, The neighborhood radius of the target task point Distribution density of all task points within, The target task point's spatial coordinates are given. For the first in the said field The spatial coordinates of each task point.

[0033] In detail, in order to increase the neighborhood radius The reasonableness of the numerical value can be assessed in different practical scenarios, adjusting the neighborhood radius accordingly. Defined as: in, This represents the average adjacent distance between all task nodes. This is a preset constant, usually 1.5.

[0034] Furthermore, a preset normalization function can be used to normalize the radial distance between the target task point and the origin, the sum of the radial distances, and the spatial coordinates of the target task point, and then perform a weighted summation to obtain the task point weight of the target task point.

[0035] For example, the task point weight of the target task point is calculated using the following formula: in, The task point weights for the target task points. This represents the normalization operator. The target task point's spatial coordinates are given. The radial distance between the target task point and the origin. The neighborhood radius of the target task point Distribution density of all task points within, , , These are preset weight parameters.

[0036] S3. Divide the target space into a preset number of sector-shaped spatial regions according to the weight of the task points.

[0037] In the actual application scenario of this invention, the difference between the sum of the task point weights of all task points in each sector spatial region and the sum of the task point weights of all task points in any other sector spatial region is less than a preset first threshold.

[0038] If joint path planning is performed directly on all task points within the target space, the dimensions of data analysis and decision-making will increase linearly with the task size, resulting in an exponential expansion of the search space during path planning. In this case, traditional single-scale or fixed-rule task allocation mechanisms often struggle to simultaneously balance efficiency and feasibility.

[0039] Therefore, this solution divides the target space into a preset number of sector-shaped spatial regions based on the task point weights, thereby reducing the spatial scale of a single region and greatly reducing the amount of data required for subsequent path planning, data analysis, and decision-making, thus improving analysis efficiency.

[0040] In this embodiment of the invention, dividing the target space into a preset number of sector-shaped spatial regions according to the task point weights includes: Establish a polar coordinate system with the takeoff point as the origin; Based on the polar coordinate system, convert the spatial coordinates of each task point in the target space into polar coordinates. The initial sector angle of each sector space region is determined according to the preset quantity; Based on the initial sector angle, the polar coordinates of all task points within each sector spatial region are sorted in ascending order to obtain a one-dimensional ordered sequence corresponding to each sector spatial region. The task point weights of each task point in the polar coordinates of each task point within each one-dimensional ordered sequence are accumulated to obtain the sum of the task point weights of each one-dimensional ordered sequence; If the difference between the sum of the task point weights of all task points in a sector spatial region and any other sector spatial region is greater than or equal to a preset first threshold, then the initial sector angle of the sector spatial region with the smallest sum of task point weights is increased by a preset value, and the step of accumulating the task point weights of the polar coordinates of each task point in each one-dimensional ordered sequence is returned. If the difference between the sum of the task point weights of all task points in any sector space region and any other sector space region is less than a preset first threshold, then the sector space region division of the target space is completed.

[0041] Specifically, the initial sector angle of each sector space region within the target space can be determined according to a preset number. For example, when the preset number is 4, the initial sector angle of each sector space region is determined to be 90 degrees, and when the preset number is 8, the initial sector angle of each sector space region is determined to be 45 degrees.

[0042] Specifically, this scheme sorts the polar coordinates of all task points within each sector spatial region in ascending order based on the initial sector angle, resulting in a one-dimensional ordered sequence corresponding to each sector spatial region: ,in, The sector angle of the sector-shaped spatial region. For the first sector within this sector-shaped spatial region Polar coordinates of each task point.

[0043] Furthermore, if the difference between the sum of the task point weights of all task points in a sector-shaped spatial region and any other sector-shaped spatial region is greater than or equal to a preset first threshold, it indicates that the sum of the task point weights in a sector-shaped spatial region is too small. Therefore, the initial sector angle of the sector-shaped spatial region with the smallest sum of task point weights can be increased by a preset value, and then the step of accumulating the task point weights of the polar coordinates of each task point in each one-dimensional ordered sequence can be returned.

[0044] Ultimately, the difference between the sum of the task weights of all task points within each sector and the sum of the task weights of all task points within any other sector is less than a preset first threshold, thus completing the sector division of the target space. This achieves flexible sector division of the target space, avoiding the potential high heterogeneity in spatial distribution and value structure of task points caused by traditional single-scale division.

[0045] By integrating information such as task point weights, radial distances, and local density that represent task point value, cost, and priority, task partitions that meet range constraints are adaptively generated to improve the efficiency of final path planning.

[0046] S4. Select any sector-shaped spatial region as the target region, and construct the task point feature representation of each task point based on the task point spatial coordinates and task feature values ​​of each task point within the target region.

[0047] In this embodiment of the invention, in order to achieve accurate and efficient path planning for any sector-shaped area, a task point feature representation for each task point can be constructed based on the task point spatial coordinates and task feature values ​​of each task point within the target area.

[0048] For example, target area It contains multiple task points, and the set of task points within this target area can be represented by the following formula: in, For the target area The set of those task points, For the target area Inner One task point, For the target area The number of internal mission points.

[0049] Furthermore, it can be done in the following ways, based on the target area The spatial coordinates of each task point and the task feature values ​​are used to construct the task point feature representation of each task point: in, For the target area Inner Task point feature representation of each task point For the target area Inner The coordinate positions of each task point in the spatial coordinate system. For the target area Inner The spatial coordinates of each task point For the target area The number of internal mission points.

[0050] In this embodiment of the invention, by constructing a task point feature representation for each task point within the target area, accurate positioning and information integration of the task points within the target area can be achieved, making the task point information data have a certain degree of structure and standardization, thereby improving the efficiency of subsequent UAV path planning using task point data.

[0051] S5. Perform feature fusion on the task point feature representations of all task points within the target area to obtain the regional feature representation of the target area.

[0052] In this embodiment of the invention, in order to achieve fast and accurate path planning for UAVs within the target area, it is necessary to construct a regional feature representation that includes the entire target area based on the task point feature representation of each task point within the target area.

[0053] In this embodiment of the invention, the step of fusing the task point feature representations of all task points within the target region to obtain the region feature representation of the target region includes: The task point feature representation of each task point in the target area is nonlinearly mapped using a preset point-level feature mapping function to obtain the mapping feature of each task point. The mapping features of all task points within the target area are aggregated by mean value processing to obtain the aggregated features of all task points within the target area. The aggregated features are numerically compressed using a preset compression mapping function to obtain the regional feature representation of the target region.

[0054] In a practical application scenario of this invention, the following algorithm can be used to implement the above-mentioned feature fusion operation of the task point feature representations of all task points within the target area: in, For target area Regional feature representation, For target area Inner Task point feature representation of each task point For the preset point-level feature mapping function, For the target area The set of those task points, For the target area Inner One task point, For the target area The number of internal mission points This is the preset compression mapping function.

[0055] In this embodiment of the invention, the regional feature representation generated by the above operations contains numerous data features of each task point within the target area, which can be used for subsequent UAV path planning for the target area, thereby improving the efficiency and accuracy of UAV path planning.

[0056] S6. Based on the sum of the task point weights of all task points within the target area, select a drone sub-cluster from the drone cluster within the target area, and select any drone within the drone sub-cluster as the target drone.

[0057] In this embodiment of the invention, since the target area is large and the number of drones in the drone cluster is large, but the number of task points and the rewards of task points are different in different fan-shaped spatial areas divided into the target area, in order to achieve efficient and orderly exploration of different fan-shaped spatial areas, it is necessary to divide the drones into subsets for each fan-shaped spatial area to achieve reasonable scheduling and allocation of drone resources.

[0058] In this embodiment of the invention, selecting a sub-cluster of drones in the target area from the drone cluster based on the sum of the task point weights of all task points within the target area includes: The total weight of the region is obtained by summing the task point weights of all task points within the target area. The sum of the task point weights of all task points within the target space is calculated to obtain the total spatial weights. Calculate the proportion of the sum of regional weights within the sum of spatial weights; According to the stated proportion, a number of drones of the same proportion are randomly selected from the drone cluster to obtain a drone sub-cluster for the target area.

[0059] For example, the sum of the regional weights is The sum of the spatial weights is The total number of drones in the human-machine cluster is The proportion of the total regional weights within the total spatial weights can then be calculated as follows: in, The sum of regional weights represents the proportion of the sum of spatial weights. This represents the sum of regional weights. This is the sum of spatial weights. This represents the total number of drones within the drone swarm. For target area Inner The task point weight of each task point For the target area The number of internal mission points For target space Inner The task point weight of each task point The target space The number of internal mission points.

[0060] In this embodiment of the invention, by taking the above steps, a drone sub-cluster that matches the task point in each sector-shaped spatial area can be selected, thereby reducing resource waste and improving the overall efficiency of drones in completing tasks.

[0061] S7. Analyze and determine the next task point corresponding to the current access task point of the target drone according to the preset algorithm, determine whether the next task point can be accessed effectively, and if it can be accessed effectively, then after the access to the current access task point is completed, control the target drone to access the next task point.

[0062] In this embodiment of the invention, when the target UAV visits any task point within the target area, in order to achieve accurate planning of the subsequent route of the target UAV, it is also necessary to obtain the remaining range of the target UAV so as to combine the range data for refined path planning.

[0063] In this embodiment of the invention, determining the next task point corresponding to the current access task point of the target UAV based on a preset algorithm includes: The spatial coordinates of the task point currently visited by the target drone are used as the machine spatial coordinates of the target drone. The modulated vector is obtained by using a preset modulation vector to perform vector modulation on the feature representation of the task point currently visited by the target UAV. Using a preset Markov decision algorithm, the probability of the target UAV accessing all task points within the target area is calculated based on the machine spatial coordinates, the modulation vector, and the regional feature representation. The task point with the highest access probability within the target area is selected as the next task point for the target UAV.

[0064] In detail, since the task points in the set of task points within the target area do not have a sequential structure when the UAV does not visit them, the task point feature representation of the currently visited task point can be nonlinearly calculated by a preset modulation vector (such as a preset weighted parameter with direction or an iterative gradient) to achieve scaling and translation of the task point feature representation, thereby achieving vector modulation of it.

[0065] Modulated vectors have better information expression capabilities, thereby improving the efficiency and accuracy of subsequent path planning analysis.

[0066] Furthermore, in this embodiment of the invention, MDP (Markov Decision Process) can be used to calculate the access probability of the target UAV to all task points in the target area based on the machine spatial coordinates, the modulation vector, and the regional feature representation. Then, the task point with the highest access probability is selected as the next task point of the target UAV, until all task points in the target area have been accessed by the UAV.

[0067] In this embodiment of the invention, the step of using a preset Markov decision algorithm to calculate the probability of the target UAV accessing all task points within the target area based on the machine spatial coordinates, the modulation vector, and the regional feature representation includes: Select task points within the target area one by one as task points to be analyzed; Calculate the spatial distance between the task point coordinates of the task to be analyzed and the machine space coordinates; Based on the spatial distance, the task point weight of the task point to be analyzed, and the regional feature representation, the Markov decision algorithm is used to perform access revenue analysis on the task point to be analyzed, and the access revenue of the task point to be analyzed is obtained. The access benefits are mapped to a preset numerical range to obtain the access probability of the task point to be analyzed.

[0068] In detail, since the task points in space are unordered and discrete scattered points before the drone visits the task points, and the essence of path planning is the selection of sequential nodes, it is equivalent to a continuous sequential decision problem.

[0069] Therefore, this paper models the path generation process of a single UAV as a discrete-time Markov decision algorithm. For the UAV path planning problem within any sector-shaped spatial region, the path planning process consists of discrete-time steps. The current state of the drone is determined by the spatial distance, the task point weight of the task point to be analyzed, and the regional feature representation. Then, the access benefit analysis of subsequent task points is performed through a preset Markov model to obtain the access probability of subsequent task points.

[0070] S8. When the target UAV visits the next task point, the next task point is taken as the new current task point and the above step S7 is repeated until there is no corresponding next task point that can be effectively visited, or until all task points in the target area have been visited.

[0071] In this embodiment of the invention, on the one hand, by integrating information such as task point weights and spatial coordinates to adaptively generate sector-shaped task partitions that satisfy UAV range constraints, the UAV path planning problem for large spaces is decomposed into path planning problems within several controllable-scale sector-shaped task partitions. This allows for fragmented decomposition of the overall planning of large-scale UAV swarms, reducing the amount of data processing and improving the efficiency of path planning. On the other hand, weight-based UAV allocation and path analysis using the regional feature representation of sector-shaped task partitions ensure that path planning within each sector-shaped task partition is based on the characteristics of the task points within that sector, reducing human intervention and route redundancy or decision confusion, thereby further improving the efficiency of path planning. Therefore, this solution can improve the efficiency of path planning for multiple UAVs.

[0072] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0073] like Figure 3 The diagram shown is a functional block diagram of a multi-UAV mission path planning device provided in an embodiment of the present invention.

[0074] This disclosure provides a multi-UAV mission path planning device, which corresponds one-to-one with the multi-UAV mission path planning method described in the above embodiments. For example... Figure 3 As shown, the multi-UAV mission path planning device 100 can be installed in an electronic device. According to its functions, the multi-UAV mission path planning device 100 includes a coordinate construction module 101, a spatial division module 102, a data processing module 103, and a path planning module 104. Detailed descriptions of each functional module are as follows: The coordinate construction module 101 is used to determine the task point spatial coordinates of each preset task point within a preset target space; The spatial division module 102 is used to obtain the preset task feature value of each task point, calculate the task point weight of each task point according to the preset weight parameter, the spatial coordinates of the task point and the task feature value, and divide the target space into a preset number of fan-shaped spatial regions according to the task point weight. Data processing module 103 is used to select any sector-shaped spatial region as the target region, construct a task point feature representation for each task point based on the task point spatial coordinates and task feature values ​​of each task point in the target region, perform feature fusion on the task point feature representations of all task points in the target region to obtain the regional feature representation of the target region, select a drone sub-cluster from the drone cluster of the target region based on the sum of the task point weights of all task points in the target region, and select any drone in the drone sub-cluster as the target drone. The path planning module 104 is used to analyze and determine the next task point corresponding to the current access task point of the target UAV according to a preset algorithm, determine whether the next task point can be accessed effectively, and if it can be accessed effectively, then after the access to the current access task point is completed, control the target UAV to access the next task point. When the target UAV accesses the next task point, the next task point is used as the new current access task point, until there is no corresponding next task point that can be accessed effectively, or until all task points in the target area have been accessed.

[0075] In this invention, the specific limitations of the multi-UAV mission path planning device can be found in the above-described limitations of the multi-UAV mission path planning method, and will not be repeated here. Each module in the aforementioned multi-UAV mission path planning device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0076] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 4As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external clients via a network connection. When the computer program is executed by the processor, it implements the functions or steps of the multi-UAV mission path planning method server-side.

[0077] In one embodiment, a computer device is provided, which may be a client, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with an external server via a network connection. When executed by the processor, the computer program implements the functions or steps of the multi-UAV mission path planning method on the client side.

[0078] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: S1. Determine the spatial coordinates of each preset task point within the preset target space; S2. Obtain the preset task feature value for each task point, and calculate the task point weight for each task point based on the preset weight parameter, the spatial coordinates of the task point, and the task feature value. S3. Divide the target space into a preset number of sector-shaped spatial regions according to the weight of the task points; S4. Select any sector-shaped spatial region as the target region, and construct the task point feature representation of each task point based on the task point spatial coordinates and task feature values ​​of each task point in the target region. S5. Perform feature fusion on the task point feature representations of all task points within the target area to obtain the regional feature representation of the target area; S6. Based on the sum of the task point weights of all task points within the target area, select a drone sub-cluster from the drone cluster within the target area, and select any drone within the drone sub-cluster as the target drone. S7. Analyze and determine the next task point corresponding to the current access task point of the target drone according to the preset algorithm, determine whether the next task point can be accessed effectively, and if it can be accessed effectively, control the target drone to access the next task point after the current access task point is completed. S8. When the target UAV visits the next task point, the next task point is taken as the new current task point and the above step S7 is repeated until there is no corresponding next task point that can be effectively visited, or until all task points in the target area have been visited.

[0079] In the several embodiments provided by this invention, it should be understood that the disclosed devices and apparatuses can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0080] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0081] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0082] In some embodiments of this example, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

[0083] The readable storage medium of the present invention stores a computer program, which, when executed by a processor of an electronic device, can perform the following: S1. Determine the spatial coordinates of each preset task point within the preset target space; S2. Obtain the preset task feature value for each task point, and calculate the task point weight for each task point based on the preset weight parameter, the spatial coordinates of the task point, and the task feature value. S3. Divide the target space into a preset number of sector-shaped spatial regions according to the weight of the task points; S4. Select any sector-shaped spatial region as the target region, and construct the task point feature representation of each task point based on the task point spatial coordinates and task feature values ​​of each task point in the target region. S5. Perform feature fusion on the task point feature representations of all task points within the target area to obtain the regional feature representation of the target area; S6. Based on the sum of the task point weights of all task points within the target area, select a drone sub-cluster from the drone cluster within the target area, and select any drone within the drone sub-cluster as the target drone. S7. Analyze and determine the next task point corresponding to the current access task point of the target drone according to the preset algorithm, determine whether the next task point can be accessed effectively, and if it can be accessed effectively, control the target drone to access the next task point after the current access task point is completed. S8. When the target UAV visits the next task point, the next task point is taken as the new current task point and the above step S7 is repeated until there is no corresponding next task point that can be effectively visited, or until all task points in the target area have been visited.

[0084] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server side and client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0085] Computer-readable storage media may also store at least one computer-executable program / instruction, such as computer-readable instructions. Computer-readable storage media include, but are not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Computer-readable storage media may include, for example, read-only memory (ROM), hard disk, flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device executes the computer-readable instructions stored on the computer-readable storage medium, the various methods described above can be performed.

[0086] In addition, the computer device may include (but is not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (e.g., keyboard, mouse, speakers, etc.).

[0087] In one embodiment, the at least one computer-executable instruction may also be compiled into or comprise a software product / computer program product, wherein one or more computer-executable instructions are executed by a processor to perform the steps of the various functions and / or methods in the embodiments described herein.

[0088] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Furthermore, any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory.

[0089] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0090] In the embodiments provided in this disclosure, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0091] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

[0092] It should be noted that any AI models, software tools, or components not belonging to this company appearing in the embodiments of this application are merely illustrative examples and do not represent actual use. All user personal information involved in the embodiments of this application has been authorized (with the knowledge and consent) by the relevant parties or has been fully authorized by all parties, and the executing entity may obtain it through various legal and compliant means. The collection, storage, use, processing, transmission, provision, and disclosure of the information, data, and signals involved all comply with relevant laws and regulations and do not violate public order and good morals.

Claims

1. A multi-UAV mission path planning method, characterized in that, The method includes: S1. Determine the spatial coordinates of each preset task point within the preset target space; S2. Obtain the preset task feature value for each task point, and calculate the task point weight for each task point based on the preset weight parameter, the spatial coordinates of the task point, and the task feature value. S3. Divide the target space into a preset number of sector-shaped spatial regions according to the weight of the task points; S4. Select any sector-shaped spatial region as the target region, and construct the task point feature representation of each task point based on the task point spatial coordinates and task feature values ​​of each task point in the target region. S5. Perform feature fusion on the task point feature representations of all task points within the target area to obtain the regional feature representation of the target area; S6. Based on the sum of the task point weights of all task points in the target area, select a drone sub-cluster from the drone cluster in the target area, and select any drone in the drone sub-cluster as the target drone. S7. Analyze and determine the next task point corresponding to the current access task point of the target drone according to the preset algorithm, determine whether the next task point can be accessed effectively, and if it can be accessed effectively, control the target drone to access the next task point after the current access task point is accessed. S8. When the target UAV visits the next task point, the next task point is taken as the new current task point and the above step S7 is repeated until there is no corresponding next task point that can be effectively visited, or until all task points in the target area have been visited.

2. The multi-UAV mission path planning method as described in claim 1, characterized in that, The step of calculating the task point weight of each task point based on preset weight parameters, the spatial coordinates of the task points, and the task feature values ​​includes: Select any task point within the target space as a target task point, and calculate the radial distance between the target task point and the origin based on the task point spatial coordinates of the target task point. Select task points within a preset neighborhood of the target task point, and calculate the sum of radial distances between the target task point and each task point within the preset neighborhood. The radial distance between the target task point and the origin, the sum of the radial distances, and the spatial coordinates of the target task point are normalized to obtain multiple normalized parameters. The task point weight of the target task point is obtained by weighting and summing multiple normalized parameters using preset weight parameters.

3. The multi-UAV mission path planning method as described in claim 1, characterized in that, The step of dividing the target space into a preset number of sector-shaped spatial regions according to the task point weights includes: Establish a polar coordinate system with the takeoff point as the origin; Based on the polar coordinate system, convert the spatial coordinates of each task point in the target space into polar coordinates. The initial sector angle of each sector space region is determined according to the preset quantity; Based on the initial sector angle, the polar coordinates of all task points within each sector spatial region are sorted in ascending order to obtain a one-dimensional ordered sequence corresponding to each sector spatial region. The task point weights of each task point in the polar coordinates of each task point within each one-dimensional ordered sequence are accumulated to obtain the sum of the task point weights of each one-dimensional ordered sequence; If the difference between the sum of the task point weights of all task points in a sector spatial region and any other sector spatial region is greater than or equal to a preset first threshold, then the initial sector angle of the sector spatial region with the smallest sum of task point weights is increased by a preset value, and the step of accumulating the task point weights of the polar coordinates of each task point in each one-dimensional ordered sequence is returned. If the difference between the sum of the task point weights of all task points in any sector space region and any other sector space region is less than a preset first threshold, then the sector space region division of the target space is completed.

4. The multi-UAV mission path planning method as described in claim 1, characterized in that, The step of fusing the feature representations of all task points within the target area to obtain the region feature representation of the target area includes: The task point feature representation of each task point in the target area is nonlinearly mapped using a preset point-level feature mapping function to obtain the mapping feature of each task point. The mapping features of all task points within the target area are aggregated by mean value processing to obtain the aggregated features of all task points within the target area. The aggregated features are numerically compressed using a preset compression mapping function to obtain the regional feature representation of the target region.

5. The multi-UAV mission path planning method as described in claim 1, characterized in that, The step of selecting a drone sub-cluster from the drone cluster within the target area based on the sum of the task point weights of all task points within the target area includes: The total weight of the region is obtained by summing the task point weights of all task points within the target area. The sum of the task point weights of all task points within the target space is calculated to obtain the total spatial weights. Calculate the proportion of the sum of regional weights within the sum of spatial weights; According to the stated proportion, a number of drones of the same proportion are randomly selected from the drone cluster to obtain a drone sub-cluster for the target area.

6. The multi-UAV mission path planning method as described in claim 1, characterized in that, The step of determining the next task point corresponding to the current access task point of the target drone based on a preset algorithm includes: The spatial coordinates of the task point currently visited by the target drone are used as the machine spatial coordinates of the target drone. The modulated vector is obtained by using a preset modulation vector to perform vector modulation on the feature representation of the task point currently visited by the target UAV. Using a preset Markov decision algorithm, the probability of the target UAV accessing all task points within the target area is calculated based on the machine spatial coordinates, the modulation vector, and the regional feature representation. The task point with the highest access probability within the target area is selected as the next task point for the target UAV.

7. The multi-UAV mission path planning method as described in claim 6, characterized in that, The step of using a preset Markov decision algorithm to calculate the probability of the target UAV accessing all task points within the target area based on the machine's spatial coordinates, the modulation vector, and the regional feature representation includes: Select task points within the target area one by one as task points to be analyzed; Calculate the spatial distance between the task point coordinates of the task to be analyzed and the machine space coordinates; Based on the spatial distance, the task point weight of the task point to be analyzed, and the regional feature representation, the Markov decision algorithm is used to perform access revenue analysis on the task point to be analyzed, and the access revenue of the task point to be analyzed is obtained. The access benefits are mapped to a preset numerical range to obtain the access probability of the task point to be analyzed.

8. A multi-UAV mission path planning device, characterized in that, The device includes: The coordinate construction module is used to determine the spatial coordinates of each preset task point within a preset target space. The spatial partitioning module is used to obtain the preset task feature value of each task point, calculate the task point weight of each task point according to the preset weight parameter, the spatial coordinates of the task point and the task feature value, and divide the target space into a preset number of sector spatial regions according to the task point weight. The data processing module is used to select any sector-shaped spatial region as the target region, construct the task point feature representation of each task point based on the task point spatial coordinates and task feature values ​​of each task point in the target region, perform feature fusion on the task point feature representations of all task points in the target region to obtain the regional feature representation of the target region, select a drone sub-cluster from the drone cluster of the target region based on the sum of the task point weights of all task points in the target region, and select any drone in the drone sub-cluster as the target drone. The path planning module is used to analyze and determine the next task point corresponding to the current access task point of the target drone according to a preset algorithm, determine whether the next task point can be accessed effectively, and if it can be accessed effectively, then after the current access task point is accessed, control the target drone to access the next task point. When the target drone accesses the next task point, the next task point is used as the new current access task point, until there is no corresponding next task point that can be accessed effectively, or until all task points in the target area have been accessed.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the multi-UAV mission path planning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the multi-UAV mission path planning method as described in any one of claims 1 to 7.

Images