UAV path method, device, UAV and readable storage medium
A UAV path planning model is constructed by using ant colony algorithm and online decision function, which solves the problem of online decision-making of benefits and costs in UAV search planning and achieves efficient path optimization.
Patent Information
- Application Number
- CN202310267909.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-03-20
- Publication Date
- 2025-09-26
- Estimated Expiration
- 2043-03-20
AI Technical Summary
Existing UAV search planning methods are unable to obtain search benefits and costs online, resulting in inefficient decision-making.
The ant colony algorithm is used to solve the traveling salesman problem, and a path planning function model is constructed, including path cost, search cost and benefit function models. The optimal solution is obtained through the online decision function to determine the optimal search path of the drone.
It achieves efficient online decision-making on the optimal search path for drones under the constraints of limited energy costs, improving search efficiency and benefits.
Smart Images
Figure CN116499462B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a method and device for unmanned aerial vehicle (UAV) path planning, an UAV, and a readable storage medium. Background Art
[0002] Unmanned aerial vehicles (UAVs), also known as drones, have been widely used in fields such as information search, environmental monitoring, border reconnaissance, and material transportation due to their advantages in maneuverability, portability, and flexibility. A UAV operation typically involves two phases: path selection and search. These two aspects are the technical foundation of UAV applications and directly impact the efficiency of the operation.
[0003] UAV search planning is an important way to collect information, and since UAV applications are superior to traditional manned aircraft methods, it is a problem that has attracted much attention. In recent years, research has been devoted to solving the joint optimization challenge of path selection and search process in UAV search planning problems. Traditional methods follow an offline two-stage paradigm, making all decisions before taking action in the order of considering path selection and search time allocation. In the first stage, only greedy search, ant colony algorithm (ACO) and other methods are used to solve the path problem separately, without considering information such as search benefits and costs of each search point. The second stage focuses on the allocation of time or cost, considering the difficulty and potential value of searching a specific point. To solve this problem, a series of methods have been proposed, including a fast scoring method based on information benefit significance
[13] , a heuristic method based on partially observable decentralized Markov process, and a time allocation based on Newton's method. These methods all have a premise, that is, before making a decision, the prior parameters of search benefits and costs need to be known.
[0004] In the actual operation of drone search, the search benefits and costs are usually obtained online rather than before the action. Therefore, a technical solution is urgently needed to achieve online acquisition of path planning information such as benefits and costs during drone search planning. Summary of the Invention
[0005] In view of this, the embodiments of the present disclosure provide a drone path planning method, device, drone, and readable storage medium, which at least partially solve the problems existing in the prior art.
[0006] In a first aspect, an embodiment of the present disclosure provides a method for planning a path for a drone, the method comprising:
[0007] Determine the starting and stopping points and all flight locations of the search area;
[0008] Constructing a path planning function model for the UAV, wherein the path planning function model includes a path cost function model for the UAV to return to the start-stop base point after passing through all flight locations, a search cost function model at all target locations, and a search benefit function model, wherein the target location is any location among the flight locations;
[0009] Under the constraint of limited energy cost, obtaining the optimal solution of the path planning function model;
[0010] An optimal search path for the UAV in the search area is determined according to the optimal solution of the path planning function model.
[0011] According to a specific implementation of the embodiment of the present disclosure, the step of constructing a path planning function model for the drone includes:
[0012] The path planning problem of the UAV returning to the starting and stopping base point after passing through all flight locations is mapped to the traveling salesman problem of integer linear programming;
[0013] Constructing a traveling salesman cost function model according to predetermined path cost parameters as a path planning function model for the UAV;
[0014] The step of obtaining the optimal solution of the path planning function model under the constraint of limited energy cost includes:
[0015] The traveling salesman cost function model is solved using an ant colony algorithm to obtain an optimal route for determining an optimal search path.
[0016] According to a specific implementation of the embodiment of the present disclosure, the step of constructing a path planning function model for the drone includes:
[0017] Obtaining online the search cost and search benefit of the UAV at each flight location;
[0018] According to the search cost of each flight location, an online decision function is used to construct a search cost function model corresponding to all flight locations;
[0019] According to the search cost of each flight location, an online decision function is used to construct a search benefit function model corresponding to all flight locations.
[0020] According to a specific implementation of the embodiment of the present disclosure, the step of obtaining the optimal solution of the path planning function model under the constraint of limited energy cost includes:
[0021] obtaining a limited energy cost of the UAV;
[0022] determining an available search cost based on the limited energy cost and the path cost corresponding to the optimal route;
[0023] Under the constraint of the available search cost, the optimal solution of the search cost function model and the search benefit function model is solved, and the flight location corresponding to the optimal solution is determined as the target location.
[0024] According to a specific implementation of the embodiment of the present disclosure, the step of determining the optimal search path of the UAV in the search area according to the optimal solution of the path planning function model includes:
[0025] determining a target site according to the search cost function model and the function model;
[0026] The target site on the optimal route is marked as a flight site where a search task needs to be performed, thereby obtaining the optimal search path.
[0027] In a second aspect, an embodiment of the present disclosure provides a drone path planning device, the drone path planning device comprising:
[0028] The location determination module is used to determine the starting and stopping points and all flight locations in the search area;
[0029] A model construction module is used to construct a path planning function model for the UAV, wherein the path planning function model includes a path cost function model for the UAV returning to the start-stop base point after passing through all flight locations, a search cost function model at all target locations, and a search benefit model, wherein the target location is any location among the flight locations;
[0030] A model solving module, used to obtain the optimal solution of the path planning function model under the constraint of limited energy cost;
[0031] The path determination module is used to determine the optimal search path of the UAV in the search area according to the optimal solution of the path planning function model.
[0032] According to a specific implementation of the embodiment of the present disclosure, the model building module is used to:
[0033] The path planning problem of the UAV returning to the starting and stopping base point after passing through all flight locations is mapped to the traveling salesman problem of integer linear programming;
[0034] Constructing a traveling salesman cost function model according to predetermined path cost parameters as a path planning function model for the UAV;
[0035] The model solving module is used to:
[0036] The traveling salesman cost function model is solved using an ant colony algorithm to obtain an optimal route for determining an optimal search path.
[0037] According to a specific implementation of the embodiment of the present disclosure, the model building module is used to:
[0038] Obtaining online the search cost and search benefit of the UAV at each flight location;
[0039] According to the search cost of each flight location, an online decision function is used to construct a search cost function model corresponding to all flight locations;
[0040] According to the search cost of each flight location, an online decision function is used to construct a search benefit function model corresponding to all flight locations;
[0041] The model solving module is used to:
[0042] obtaining a limited energy cost of the UAV;
[0043] determining an available search cost based on the limited energy cost and the path cost corresponding to the optimal route;
[0044] Under the constraint of the available search cost, the optimal solution of the search cost function model and the search benefit function model is solved, and the flight location corresponding to the optimal solution is determined as the target location.
[0045] In a third aspect, an embodiment of the present disclosure further provides a drone, comprising a body, a memory, and a processor, wherein the memory stores a computer program, and when the processor runs the computer program, the drone path planning method described in any one of the first aspects is executed.
[0046] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium storing a computer program, which, when running on a processor, executes the drone path planning method described in any one of the first aspects.
[0047] The drone route planning scheme in the embodiment of the present disclosure includes: determining the start-stop base point and all flight locations in the search area; constructing a path planning function model for the drone, wherein the path planning function model includes a path cost function model for the drone returning to the start-stop base point after passing through all flight locations from the start-stop base point, a search cost function model at all target locations, and a search benefit function model, wherein the target location is any location among the flight locations; obtaining the optimal solution of the path planning function model under the constraint of limited energy cost; and determining the optimal search path of the drone in the search area based on the optimal solution of the path planning function model. The scheme of the present disclosure proposes a new online drone search planning problem to solve the decision-making challenges brought about by knowing both the benefits and costs of the search action in an online manner. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.
[0049] Figure 1 A flowchart of a drone path planning method provided by an embodiment of the present disclosure;
[0050] Figure 2 A schematic diagram of a path involved in the drone path planning method provided in an embodiment of the present disclosure;
[0051] Figure 3 A schematic diagram of the performance involved in the drone planning method provided in an embodiment of the present disclosure;
[0052] Figure 4 A schematic diagram of the path involved in the drone planning method provided in an embodiment of the present disclosure;
[0053] Figure 5 A block diagram of a drone path planning device according to an embodiment of the present disclosure. Implementation Method
[0054] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.
[0055] The following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.
[0056] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on this disclosure, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement an apparatus and / or practice a method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this apparatus and / or practice this method.
[0057] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The illustrations only show components related to the present disclosure and are not drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component can be changed at will, and the component layout type may also be more complicated.
[0058] Additionally, in the following description, specific details are provided to provide a thorough understanding of the examples. However, one skilled in the art will appreciate that the aspects described can be practiced without these specific details. Example
[0059] The present disclosure provides a method for planning a path for a drone.
[0060] For details, see Figure 1 ,The UAV path planning method provided mainly includes the following steps:
[0061] Step S101, determining the starting and stopping base points and all flight locations of the search area;
[0062] The drone path planning method provided in this embodiment is mainly used to provide a scenario in which a drone passes through multiple specific locations in a specific area to search for information. The preset area is defined as the search area, and the locations that need to be passed through are defined as flight locations. The search area also includes a location for the drone to take off and land, which is defined as the start and stop base point.
[0063] The energy cost that drones can provide is limited. The path costs and search benefits consumed by different locations are also different. In addition, the search energy costs required to be consumed at different flight locations are different. The limited energy cost is not enough to support the search consumption of all flight locations in the search area. Therefore, it is necessary to screen low-cost paths and target locations for executing search information and obtain the maximum search benefit. It should be noted that the purpose of drone path planning is to plan a travel route that covers all flight locations and determine the flight locations where search operations are performed, so as to obtain a larger search benefit at a smaller energy cost. In addition, the location where the search operation is performed can be defined as the target location. The drone should choose a route that only allows each location to be visited once, with the starting and ending points being the same point, that is, the start and stop base point.
[0064] Step S102: constructing a path planning function model for the UAV, wherein the path planning function model includes a path cost function model for the UAV returning to the start-stop base point after passing through all flight locations, a search cost function model at all target locations, and a search benefit function model, wherein the target location is any location among the flight locations;
[0065] Based on the above analysis, the UAV path planning solution is divided into two stages: determining the optimal path and determining the target location for the search operation. Considering that the cost of collecting information at the search point is usually not known in advance before the UAV arrives at the evaluation and search points, the corresponding search cost and search benefit must be determined online. Therefore, when constructing the path planning function model, it can be divided into three parts: the path cost function model for the UAV to return to the starting and stopping base point after passing through all flight locations, the search cost function model at all target locations, and the search benefit function model.
[0066] To facilitate path planning, the corresponding search area is network It is a directed network, and the flight sites in the search area correspond to a set of vertices in the directed network. , the paths between adjacent flight sites are the edges in the directed network, that is, the directed network consists of a set of edges .
[0067] At each search point At, use search cost (vector ) to represent the search energy cost, path cost (vector ) indicates that the edge The energy cost of moving. The limited energy budget of the UAV is represented by (vector ) represents the binary decision variable for search selection. If the drone chooses to consume energy To search for vertices ,but The corresponding payoff for searching at a vertex is expressed as .use (vector ) represents the binary decision variable for path selection. If the decision maker chooses to go through the edge ,but The purpose of path planning is to maximize The cumulative search revenue on Maximize, and the total cost is within the limited cost budget, that is
[0068] (1)
[0069] in represents the cumulative cost of the move, Represents the cumulative cost of the search.
[0070] like Figure 2 As shown, the starting and stopping base points of the drone are , its search cost and benefit pair is , the drone does not need to search for information on its base. and consume the corresponding path cost , the drone can reach the top , you can get the search cost and benefits online immediately The drone should select a route in advance and decide online whether to search for the vertices it encounters. The specific online acquisition of search costs and benefits is directly determined by the actual situation and will not be explained in detail in this solution.
[0071] Step S103, obtaining the optimal solution of the path planning function model under the constraint of limited energy cost;
[0072] In order to solve the problem of online search planning of drones, this embodiment adopts an ant colony algorithm to solve the path selection problem, and then performs online search decision-making through online linear programming.
[0073] Regarding the path planning stage, according to a specific implementation of an embodiment of the present disclosure, the step of constructing a path planning function model for the drone may specifically include:
[0074] The path planning problem of the UAV returning to the starting and stopping base point after passing through all flight locations is mapped to the traveling salesman problem of integer linear programming;
[0075] Constructing a traveling salesman cost function model according to predetermined path cost parameters as a path planning function model for the UAV;
[0076] The step of obtaining the optimal solution of the path planning function model under the constraint of limited energy cost includes:
[0077] The traveling salesman cost function model is solved using an ant colony algorithm to obtain an optimal route for determining an optimal search path.
[0078] Path selection can be modeled as an offline UAV path planning problem. Given the network G and the corresponding parameters c, the online parameters and r are unknown. In this case, in order to minimize the traveling salesman cost, this can be expressed as the traveling salesman problem (TSP) in integer linear programming (ILP). It is shown below:
[0079]
[0080] (2)
[0081] (3)
[0082] (4)
[0083] (5)
[0084] (6)
[0085] in is an auxiliary integer variable introduced to accumulate the path cost given the path selection vector y. Constraint (2) states that the drone can only visit one vertex after visiting vertex i, while constraint (3) ensures that the drone can only visit a vertex from the previous vertex. Constraints (4)-(6) are MTZ constraints, which enforce that the path is fully connected, i.e., without subloops.
[0086] The Traveling Salesman Problem (TSP) is an NP-hard problem. Ant Colony Algorithm can obtain satisfactory solutions when solving large-scale problems. Therefore, the first stage of offline UAV route planning is formulated as the Traveling Salesman Problem (TSP) and solved by Ant Colony Algorithm. Specifically, the suboptimal solution of ILP-TSP includes the suboptimal objective and the corresponding vertex search sequence ,in By path planning For simplicity, let , we can get . You can also use and A sequence of parameters representing the corresponding costs and benefits of online search.
[0087] Regarding the search and planning stage, according to another specific implementation of the embodiment of the present disclosure, the step of constructing a path planning function model for the drone includes:
[0088] Obtaining online the search cost and search benefit of the UAV at each flight location;
[0089] According to the search cost of each flight location, an online decision function is used to construct a search cost function model corresponding to all flight locations;
[0090] According to the search cost of each flight location, an online decision function is used to construct a search benefit function model corresponding to all flight locations.
[0091] Furthermore, according to a specific implementation of the embodiment of the present disclosure, the step of obtaining the optimal solution of the path planning function model under the constraint of limited energy cost includes:
[0092] obtaining a limited energy cost of the UAV;
[0093] determining an available search cost based on the limited energy cost and the path cost corresponding to the optimal route;
[0094] Under the constraint of the available search cost, the optimal solution of the search cost function model and the search benefit function model is solved, and the flight location corresponding to the optimal solution is determined as the target location.
[0095] Online search decision can be modeled as an online UAV search planning problem. In the previous stage, the optimal route for the UAV is obtained, and based on this, it is decided whether to search and collect information about the current flight location. The goal of this stage is to make Maximize the cumulative search profit on Maximize, after removing the path cost, the remaining energy supply should be greater than 0, that is Assume that all online data search costs The search reward r is known in advance, and the offline UAV search planning problem (OffSP) is formulated as the following integer linear programming:
[0096]
[0097] (7);
[0098] When all parameters are known, the optimal solution can be obtained and .
[0099] because r is learned online, that is, it cannot be obtained directly and Different from the setting of P-OffSP, the decision maker only knows the historical information before making a decision on the search option at time t. and current observation information and ,Right now Therefore, the online UAV search planning (OnSP) can be formulated as:
[0100]
[0101] in represents the online decision function of the decision maker. In order to design an approximate optimal function , first relax the integer constraints in P-OffSP , which leads to the following linear programming:
[0102] (7)
[0103] Then by introducing the dual decision variable and , we can get the dual planning of P-OffSP:
[0104]
[0105] (8)
[0106] use 、 and Denotes the optimal solution of the LP-OffSP and DLP-OffSP problems. According to the complementary conditions, we have:
[0107] (9)
[0108] ONSP's online solution algorithm is as follows:
[0109] Input: n, coefficient pairs displayed online ( )
[0110] Parameter: Learning rate
[0111] Output: A series of online search decisions x
[0112] 1: Another
[0113] 2: Initialization
[0114] 3:
[0115] 4: Setup
[0116] (10)
[0117] 5: Calculation
[0118]
[0119] 6: end for
[0120] 7: Return .
[0121] And when When the decision variable May be a non-integer.
[0122] Based on equation (9), an online algorithm can be designed to solve the online UAV search planning problem using the proposed online linear programming method. In order to evaluate the quality of the algorithm, the expected optimality gap between OffSP and OnSP, i.e., the average regret, can be calculated as follows:
[0123]
[0124] in and denote the cumulative benefits of the optimal offline solution and the online algorithm, respectively. When assuming that the online parameters r and a are sampled independently and identically distributed (iid) from an unknown distribution, which is usually reasonable in practical applications, Algorithm 3.2 approximates the optimal solution and achieves the OffSP problem. Average regret.
[0125] Step S104: determining an optimal search path for the UAV in the search area according to the optimal solution of the path planning function model.
[0126] Specifically, the step of determining the optimal search path of the UAV in the search area according to the optimal solution of the path planning function model includes:
[0127] determining a target site according to the search cost function model and the function model;
[0128] The target site on the optimal route is marked as a flight site where a search task needs to be performed, thereby obtaining the optimal search path.
[0129] In this way, online planning for the UAV path can be completed. On this basis, the performance of the proposed two-stage solution method for the online UAV search planning problem can also be evaluated on a commonly used dataset called TSPLIB. The programming and algorithm proposed in this paper were tested on a Windows 10 (64) computer equipped with an Intel Core-i7 CPU and 16.0 GB RAM using the Gurobi 9.0.1 solver.
[0130] First, the performance of the ant colony algorithm in offline UAV path planning is analyzed. On the problem pr1002 with 1002 cities, the process of minimizing the path cost using the ant colony algorithm is as follows: Figure 3 In this problem, the optimized route is as follows Figure 4 As shown in Figure 2, the cost after 100 iterations of the ant colony algorithm is 305,061, while the optimal cost is 259,045. The corresponding competitive ratio is about 1.18. Then, 10 pairs of All and r are uniformly distributed on [1,10].
[0131] For UAV search planning, the performance of online planning algorithm and offline optimization algorithm in several cases of different networks is shown in Table 1. The results show that the online planning algorithm proposed in this paper is more efficient than the offline optimization algorithm. When n is larger, an online near-optimal solution with a competition ratio greater than 90% can be obtained. As n increases, the competition ratio of the online planning algorithm increases, which is consistent with the The theoretical analysis of average regret is consistent. More importantly, in actual online decision-making applications, only online algorithms can be applied, and offline optimization solutions cannot be obtained.
[0132] Table 1
[0133]
[0134] The drone path planning method provided in the above embodiment solves the online decision-making problem in drone search planning, especially the situation where the benefits and costs of each search point are not available to the decision maker. A new type of online drone search planning problem is proposed to solve the decision-making challenge brought about by knowing the benefits and costs of the search action in an online manner. By decoupling the problem into a traveling salesman problem and an online resource planning problem, a two-stage solution process is proposed. Specifically, the method solves the traveling salesman problem through an ant colony algorithm to obtain route planning, performs online decision-making through online linear programming, and proves that the method is approximately optimal. The effectiveness of the two-stage method proposed in this paper has been verified in a wide range of application data sets, and the experimental results show the excellent performance of online search decision-making.
[0135] Example 2
[0136] In addition, an embodiment of the present disclosure provides a drone path planning device 500, wherein the drone path planning device 500 includes:
[0137] The location determination module 501 is used to determine the starting and stopping points and all flight locations in the search area;
[0138] Model construction module 502 is used to construct a path planning function model for the UAV, wherein the path planning function model includes a path cost function model for the UAV returning to the start-stop base point after passing through all flight locations, a search cost function model at all target locations, and a search benefit model, wherein the target location is any location among the flight locations;
[0139] A model solving module 503 is used to obtain the optimal solution of the path planning function model under the constraint of limited energy cost;
[0140] The path determination module 504 is configured to determine an optimal search path for the UAV in the search area according to the optimal solution of the path planning function model.
[0141] According to a specific implementation of the embodiment of the present disclosure, the model building module 502 is used to:
[0142] The path planning problem of the UAV returning to the starting and stopping base point after passing through all flight locations is mapped to the traveling salesman problem of integer linear programming;
[0143] Constructing a traveling salesman cost function model according to predetermined path cost parameters as a path planning function model for the UAV;
[0144] The model solving module 503 is used to:
[0145] The traveling salesman cost function model is solved using an ant colony algorithm to obtain an optimal route for determining an optimal search path.
[0146] According to a specific implementation of the embodiment of the present disclosure, the model building module 502 is used to:
[0147] Obtaining online the search cost and search benefit of the UAV at each flight location;
[0148] According to the search cost of each flight location, an online decision function is used to construct a search cost function model corresponding to all flight locations;
[0149] According to the search cost of each flight location, an online decision function is used to construct a search benefit function model corresponding to all flight locations;
[0150] The model solving module 503 is used to:
[0151] obtaining a limited energy cost of the UAV;
[0152] determining an available search cost based on the limited energy cost and the path cost corresponding to the optimal route;
[0153] Under the constraint of the available search cost, the optimal solution of the search cost function model and the search benefit function model is solved, and the flight location corresponding to the optimal solution is determined as the target location.
[0154] The drone path planning device provided in this embodiment first determines the start / stop base points and all flight locations in a search area, then constructs a path planning function model for the drone. The path planning function model includes a path cost function model for the drone returning to the start / stop base point after passing through all flight locations, a search cost function model at all target locations, and a search benefit function model, where a target location is any location among the flight locations. Under the constraint of a limited energy cost, the optimal solution of the path planning function model is obtained. Based on the optimal solution of the path planning function model, the optimal search path for the drone in the search area is determined. The disclosed solution proposes a new online drone search planning problem to address the decision-making challenges posed by knowing both the benefits and costs of a search operation online. The specific implementation process of the provided drone path planning device can be found in the specific implementation process of the drone path planning method provided in the aforementioned embodiment, and will not be further elaborated here.
[0155] Example 3
[0156] In addition, an embodiment of the present disclosure provides a drone, including a body, a memory, and a processor, wherein the memory stores a computer program, and when the computer program runs on the processor, it executes the drone path planning method provided in the above-mentioned method embodiment 1.
[0157] In addition, the drone also includes some functional modules not shown, which will not be described here.
[0158] In addition, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the drone path planning method provided in the aforementioned embodiment is implemented.
[0159] In this embodiment, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
[0160] The drone and computer-readable storage medium provided in this embodiment first determine the start / stop base point and all flight locations in a search area, then construct a path planning function model for the drone. The path planning function model includes a path cost function model for the drone traveling from the start / stop base point through all flight locations and back to the start / stop base point, a search cost function model for all target locations, and a search benefit function model, where a target location is any of the flight locations. Under the constraint of a limited energy cost, the optimal solution of the path planning function model is obtained. Based on the optimal solution of the path planning function model, the optimal search path for the drone within the search area is determined. The disclosed solution proposes a new online drone search planning problem to address the decision-making challenges posed by knowing both the benefits and costs of a search operation online. The specific implementation of the drone and computer-readable storage medium provided herein can be found in the specific implementation of the drone path planning method described in Example 1 above, and will not be repeated here to avoid repetition.
[0161] The above description is merely a specific embodiment of the present disclosure, but the scope of protection of the present disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this disclosure should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure should be based on the scope of protection of the claims.
Claims
1. A UAV path planning method, characterized in that: The UAV path planning method includes: Determine the starting and stopping points and all flight locations of the search area; Constructing a path planning function model for the UAV, wherein the path planning function model includes a path cost function model for the UAV to return to the start-stop base point after passing through all flight locations, a search cost function model at all target locations, and a search benefit function model, wherein the target location is any location among the flight locations; Under the constraint of limited energy cost, obtaining the optimal solution of the path planning function model; Determine the optimal search path of the UAV in the search area according to the optimal solution of the path planning function model; The step of constructing the path planning function model of the UAV includes: The path planning problem of the UAV returning to the starting and stopping base point after passing through all flight locations is mapped to the traveling salesman problem of integer linear programming; Constructing a traveling salesman cost function model according to predetermined path cost parameters as a path planning function model for the UAV; The offline UAV search planning problem is formulated as the following integer linear program: ; ,in Represents the corresponding benefit of searching at the vertex, the vector Represents the binary decision variables for search choices, vector Represents the consumption of search energy cost, vector represents the energy budget of the UAV, represents the path cost; The integer linear programming to relax the offline UAV search planning problem is: ; Introducing dual decision variables and , and obtain the dual planning: ; ； use 、 and Indicates a problem The optimal solution of DLP-OffSP is obtained according to the complementary conditions: ; Among them, setting the learning rate = ,e= , and initialize =0, and set ; =max{ ( ),0}; get .
2. The UAV path planning method according to claim 1, characterized in that: The step of determining an optimal search path for the UAV in the search area according to the optimal solution of the path planning function model comprises: determining a target site according to the search cost function model and the function model; The target site is marked as a flight site where a search mission needs to be performed, and the optimal search path is obtained.
3. A UAV path planning device, characterized in that: The UAV path planning device comprises: The location determination module is used to determine the starting and stopping points and all flight locations in the search area; A model construction module is used to construct a path planning function model for the UAV, wherein the path planning function model includes a path cost function model for the UAV returning to the start-stop base point after passing through all flight locations, a search cost function model at all target locations, and a search benefit model, wherein the target location is any location among the flight locations; A model solving module, used to obtain the optimal solution of the path planning function model under the constraint of limited energy cost; a path determination module, configured to determine an optimal search path for the UAV in the search area according to an optimal solution of the path planning function model; Wherein, the model building module is used to: The path planning problem of the UAV returning to the starting and stopping base point after passing through all flight locations is mapped to the traveling salesman problem of integer linear programming; Constructing a traveling salesman cost function model according to predetermined path cost parameters as a path planning function model for the UAV; The model solving module is used to: The offline UAV search planning problem is formulated as the following integer linear program: ; ,in Represents the corresponding benefit of searching at the vertex, the vector Represents the binary decision variables for search choices, vector Represents the consumption of search energy cost, vector represents the energy budget of the UAV, represents the path cost; The integer linear programming to relax the offline UAV search planning problem is: ; Introducing dual decision variables and , and obtain the dual planning: ; ； use and Indicates a problem The optimal solution of DLP-OffSP is obtained according to the complementary conditions: ; Among them, setting the learning rate = ,e= , and initialize =0, and set ; =max{ ( ),0}; get .
4. A drone, characterized in that: The method comprises a body, a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program, the method executes the UAV path planning method according to any one of claims 1 to 2.
5. A computer-readable storage medium, characterized in that It stores a computer program, which, when running on a processor, executes the drone path planning method according to any one of claims 1 to 2.