Unmanned aerial vehicle path planning method based on improved skill optimization algorithm

Through an improved skill optimization algorithm, combined with adaptive weights, disturbance and trap avoidance factors, the UAV path planning is optimized, which solves the problems of local optimality and poor power prediction accuracy, and achieves a more efficient and safe drone flight path and power management.

CN118915792AInactive Publication Date: 2024-11-08UNIV FOR SCI & TECH ZHENGZHOU
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411070395.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing skill optimization algorithms are prone to falling into local optimization in drone path planning and cannot obtain an ideal flight path. At the same time, the accuracy of power prediction is poor, which affects the reliability of power system management.

Method used

Using an improved skill optimization algorithm, the flight cost function is constructed by building a flight environment model, setting the starting point, target point and no-fly zone, and introducing adaptive weight factors, perturbation factors and trap avoidance factors during the iterative update process to optimize candidates Decompose the population to obtain the optimal track path.

Benefits of technology

It effectively avoids local optimal problems, improves the efficiency and safety of path planning, reduces the cost of drone flight, and improves the reliability of power system management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118915792A_ABST
    Figure CN118915792A_ABST
Patent Text Reader

Abstract

The invention provides an unmanned aerial vehicle path planning method based on an improved skill optimization algorithm, and the method comprises the steps: constructing a flight cost function of an unmanned aerial vehicle according to the flight cost and flight safety of the unmanned aerial vehicle; the candidate solution population is initialized; an improved skill optimization algorithm is adopted to carry out iteration updating of set times on the candidate solution population so as to obtain an optimized candidate solution population, and the iteration updating at least comprises the steps that in the exploration stage, according to the number of times of iteration updating, a self-adaptive weight factor is generated, and the number of times of iteration updating is set; the candidate solution population is updated according to the self-adaptive weight factor; and obtaining a target candidate solution from the optimized candidate solution population by using the flight cost function, and constructing a flight path of the unmanned aerial vehicle according to the starting point, the destination point and the target candidate solution. According to the technical scheme of the invention, local optimum in the process of planning the flight path of the unmanned aerial vehicle can be avoided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) path planning, and in particular to a UAV path planning method based on an improved skill optimization algorithm. Background Art

[0002] Drones have the advantages of strong maneuverability, controllability and flexibility, and are widely used in a wide range of work scenarios in the fields of military, logistics, agriculture, search and rescue, environmental monitoring, etc. Drone trajectory planning refers to a series of technologies and methods to design the optimal flight route from the starting point to the end point for the drone under given environment and mission requirements, while also meeting the requirements of avoiding obstacles, optimizing path length and flight time, and ensuring flight safety.

[0003] Existing UAV trajectory planning methods can be roughly divided into three categories: search-based algorithms, sampling-based algorithms, and intelligent optimization-based algorithms. Search-based algorithms include Dijkstra algorithm, fast expansion random number algorithm, and artificial trial field method. This type of algorithm is highly efficient and can find the global optimal trajectory, but it cannot handle the multi-constraint problem in the UAV path planning process. Although intelligent optimization algorithms have the advantages of simple parameters, flexible calculation, and fast search speed, they are prone to falling into local optimality.

[0004] The skill optimization algorithm is a type of intelligent optimization algorithm that simulates the human effort to acquire and improve skills, but it still has the problem of falling into local optimality and converging prematurely, and cannot make the obtained flight path achieve the ideal effect. Summary of the invention

[0005] The purpose of the present invention is to provide a UAV path planning method based on an improved skill optimization algorithm, which can solve the problem that in the process of planning the flight path of a UAV using a skill optimization algorithm in the prior art, the obtained flight path is not ideal due to falling into a local optimum.

[0006] It is used to solve the problem of poor accuracy in electricity consumption prediction in the prior art, so as to achieve the purpose of improving the reliability of power system management and operation.

[0007] Specifically, the present invention provides a UAV path planning method based on an improved skill optimization algorithm, the method comprising:

[0008] Constructing a flight environment model according to the working environment of the UAV, and setting a starting point, a target point and a no-fly zone in the flight environment model;

[0009] Constructing a flight cost function of the UAV according to the flight cost and flight safety of the UAV;

[0010] Initializing a candidate solution population, wherein the candidate solution population includes a plurality of candidate solutions, each of the candidate solutions including position information of a plurality of track points;

[0011] The candidate solution population is iteratively updated a set number of times using an improved skill optimization algorithm to obtain an optimized candidate solution population, wherein the iterative update at least includes: in an exploration phase, generating an adaptive weight factor according to the number of iterative updates, and updating the candidate solution population according to the adaptive weight factor;

[0012] The flight cost function is used to obtain a target candidate solution from the optimized candidate solution population, and a track path of the UAV is constructed according to the starting point, the destination point and the target candidate solution.

[0013] Furthermore, generating the adaptive weight factor according to the number of iterative updates includes:

[0014] Assume that the number of iterative updates is t, the set number is T, and the adaptive weight factor w in the tth iterative update is t ,but:

[0015]

[0016] Furthermore, the improved skill optimization algorithm is used to iteratively update the candidate solution population for a set number of times, and further includes:

[0017] In the development phase, a disturbance factor is generated according to the number of iterative updates, and the candidate solution population is updated according to the disturbance factor.

[0018] Furthermore, the generating of the disturbance factor according to the number of iterative updates includes:

[0019] Assume that the number of iterations is t, the number of settings is T, and the disturbance factor in the tth iteration is q t ,but:

[0020]

[0021] Wherein, a and g are both constants.

[0022] Furthermore, the iterative updating of the candidate solution population by using the improved skill optimization algorithm for a set number of times also includes:

[0023] In the development phase, a trap avoidance factor of the candidate solution population is obtained, and the candidate solution population is updated according to the trap avoidance factor.

[0024] Furthermore, the step of obtaining the trap avoidance factor of the candidate solution population includes:

[0025] Get the first random number u updated in the tth iteration 1,t and the second random number u 2,t , and suppose that in the t-th iteration update, the trap avoidance factor of the m-th track point in the i-th candidate solution is

[0026] In the first random number u 1,t When the value is less than the first set threshold, the following:

[0027]

[0028] In the first random number u 1,t When the value is greater than or equal to the first set threshold:

[0029]

[0030] Among them, θ 1,t is a random number with a value range greater than -1 and less than 1 in the t-th iteration update process, θ 2,t is a random number greater than -0.5 and less than 0.5 during the t-th iteration update process, r t is a random number whose value range is greater than or equal to 0 and less than or equal to 1 during the t-th iteration update process, K i,m,t The position information of the mth track point in the i-th candidate solution before the t-th iteration update, E i,m,t represents the expert set of the mth track point in the ith candidate solution in the tth iteration update, The position information of the m-th track point in the ith candidate solution at the next iteration update of the t-th iteration update.

[0031] Furthermore, constructing a flight cost function of the drone according to the flight cost and flight safety of the drone includes:

[0032] Constructing a track length cost function of the UAV according to the flight length of the human-machine;

[0033] Constructing a height cost function of the UAV according to the flight height of the UAV;

[0034] Constructing a turning angle cost function of the UAV according to the turning angle of the UAV;

[0035] The flight cost function is constructed according to the track length cost function, the altitude cost function and the turning angle cost function.

[0036] Furthermore, constructing the flight cost function according to the track length cost function, the altitude cost function and the turning angle cost function includes:

[0037] Respectively obtaining weight values ​​of the track length cost function, the height cost function and the turning angle cost function;

[0038] Using each of the weight values ​​to modify the track length cost function, the altitude cost function and the turning angle cost function;

[0039] The flight cost function is obtained according to the modified track length cost function, altitude cost function and turning angle cost function.

[0040] Furthermore, the flight environment model is constructed according to the working environment of the UAV, including:

[0041] The function simulation method is used to simulate the topographical features of the working environment, and an obstacle model is added according to the actual terrain of the working environment to construct the flight environment model.

[0042] Furthermore, the adopting of the improved skill optimization algorithm to iteratively update the candidate solution population for a set number of times includes:

[0043] After the candidate solution is updated in the exploration phase and / or after the candidate solution is updated in the development phase, the method further includes:

[0044] After the candidate solution is updated, the flight cost function is used to calculate the flight cost of the candidate solution before and after the update;

[0045] If the updated flight cost of the candidate solution is greater than the flight cost before the update, retain the candidate solution before the update;

[0046] If the flight cost of the candidate solution before updating is greater than the flight cost after updating, the updated candidate solution is retained.

[0047] The UAV path planning method of the present invention uses an improved skill optimization algorithm to iteratively optimize the candidate solution population for a set number of times to obtain the optimal candidate solution population, and selects the target candidate solution from the optimal candidate solution population according to the flight cost function of the UAV; then the UAV's track path is constructed according to the selected target candidate solution, the starting point and the target point to reduce the cost of the UAV flight and improve the safety of the UAV flight. In addition, in the technical solution of the present invention, an adaptive weight factor is introduced into the improved skill optimization algorithm in the exploration phase in the exploration phase, which can effectively avoid the problem of the algorithm falling into the local optimum and improve the work efficiency of iterative optimization of the candidate solution cluster.

[0048] Based on the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings, those skilled in the art will become more aware of the above and other objects, advantages and features of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Hereinafter, some specific embodiments of the present invention will be described in detail in an exemplary and non-limiting manner with reference to the accompanying drawings. The same reference numerals in the accompanying drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the accompanying drawings:

[0050] Figure 1 is a schematic flow chart of a UAV path planning method based on an improved skill optimization algorithm according to an embodiment of the present invention;

[0051] Figure 2 is a schematic flow chart of updating a candidate solution cluster during a development phase according to an embodiment of the present invention;

[0052] Figure 3 is a schematic flow chart of updating a candidate solution cluster during a development phase according to another embodiment of the present invention;

[0053] Figure 4 is a schematic flow chart of adding a threat area model to a flight environment model according to an embodiment of the present invention;

[0054] Figure 5 is a schematic flow chart of correcting an updated candidate solution according to a flight cost function according to another embodiment of the present invention. DETAILED DESCRIPTION

[0055] Refer to the following Figures 1 to 5 To describe a UAV path planning method based on an improved skill optimization algorithm in an embodiment of the present invention. In the description of this embodiment, it should be understood that the terms "first" and "second" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features, that is, include one or more of the features. In the description of the present invention, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. When a feature "includes or contains" one or some of the features it covers, unless otherwise specifically described, this indicates that other features are not excluded and may further include other features.

[0056] See also Figure 1 , Figure 1 What is shown is a UAV path planning method based on an improved skill optimization algorithm according to an embodiment of the present invention, the method comprising the following steps:

[0057] Step S101: constructing a flight environment model according to the working environment of the UAV, and setting a starting point, a target point and a no-fly zone in the flight environment model;

[0058] Step S102: constructing a flight cost function of the UAV according to the flight cost and flight safety of the UAV;

[0059] Step S103: Initializing a candidate solution population, wherein the candidate solution population includes multiple candidate solutions, and each candidate solution includes location information of multiple track points;

[0060] Step S104: using an improved skill optimization algorithm to iteratively update the candidate solution population for a set number of times to obtain an optimal candidate solution population;

[0061] Step S105: using the flight cost function of the drone to obtain a target candidate solution from the optimal candidate solution population, and constructing the trajectory path of the drone based on the target candidate solution, as well as the above starting point and target point.

[0062] In the above step S101, the flight environment model constructed is a simulation model for imitating the working environment of a drone, and the flight environment model is set in a three-dimensional coordinate system, in which the height direction is the z-axis direction and the plane where the x-axis and y-axis are located is the ground plane.

[0063] After building the flight environment model, obtain the starting point and target point of the UAV flight mission and set them in the flight environment model. Suppose the coordinates of the starting point are (x0, y0, z0) and the coordinates of the target point are (x N+1 ,y N+1 ,z N+1 ), where N is the total number of track points during the drone's flight mission. Then, according to the areas in the working environment that pose a threat to drone flight, no-fly zones are set in the flight environment, and the center position of each no-fly zone is obtained to avoid the no-fly zones during the process of controlling the drone to perform the flight mission, thereby improving the safety of drone flight.

[0064] In the above step S102, the flight distance and flight altitude of the drone will affect the flight cost of the drone, and the flight altitude of the drone and the distance from the no-fly zone will affect the flight safety of the drone. Therefore, the flight cost function of the drone can be constructed based on the flight distance, flight altitude and distance from the no-fly zone of the drone.

[0065] In the above step S103, the position information of each track point may be randomly acquired first to obtain a candidate solution, and then multiple candidate solutions may be obtained in sequence, thereby obtaining a candidate solution population composed of multiple candidate solutions, and completing the comprehensive initialization of the candidate solutions.

[0066] In the above step S104, the iterative update of the candidate solution population using the improved skill optimization algorithm includes an exploration phase and a development phase, wherein the exploration phase is to acquire skills from experts, which is equivalent to searching for new solutions in the optimization process; the development phase is the skill improvement based on time and personal efforts, that is, fine-tuning on the basis of the existing solution to obtain the optimal solution.

[0067] In this embodiment, during the iterative update of the candidate solution population, if the number of iterative updates is less than the set iteration threshold, the candidate solution cluster is updated in the exploration phase; if the number of iterative updates is greater than or equal to the set iteration threshold, the candidate solution cluster is updated in the development phase.

[0068] In the exploration phase, the methods for updating the candidate solution population include:

[0069] An adaptive weight factor is generated according to the number of iterative updates, and the candidate solution population is updated according to the adaptive weight factor. The formula used for the update is:

[0070]

[0071] in, is the i-th candidate solution after the t-th iteration update in the exploration phase; K represents the position information of the mth track point in the i-th candidate solution; i,m,t is the position information of the mth track point in the ith candidate solution before the tth iteration update; r t is a random number in the range [0,1] in the tth iteration update; E i,m,t I represents the expert set of the mth track point in the ith candidate solution in the tth iteration update, and the expert set contains the position information of the mth track point of a random candidate solution among the candidate solutions whose flight cost is smaller than that of the ith candidate solution in the tth iteration; t is the value randomly selected from the set {1,2} in the tth iteration update; w t is the adaptive weight factor in the tth iteration update, and the larger the number of iteration updates t, the larger the adaptive weight factor w t The smaller the value of .

[0072] In the exploration phase, the adaptive weight factor w t After being introduced into the improved skill optimization algorithm, the adaptive weight factor w t When it is larger, it can enhance the global search ability of the algorithm, and when the adaptive weight factor w tWhen it is small, the algorithm’s search accuracy can be improved and the algorithm’s search speed can be accelerated. Therefore, after introducing the adaptive weight factor, the algorithm’s search accuracy and search speed can be gradually reduced during the iterative update process, but the search capability can be gradually increased, thus avoiding the problem of falling into the local optimum during the exploration phase.

[0073] In this embodiment, in the development stage, the method of updating the candidate solution population using the improved skill optimization algorithm includes:

[0074] In the random number r t When less than 0.5:

[0075]

[0076] In the random number r t When greater than or equal to 0.5:

[0077]

[0078] in, is the i-th candidate solution after the t-th iteration update in the development phase, is the position information of the mth track point in the ith candidate solution, K i,m,t is the position information of the mth track point in the ith candidate solution before the tth iteration update in the development phase, t is the number of iteration updates, lb and ub are the lower and upper bounds of the search space respectively. Suppose the lower limits of the preset search space on the x-axis, y-axis and z-axis are x min ,y min and z min , the upper limits of the x-axis, y-axis, and z-axis are x max ,y max and z max , then the coordinates of lb are (x min ,y min ,z min ), the coordinates of ub are (x max ,y max ,z max ).

[0079] After the candidate solution population is iteratively updated a set number of times using the improved skill optimization algorithm, the obtained candidate solution population is the optimal candidate solution population.

[0080] In the above step S105, the flight cost function of the drone is used to calculate the flight cost of each candidate solution in the optimal candidate solution population, and then the candidate solution with the smallest flight cost is used as the target candidate solution. Then, each track point in the target candidate solution is connected in sequence, and the first track point in the target candidate solution is connected to the starting point, and the last track point in the target candidate solution is connected to the target point to obtain the track path of the drone.

[0081] In summary, the UAV path planning method of this embodiment uses an improved skill optimization algorithm to iteratively optimize the candidate solution population for a set number of times to obtain the optimal candidate solution population, and selects the target candidate solution from the optimal candidate solution population according to the flight cost function of the UAV; then, the UAV's track path is constructed according to the selected target candidate solution, the starting point and the target point to reduce the cost of UAV flight and improve the safety of UAV flight. In addition, in this embodiment, the improved skill optimization algorithm introduces an adaptive weight factor in the exploration stage, which can effectively avoid the problem of the algorithm falling into the local optimum and improve the work efficiency of iterative optimization of the candidate solution cluster.

[0082] In some embodiments of the present invention, the calculation formula used to generate the adaptive weight factor according to the number of iterative updates in step S104 is:

[0083]

[0084] In the above formula, t is the number of iterative updates, T is the total number of iterative updates, and w t is the self-use weight factor in the t-th iteration update process.

[0085] Through the technical solution of this embodiment, an adaptive weight factor can be calculated according to the number of iterations, so that the adaptive weight factor can change with the number of iterations, thereby improving the reliability of updating the candidate solution cluster.

[0086] In some embodiments of the present invention, during the development phase, an improved skill optimization algorithm is used to update the candidate solution population. Figure 2 As shown, including:

[0087] Step S201: generating a disturbance factor according to the number of iterative updates;

[0088] Step S202: updating the candidate population according to the obtained disturbance factor.

[0089] In this embodiment, the disturbance factor in the tth iteration update is assumed to be q t , then the method of using the improved skill optimization algorithm to iteratively update the candidate solution population for the tth time includes:

[0090] In the random number rt When less than 0.5:

[0091]

[0092] In the random number r t When greater than or equal to 0.5:

[0093]

[0094] Through the technical solution of this embodiment, the improved skill optimization algorithm introduces a disturbance factor in the development stage, which can gradually reduce the search range of the algorithm, thereby enhancing the local search capability, so as to effectively avoid falling into the problem of local optimality and further improve the work efficiency of iterative optimization of candidate solution clusters.

[0095] In some embodiments of the present invention, the improved skill optimization algorithm is used to iteratively update the candidate solution population, and the disturbance factor q is obtained according to the number of iterative updates t. t The methods include:

[0096]

[0097] Wherein, a and g are both constants, and the value of a is preferably 0.9, and the value of g is preferably 0.4.

[0098] Through the technical solution of this embodiment, the disturbance factor can be obtained according to the number of iterative updates, so as to improve the reliability of the candidate solution cluster update in the development stage.

[0099] In some embodiments of the present invention, during the development phase, an improved skill optimization algorithm is used to iteratively update the candidate solution population. Figure 3 As shown, including:

[0100] Step S301: Obtain the trap avoidance factor of the candidate solution population;

[0101] Step S302: updating the candidate solution population according to the obtained trap avoidance factor.

[0102] In this embodiment, the trap avoidance factor of the mth track point of the ith candidate solution in the tth iteration update is The methods for updating the candidate population during the development phase include:

[0103] First, a random value rand is obtained in the range of greater than or equal to 0 and less than or equal to 1; then, according to the random value rand and the trap avoidance factor, Get a new candidate solution population, including:

[0104] Assume that in the development phase of the tth iteration update, the trap avoidance factor is introduced as In the candidate solution population after , the mth track point of the ith candidate solution is but

[0105] When the random value rand is less than 0.6:

[0106]

[0107] When the random value rand is greater than or equal to 0.6:

[0108]

[0109] in, The position information of the m-th track point in the ith candidate solution at the next iteration update of the t-th iteration update, and

[0110]

[0111]

[0112] In the above formulas, r 1,t and r 2,t is a random number generated during the t-th iteration update process, and the value range of the random number is greater than 0 and less than 1; K i,t represents the i-th candidate solution in the candidate solution cluster before the t-th update, max(K i,t ) is the maximum value matrix of the i-th candidate solution, min(K i,t ) is the minimum matrix of the i-th candidate solution, and:

[0113]

[0114] in, is the i-th candidate solution K i,t The maximum value of each track point on the x-axis coordinate, is the candidate solution K i,t The maximum value of each track point on the y-axis coordinate, is the candidate solution K i,t The maximum value of each track point on the z-axis coordinate; is the candidate solution K i,t The minimum value of each track point on the x-axis coordinate, is the candidate solution K i,t The minimum value of each track point on the y-axis coordinate, is the candidate solution K i,t The minimum value of each track point on the z-axis coordinate.

[0115] During the development phase, the methods for updating the candidate solution population include:

[0116] In the random number r t When less than 0.5:

[0117]

[0118] In the random number r t When greater than or equal to 0.5:

[0119]

[0120] Through the technical solution of this embodiment, a trap avoidance factor is introduced in the development stage of the improved skill optimization algorithm. Since the trap avoidance factor generates a candidate solution with superior performance by using additional candidate solutions, the quality of the updated candidate solution is improved, so as to effectively avoid falling into the problem of local optimality and further improve the work efficiency of iterative optimization of candidate solution clusters.

[0121] In some embodiments of the present invention, the method of obtaining the trap avoidance factor of the candidate population includes:

[0122] First, get the first random number u in the tth iteration update 1,t and the second random number u 2,t , the acquisition methods include:

[0123] u 1,t =s t ×3×rand+(1-s t )

[0124] u 2,t =s t ×rand+(1-s t )

[0125] Among them, s t is the binary random number updated in the tth iteration, that is, s t The value is 0 or 1.

[0126] Then, if the first random number u 1,t Less than 0.5, then

[0127]

[0128] If the first random number u 1,t is greater than or equal to 0.5, then

[0129]

[0130] Among them, θ 1,t is a random number with a value range greater than -1 and less than 1, θ 2,t is a random number greater than -0.5 and less than 0.5.

[0131] In some embodiments of the present invention, in the above step S102, the method for constructing the flight cost function of the drone includes:

[0132] According to the flight length of the UAV, a cost function of the UAV's track length is constructed;

[0133] According to the flight altitude of the UAV, a flight altitude cost function of the UAV is constructed;

[0134] According to the turning angle of the UAV, construct the turning angle cost function of the UAV;

[0135] The flight cost function of the UAV is constructed based on the track length cost function, flight altitude cost function and turning angle cost function of the UAV.

[0136] In this embodiment, the starting point is set as the 0th track point, the target point is set as the M+1th track point, and the track length cost function of the UAV is P(path), then

[0137]

[0138] Among them, P j represents the distance between the j-th track point and the j-1-th track point, and

[0139] P j =||L(j)-L(j-1)||2

[0140] In the above formula, L(j) represents the jth track point.

[0141] In this embodiment, the coordinate of each track point on the z-axis is used as the height of the corresponding track point, and the height of the j-th track point is set as Z(j), and the flight height cost function of the drone is H(height), then

[0142]

[0143] in, is the average height of each track point, that is

[0144]

[0145] In this embodiment, let the segment vector of the j-th route be a j , where the first route is the route between the first track point and the starting point, the M+1th route is the route between the target point and the Mth track point, and the jth turning angle of the drone is γ j ,but

[0146]

[0147] Assume the turning function of the drone is T(turn), then:

[0148]

[0149] Among them, Φ is the upper threshold of the turning angle of the drone.

[0150] After obtaining the UAV's track length cost function P (path), flight altitude cost function H (height) and turning function T (turn), the above track length cost function, flight altitude cost function and turning function are fused to obtain the UAV's flight cost function.

[0151] Through the technical solution of this embodiment, the flight cost function of the UAV can be constructed according to the flight length, flight altitude and flight angle of the UAV, so that the flight cost function can comprehensively reflect the flight cost and flight safety of the UAV, thereby improving the reliability of the flight cost calculation of each candidate solution in the candidate solution cluster.

[0152] In some embodiments of the present invention, a method for constructing a flight cost function of a drone according to a track length cost function, a flight altitude cost function, and a turning angle cost function of the drone includes:

[0153] First, obtain the weights of the UAV's track length cost function, flight altitude cost function, and turning angle cost function;

[0154] Then, according to the above weights, the track length cost function, flight altitude cost function and turning angle cost function of the UAV are corrected respectively, and the flight cost function of the UAV is constructed according to the weights of the corrected track length cost function, flight altitude cost function and turning angle cost function.

[0155] In this embodiment, the weight v1 of the UAV's track length cost function, the flight height cost function v2 and the turning angle cost function v3 are assumed, and the UAV's flight cost function is F cost ,but:

[0156] F cost =v1P(path)+v2H(height)+v3T(turn)

[0157] and

[0158] v1+v2+v3=1

[0159] Through the technical solution of this embodiment, the flight cost function of the UAV can be constructed based on the corrected track length cost function, flight altitude cost function and turning angle cost function to improve the reliability of the flight cost function and ensure the reliability of the flight cost calculation of the candidate solutions in the candidate solution cluster.

[0160] In some embodiments of the present invention, the method of constructing a flight environment model according to the working environment of the drone includes:

[0161] The function simulation method is used to simulate the topographical features of the working environment, and the obstacle model is added according to the actual terrain of the working environment to construct the flight environment model.

[0162] In this embodiment, the expression of the function simulation method is:

[0163]

[0164] Among them, (x, y) is the coordinate of a point in the working environment projected on the horizontal plane, z(x, y) represents the height of the point, that is, the coordinate of the point on the z-axis; α, β, δ, ε, η and λ are constants, and the topographical features can be changed by changing these constants. Therefore, in this embodiment, the constants α, β, δ, ε, η and λ to simulate the topographic characteristics of the working environment.

[0165] In the actual terrain of the working environment, in addition to landform features, there are also obstacles such as mountains and hills that affect the flight of drones. Therefore, after simulating the landform features of the working environment, it is necessary to superimpose the obstacle model on the basis of the landform features.

[0166] In this embodiment, the expression of the superimposed obstacle model is as follows:

[0167]

[0168] In the above formula, h j Expressed as the height of the obstacle; h p is the reference terrain of the obstacle; (x pj ,y pj ) represents the coordinates of the center position of each j-th obstacle; a j and b j They represent the slopes of the j-th obstacle along the x-axis and y-axis respectively.

[0169] According to the above method, the flight environment model constructed is:

[0170] Z(x,y)=max[z(x,y),h(x,y)]

[0171] Furthermore, since there may be areas in the working environment that pose a threat to drone flight, such as radar areas, confidential areas, and areas with dense magnetic fields, drones are easily disturbed when flying in these areas, resulting in flight accidents. Therefore, after constructing the flight environment model, it is also necessary to add a threat area model to the flight environment model.

[0172] In this embodiment, the method for adding a threat area model to the flight environment model is as follows: Figure 4 As shown, including:

[0173] Step S401: obtaining the center position and area range of the threat area, and obtaining a circular area that can cover the area range;

[0174] Step S402: Taking the center position of the threat area as the center of the circle, a cylindrical model is established according to the center area of ​​the circle. The cylindrical model is the threat area model used to simulate the threat area.

[0175] Through the technical solution of this embodiment, the function simulation method can be used to simulate the topographical features of the working environment, and obstacle models and threat area models can be added to improve the accuracy of the simulation of the working environment, thereby improving the accuracy and reliability of the generated track path.

[0176] In some embodiments of the present invention, in step S104, after each iteration of updating the candidate solution population using the improved skill optimization algorithm, the updated candidate solution is corrected according to the flight cost function. The correction method is as follows: Figure 5 As shown, including:

[0177] Step S501: using a flight cost function to calculate the flight cost of the candidate solution before and after the update;

[0178] Step S502: Determine whether the flight cost of the candidate solution after updating is greater than the flight cost before updating;

[0179] Step S503: According to whether the flight cost of the candidate solution after updating is greater than the flight cost before updating, the candidate solution is modified, specifically including:

[0180] If the flight cost of the candidate solution after the update is greater than the flight cost before the update, the candidate solution before the update is retained and the candidate solution after the update is deleted;

[0181] If the updated flight cost of the candidate solution is less than the flight cost before the update, the updated candidate solution is retained and the candidate solution before the update is deleted.

[0182] Through the technical solution of this embodiment, it can be ensured that after the candidate solution population is updated, the flight cost of each candidate solution in the candidate solution population is less than or equal to the flight cost of each candidate solution in the candidate solution population before the update, thereby ensuring the reliability of the optimization of the candidate solution population.

[0183] At this point, those skilled in the art should recognize that, although multiple exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications that conform to the principles of the present invention can still be directly determined or derived based on the content disclosed in the present invention without departing from the spirit and scope of the present invention. Therefore, the scope of the present invention should be understood and recognized as covering all these other variations or modifications.

Claims

1. A UAV path planning method based on an improved skill optimization algorithm, characterized in that: The method comprises: Constructing a flight environment model according to the working environment of the UAV, and setting a starting point, a target point and a no-fly zone in the flight environment model; Constructing a flight cost function of the UAV according to the flight cost and flight safety of the UAV; Initializing a candidate solution population, wherein the candidate solution population includes a plurality of candidate solutions, each of the candidate solutions including position information of a plurality of track points; The candidate solution population is iteratively updated a set number of times using an improved skill optimization algorithm to obtain an optimized candidate solution population, wherein the iterative update at least includes: in an exploration phase, generating an adaptive weight factor according to the number of iterative updates, and updating the candidate solution population according to the adaptive weight factor; The flight cost function is used to obtain a target candidate solution from the optimized candidate solution population, and a track path of the UAV is constructed according to the starting point, the destination point and the target candidate solution.

2. The UAV path planning method according to claim 1, characterized in that: The step of generating the adaptive weight factor according to the number of iterative updates includes: Assume that the number of iterative updates is t, the set number is T, and the adaptive weight factor w in the tth iterative update is t ,but:

3. The UAV path planning method according to claim 1, characterized in that: The method of using the improved skill optimization algorithm to iteratively update the candidate solution population for a set number of times also includes: In the development phase, a disturbance factor is generated according to the number of iterative updates, and the candidate solution population is updated according to the disturbance factor.

4. The UAV path planning method according to claim 3, characterized in that: The generating of the disturbance factor according to the number of iterative updates comprises: Assume that the number of iterations is t, the number of settings is T, and the disturbance factor in the tth iteration is q t ,but: Wherein, a and g are both constants.

5. The UAV path planning method according to claim 1, characterized in that: The method of using the improved skill optimization algorithm to iteratively update the candidate solution population for a set number of times also includes: In the development phase, a trap avoidance factor of the candidate solution population is obtained, and the candidate solution population is updated according to the trap avoidance factor.

6. The UAV path planning method according to claim 5, characterized in that: The step of obtaining the trap avoidance factor of the candidate solution population includes: Get the first random number u updated in the tth iteration 1,t and the second random number u 2,t , and suppose that in the t-th iteration update, the trap avoidance factor of the m-th track point in the i-th candidate solution is In the first random number u 1,t When the value is less than the first set threshold, the following: In the first random number u 1,t When the value is greater than or equal to the first set threshold: Among them, θ 1,t is a random number with a value range greater than -1 and less than 1 in the t-th iteration update process, θ 2,t is a random number greater than -0.5 and less than 0.5 during the t-th iteration update process, r t is a random number whose value range is greater than or equal to 0 and less than or equal to 1 during the t-th iteration update process, K i,m,t The position information of the mth track point in the i-th candidate solution before the t-th iteration update, E i,m,t represents the expert set of the mth track point in the ith candidate solution in the tth iteration update, The position information of the m-th track point in the ith candidate solution at the next iteration update of the t-th iteration update.

7. The UAV path planning method according to claim 1, characterized in that: The step of constructing a flight cost function of the drone according to the flight cost and flight safety of the drone includes: Constructing a track length cost function of the UAV according to the flight length of the human-machine; Constructing a height cost function of the UAV according to the flight height of the UAV; Constructing a turning angle cost function of the UAV according to the turning angle of the UAV; The flight cost function is constructed according to the track length cost function, the altitude cost function and the turning angle cost function.

8. The UAV path planning method according to claim 7, characterized in that: The step of constructing the flight cost function according to the track length cost function, the altitude cost function and the turning angle cost function comprises: Respectively obtaining weight values ​​of the track length cost function, the height cost function and the turning angle cost function; Using each of the weight values ​​to modify the track length cost function, the altitude cost function and the turning angle cost function; The flight cost function is obtained according to the modified track length cost function, altitude cost function and turning angle cost function.

9. The UAV path planning method according to claim 1, characterized in that: The flight environment model is constructed according to the working environment of the UAV, including: The function simulation method is used to simulate the topographical features of the working environment, and an obstacle model is added according to the actual terrain of the working environment to construct the flight environment model.

10. The UAV path planning method according to claim 1, characterized in that: The step of using the improved skill optimization algorithm to iteratively update the candidate solution population for a set number of times includes: After the candidate solution is updated, the flight cost function is used to calculate the flight cost of the candidate solution before and after the update; If the updated flight cost of the candidate solution is greater than the flight cost before the update, retain the candidate solution before the update; If the flight cost of the candidate solution before updating is greater than the flight cost after updating, the updated candidate solution is retained.