Rescue unmanned aerial vehicle path-finding method and system based on improved whale algorithm

Abstract

The invention discloses a rescue unmanned aerial vehicle path-finding method and system based on an improved whale algorithm, and the method comprises the steps: collecting the current two-dimensional coordinates of an unmanned aerial vehicle, and setting the coordinates of a to-be-rescued person; constructing a target function by taking the minimum load power, the minimum cost and the shortest rescue distance as targets; substituting the current two-dimensional position of the unmanned aerial vehicle into the target function to evaluate the fitness, introducing Chebyshev mapping and adaptive weight to improve the whale algorithm, and using the improved whale algorithm to obtain the optimal next two-dimensional position of the unmanned aerial vehicle; forming an optimal operation path of the unmanned aerial vehicle according to the optimal two-dimensional position, and adjusting the direction of the unmanned aerial vehicle based on the optimal path; according to the method, the shortest distance, the minimum load power and the minimum rescue cost serve as objective functions, the whale algorithm is used for optimizing the rescue path, the rescue efficiency can be improved, and the rescue cost can be saved.

Description

Technical Field

[0001] The present invention relates to a rescue drone path-finding method and system, and in particular to a rescue drone path-finding method and system based on an improved whale algorithm, and belongs to the field of energy-saving control and intelligent transportation technology. Background Art

[0002] With the intensification of global climate change and environmental problems, natural disasters are occurring more frequently, especially in complex terrain areas such as grasslands, where the difficulty of search and rescue missions continues to increase. Globally, the number of grassland fires, missing persons and other incidents has increased year by year, placing a heavy burden on society and the environment. Traditional search and rescue activities often rely on a large number of manpower. However, the complex grassland terrain and severe weather conditions often make these methods inefficient, costly, and accompanied by extremely high safety risks. Therefore, the development of more efficient and intelligent search and rescue systems has become one of the key areas of technological progress.

[0003] Intelligent algorithms are an important part of intelligent search and rescue systems. In particular, intelligent algorithms play a vital role in path planning. The whale algorithm used in this patent is an innovative heuristic algorithm that imitates the predation behavior of whales and optimizes the search process by simulating the spiral encirclement and bubble net predation strategies of whales. The whale algorithm has been widely used in many fields such as path planning, image processing, neural network training and machine learning parameter optimization.

[0004] Although the Whale Optimization Algorithm (WOA) performs well in many optimization problems, it also has some defects and limitations. When dealing with complex environments, the traditional whale algorithm is prone to falling into local optimality, slow convergence, and premature convergence of the algorithm search process. This patent combines Chebyshev chaotic mapping and adaptive weight factors to effectively enhance the global search capability of the whale algorithm, speed up the convergence speed, reduce the risk of falling into local optimal solutions, and improve the stability and solution accuracy of the algorithm. Summary of the invention

[0005] Purpose of the invention: The purpose of the present invention is to provide a rescue drone path-finding method and system based on an improved whale algorithm that can improve rescue efficiency and save rescue costs.

[0006] Technical solution: The present invention discloses a method for finding a path for a rescue drone based on an improved whale algorithm, comprising:

[0007] (1) Collect the current two-dimensional coordinates of the drone and set the coordinates of the person to be rescued;

[0008] (2) Construct the objective function with the goal of minimum load power, minimum cost and shortest rescue distance;

[0009] (3) Substitute the current two-dimensional position of the UAV into the objective function to evaluate the fitness, introduce the Chebyshev mapping and adaptive weight to improve the whale algorithm, and use the improved whale algorithm to obtain the optimal next two-dimensional position of the UAV;

[0010] (4) The optimal operation path of the UAV is formed according to the optimal two-dimensional position, and the direction of the UAV is adjusted based on the optimal path.

[0011] Furthermore, the objective function formula in step (2) is as follows:

[0012]

[0013] Where Z is the cost, st is the constraint condition, constraint (a) is used to avoid forming a loop containing dangerous areas, and constraint (b) is used to ensure that the initial starting point and final point of the rescue drone are both the rescue center; N is the set of people to be rescued, N = {0, 1, ..., l}, K is the set of rescue drones, K = {1, 2, ..., m}; λ is used to balance the minimum cost and minimum power consumption, and is evenly distributed in [0, 1]; c d is the cost of launching a rescue drone; c t is the fuel consumption cost per unit distance of a single rescue drone; is the distance between the rescue drone k and the rescued person i and the rescued person j; is the driving variable of the rescue drone k from the rescued person i to the rescued person j.

[0014] Furthermore, the whale algorithm in step (3) is implemented as follows:

[0015] Initialization parameters, whale population, maximum number of iterations t max , the current location of the drone; the population matrix of the whale algorithm is as follows:

[0016]

[0017] Where X is the whale population; X g is the g-th whale, g=1,2,…,G; x gh is the position of the g-th whale in the h-th dimension, h = 1, 2, …, d; G is the size of the whale population; d is the dimension of the solution;

[0018] Assume that the position of the whale closest to the prey is equivalent to the current optimal solution. The other whales update their positions according to the best whale position and gradually move closer to surround the prey. The mathematical model is shown in equations (3) and (4):

[0019]

[0020]

[0021] Where t represents the current iteration number; It represents the optimal position that the whale population has experienced so far; Represents the current whale position; and is the coefficient vector, which is determined by equations (5) and (6) respectively; by adjusting and The value of can control the whale to reach any position near the best whale; among them, as the number of iterations t increases, Decrease linearly from 2 to 0; is a random vector in [0,1], where t max Indicates the maximum number of iterations;

[0022] Update the position of whale population members in the search space. Let P be a random number between [0,1]. When P≥0.5, whales capture food in a spiral approach. Its mathematical model is shown in formula (8). When P<0.5, whales shrink the encirclement. The formula for whales to surround prey in a spiral form is shown in (9):

[0023]

[0024] Where b is a constant coefficient used to limit the logarithmic spiral form; l represents a random number between [-1,1];

[0025]

[0026] In the formula It means that the whale randomly selects the location of an individual from the population as the target location;

[0027] In the prey search phase, the whale conducts global exploration. When |A| < 1 control parameter, it searches for the local optimal solution, and its mathematical model is formula (11); when |A| ≥ 1 control parameter, the algorithm conducts random search, and its mathematical model is formula (12);

[0028]

[0029] Calculate the fitness of each individual and record the optimal individual position;

[0030] Determine whether the termination condition is met, that is, whether the destination is reached. If the termination condition is not met, repeat this step; if the constraint condition is met, output the optimal operation path.

[0031] Furthermore, the step (3) introduces Chebyshev mapping and adaptive weights to improve the whale algorithm, including:

[0032] Introducing an adaptive weight factor, in the early stage of iteration, the whale uses a larger weight to update the position, improving the global search ability of the whale algorithm. In the later stage of iteration, the whale uses a smaller weight to update the position, improving the local optimization ability of the whale algorithm. The improved formula is as follows:

[0033]

[0034] The Chebyshev chaotic map is used to generate population positions and enhance the diversity of whale populations. Its mathematical expression is as follows:

[0035] x n+1 = cos(α arccos x n ),x n ∈[-1,1] (15)

[0036] Where x n is the value at the nth iteration, n is a positive integer, α is a parameter, and its value is an integer greater than 1.

[0037] Based on the same inventive concept, the present invention also provides a rescue drone path-finding system based on an improved whale algorithm, comprising:

[0038] The initialization module is used to collect the current two-dimensional coordinates of the drone and set the coordinates of the person to be rescued;

[0039] The objective function building module is used to build the objective function with the minimum load power, the minimum cost and the shortest rescue distance as the goals;

[0040] The calculation module is used to substitute the current two-dimensional position of the UAV into the objective function to evaluate the fitness, introduce Chebyshev mapping and adaptive weight to improve the whale algorithm, and use the improved whale algorithm to obtain the optimal next two-dimensional position of the UAV;

[0041] The path-finding module is used to form the optimal operation path of the drone according to the optimal two-dimensional position and adjust the direction of the drone based on the optimal path.

[0042] Furthermore, the objective function formula of the objective function construction module is as follows:

[0043]

[0044] Where Z is the cost, st is the constraint condition, constraint (a) is used to avoid forming a loop containing dangerous areas, and constraint (b) is used to ensure that the initial starting point and final point of the rescue drone are both the rescue center; N is the set of people to be rescued, N = {0, 1, ..., l}, K is the set of rescue drones, K = {1, 2, ..., m}; λ is used to balance the minimum cost and minimum power consumption, and is evenly distributed in [0, 1]; cd is the cost of launching a rescue drone; c t is the fuel consumption cost per unit distance of a single rescue drone; is the distance between the rescue drone k and the rescued person i and the rescued person j; is the driving variable of the rescue drone k from the rescued person i to the rescued person j.

[0045] Furthermore, the whale algorithm of the calculation module is implemented as follows:

[0046] Initialization parameters, whale population, maximum number of iterations t max , the current location of the drone; the population matrix of the whale algorithm is as follows:

[0047]

[0048] Where X is the whale population; X g is the g-th whale, g=1,2,…,G; x gh is the position of the g-th whale in the h-th dimension, h = 1, 2, …, d; G is the size of the whale population; d is the dimension of the solution;

[0049] Assume that the position of the whale closest to the prey is equivalent to the current optimal solution. The other whales update their positions according to the best whale position and gradually move closer to surround the prey. The mathematical model is shown in equations (3) and (4):

[0050]

[0051] Where t represents the current iteration number; It represents the optimal position that the whale population has experienced so far; Represents the current whale position; and is the coefficient vector, which is determined by equations (5) and (6) respectively; by adjusting and The value of can control the whale to reach any position near the best whale; among them, as the number of iterations t increases, Decrease linearly from 2 to 0; is a random vector in [0,1], where t max Indicates the maximum number of iterations;

[0052] Update the position of whale population members in the search space. Let P be a random number between [0,1]. When P≥0.5, whales capture food in a spiral approach. Its mathematical model is shown in formula (8). When P<0.5, whales shrink the encirclement. The formula for whales to surround prey in a spiral form is shown in (9):

[0053]

[0054] Where b is a constant coefficient used to limit the logarithmic spiral form; l represents a random number between [-1,1];

[0055]

[0056] In the formula It means that the whale randomly selects the location of an individual from the population as the target location;

[0057] In the prey search phase, the whale conducts global exploration. When |A| < 1 control parameter, it searches for the local optimal solution, and its mathematical model is formula (11); when |A| ≥ 1 control parameter, the algorithm conducts random search, and its mathematical model is formula (12);

[0058]

[0059] Calculate the fitness of each individual and record the optimal individual position;

[0060] Determine whether the termination condition is met, that is, whether the destination is reached. If the termination condition is not met, repeat this step; if the constraint condition is met, output the optimal operation path.

[0061] Furthermore, the calculation module introduces Chebyshev mapping and adaptive weights to improve the whale algorithm, including:

[0062] Introducing an adaptive weight factor, in the early stage of iteration, the whale uses a larger weight to update the position, improving the global search ability of the whale algorithm. In the later stage of iteration, the whale uses a smaller weight to update the position, improving the local optimization ability of the whale algorithm. The improved formula is as follows:

[0063]

[0064] The Chebyshev chaotic map is used to generate population positions and enhance the diversity of whale populations. Its mathematical expression is as follows:

[0065] x n+1 = cos(α arccos x n ),x n ∈[-1,1] (15)

[0066] Where x n is the value at the nth iteration, n is a positive integer, α is a parameter, and its value is an integer greater than 1.

[0067] Based on the same inventive concept, the present invention also provides a computing device, including: one or more processors, one or more memories and one or more programs, wherein the programs are stored in the memories and configured to be executed by the processors, and when the programs are loaded into the processors, the steps of the rescue drone pathfinding method based on the improved whale algorithm as described in any one of the above items are implemented.

[0068] Based on the same inventive concept, the present invention also provides a storage medium, which stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a processor, the processor executes the steps of the rescue drone pathfinding method based on the improved whale algorithm according to any of the above items.

[0069] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: 1. The present invention takes minimum cost (minimum load power, minimum rescue drones and shortest rescue distance) as the objective function, and uses the whale algorithm to optimize the rescue path, which can improve rescue efficiency and save rescue costs; in addition, the present invention optimizes the algorithm by introducing an adaptive weight factor; 2. The present invention introduces the Chebyshev chaotic map, which improves the global search ability of the algorithm through its good randomness and ergodicity, and avoids falling into local optimality; combined with the adaptive weight factor, the algorithm can dynamically adjust the search intensity according to different stages, enhance the global search in the early stage, and improve the local accuracy in the later stage; this improvement effectively accelerates the convergence speed, improves the optimization performance and stability, and adapts to the path planning needs of complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0071] Figure 2 This is a flow chart of the whale algorithm according to an embodiment of the present invention;

[0072] Figure 3 It is a graph of the iteration of the whale algorithm of an embodiment of the present invention;

[0073] Figure 4 It is an iterative curve diagram of an improved whale algorithm based on Chebyshev mapping according to an embodiment of the present invention;

[0074] Figure 5 It is an iterative curve diagram of the whale algorithm based on Chebyshev mapping and weight improvement according to an embodiment of the present invention;

[0075] Figure 6 It is an improved whale algorithm path planning diagram of an embodiment of the present invention;

[0076] Figure 7 is a path planning diagram for different obstacles of the improved whale algorithm according to an embodiment of the present invention;

[0077] Figure 8 This is a whale algorithm path planning diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0078] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0079] As attached Figure 1 As shown, the rescue drone path-finding method based on the improved whale algorithm of this embodiment includes:

[0080] (1) Collect the current two-dimensional coordinates of the drone and set the coordinates of the person to be rescued;

[0081] (2) Construct the objective function with the goal of minimum load power, minimum cost and shortest rescue distance;

[0082] (3) Substitute the current two-dimensional position of the UAV into the objective function to evaluate the fitness, introduce the Chebyshev mapping and adaptive weight to improve the whale algorithm, and use the improved whale algorithm to obtain the optimal next two-dimensional position of the UAV;

[0083] (4) The optimal operation path of the UAV is formed according to the optimal two-dimensional position, and the direction of the UAV is adjusted based on the optimal path.

[0084] Specifically, in step (1), the two-dimensional position of the drone and the coordinates of the destination and the end point are collected.

[0085] The objective function formula and constraints in step (2) are as follows:

[0086]

[0087] Where Z is the cost, st is the constraint condition, constraint (a) is used to avoid forming a loop containing dangerous areas, and constraint (b) is used to ensure that the initial starting point and final point of the rescue drone are both the rescue center; N is the set of people to be rescued, N = {0, 1, ..., l}, K is the set of rescue drones, K = {1, 2, ..., m}; λ is used to balance the minimum cost and minimum power consumption, and is evenly distributed in [0, 1]; c d is the cost of launching a rescue drone; c t is the fuel consumption cost per unit distance of a single rescue drone; is the distance between the rescue drone k and the rescued person i and the rescued person j; is the driving variable of the rescue drone k from the rescued person i to the rescued person j.

[0088] In step (3), if Figure 2 As shown in the figure, the process of using the improved whale algorithm to obtain the optimal next three-dimensional position of the rescue drone is as follows:

[0089] Initialization parameters, whale population, maximum number of iterations t max , the current location of the drone; the population matrix of the whale algorithm is as follows:

[0090]

[0091] Where X is the whale population; X g is the g-th whale, g=1,2,…,G; x gh is the position of the gth whale in the hth dimension, h = 1, 2, …, d; G is the size of the whale population; d is the dimension of the solution.

[0092] Whales collect the location information of their prey and surround them, but since whales cannot know the location of their prey in advance, they assume that the location of the whale closest to the prey is equivalent to the current optimal solution. Other whales update their positions based on the best whale position and gradually move closer to surround the prey. The mathematical model is shown in equations (3) and (4):

[0093]

[0094] Where t represents the current iteration number; It represents the optimal position that the whale population has experienced so far; Represents the current whale position; and is the coefficient vector, which is determined by equations (5) and (6) respectively; by adjusting and The value of can control the whale to reach any position near the best whale; among them, as the number of iterations t increases, Decrease linearly from 2 to 0; is a random vector in [0,1], where t max Indicates the maximum number of iterations;

[0095] Update the position of whale population members in the search space. The whale bubble net attacks prey in two ways: shrinking encirclement mechanism and spiral approach mechanism. P is a random number between [0,1]. When P≥0.5, the whale captures food in a spiral approach. Its mathematical model is shown in formula (7). When P<0.5, the whale shrinks the encirclement circle. The formula for the whale to surround the prey in a spiral form is shown in (8):

[0096]

[0097] Where b is a constant coefficient used to limit the logarithmic spiral form, usually 1; l represents a random number between [-1,1].

[0098]

[0099] In the formula It means that the whale randomly selects the location of an individual from the population as the target location;

[0100] Searching for prey (global exploration phase) In this phase, the whale conducts global exploration. When |A| < 1 control parameter, a local optimal solution search is performed, and its mathematical model is formula (11); when |A| ≥ 1 control parameter, the algorithm performs a random search, and its mathematical model is formula (12).

[0101]

[0102] Calculate the fitness of each individual and record the optimal individual position.

[0103] Determine whether the termination condition is met, that is, whether the destination is reached. If the termination condition is not met, repeat this step; if the constraint condition is met, output the optimal operation path.

[0104] Furthermore, the step (3) introduces Chebyshev mapping and adaptive weights to improve the whale algorithm, including:

[0105] Introducing an adaptive weight factor, in the early stage of iteration, the whale uses a larger weight to update the position, improving the global search ability of the whale algorithm. In the later stage of iteration, the whale uses a smaller weight to update the position, improving the local optimization ability of the whale algorithm. The improved formula is as follows:

[0106]

[0107] The Chebyshev chaotic map is used to generate population positions and enhance the diversity of whale populations. Its mathematical expression is as follows:

[0108] x n+1 = cos(α arccos x n ),x n ∈[-1,1] (15)

[0109] Where x n is the value at the nth iteration, n is a positive integer, α is a parameter, and its value is an integer greater than 1.

[0110] This embodiment mainly uses the whale optimization algorithm to realize the shortest path planning between the rescue center and the distress point. When planning the path, it is necessary to first obtain environmental information and establish an environmental map. The present invention uses the grid method to establish the environmental map for simulation. The simulation adopts the whale optimization algorithm to achieve the fastest traversal of evenly distributed coordinate points in the area. The algorithm simulation process is completed in the Matlab environment. Simulation areas are established for different environments and dangerous areas are regarded as obstacles, that is, unreachable areas. The obstacles of path planning are randomly set during the simulation process. In real life, it is completed by combining actual weather maps and electronic maps. The coordinate points of the fastest traversal in the area are randomly and evenly generated by the simulation software.

[0111] like Figure 3 , Figure 4 and Figure 5 As shown in the figure, the curve of the average fitness value of individuals in the whale algorithm population changing with the number of iterations is obtained through simulation. Figure 3 This is the convergence curve of the whale algorithm. After about 9 iterations, the algorithm converges and the fitness values ​​of individuals in the population are all 37.8995. Figure 4 This is the convergence curve of the whale algorithm improved based on Chebyshev mapping. After about 20 iterations, the algorithm converges and the fitness values ​​of individuals in the population are all 36.1421. It can be seen from the figure that the whale algorithm improved by Chebyshev mapping can generate population positions with good randomness, which helps to improve the stability of the algorithm. Figure 5 This is the convergence curve of the improved whale algorithm based on adaptive weights and Chebyshev mapping. After about 22 iterations, the algorithm converges and the fitness values ​​of individuals in the population are all 32.1421. It can be seen from the figure that the whale algorithm combined with adaptive weights and Chebyshev mapping can effectively improve the global search capability of the whale algorithm, accelerate the convergence speed, reduce local optimal traps, and improve the stability and solution accuracy of the algorithm.

[0112] like Figure 6 and Figure 7 As shown, in order to verify the effectiveness and stability of the improved whale algorithm, this embodiment models two different grid maps of obstacle areas to restore the actual scene to the greatest extent. The obstacles are modeled between the starting point and the end point. Figure 6 and Figure 7 It can be clearly seen that when the rescue drones have the same starting point and end point, but the areas of the obstacles are different, the simulation results obtain different paths. Therefore, the simulation results show that the whale optimization algorithm used in the present invention can adapt well to changes in environmental information. Figure 7 and Figure 8 As shown in the figure, for the rescue drone in the same map, the whale optimization algorithm has higher solution accuracy and better path planning effect than the whale algorithm.

[0113] Based on the same inventive concept, this embodiment also provides a rescue drone path-finding system based on an improved whale algorithm, including:

[0114] The initialization module is used to collect the current two-dimensional coordinates of the drone and set the coordinates of the person to be rescued;

[0115] The objective function building module is used to build the objective function with the minimum load power, the minimum cost and the shortest rescue distance as the goals;

[0116] The calculation module is used to substitute the current two-dimensional position of the UAV into the objective function to evaluate the fitness, introduce Chebyshev mapping and adaptive weight to improve the whale algorithm, and use the improved whale algorithm to obtain the optimal next two-dimensional position of the UAV;

[0117] The path-finding module is used to form the optimal operation path of the drone according to the optimal two-dimensional position and adjust the direction of the drone based on the optimal path.

[0118] Furthermore, the objective function formula of the objective function construction module is as follows:

[0119]

[0120] Where Z is the cost, st is the constraint condition, constraint (a) is used to avoid forming a loop containing dangerous areas, and constraint (b) is used to ensure that the initial starting point and final point of the rescue drone are both the rescue center; N is the set of people to be rescued, N = {0, 1, ..., l}, K is the set of rescue drones, K = {1, 2, ..., m}; λ is used to balance the minimum cost and minimum power consumption, and is evenly distributed in [0, 1]; c d is the cost of launching a rescue drone; c t is the fuel consumption cost per unit distance of a single rescue drone; is the distance between the rescue drone k and the rescued person i and the rescued person j; is the driving variable of the rescue drone k from the rescued person i to the rescued person j.

[0121] Furthermore, the whale algorithm of the calculation module is implemented as follows:

[0122] Initialization parameters, whale population, maximum number of iterations t max , the current location of the drone; the population matrix of the whale algorithm is as follows:

[0123]

[0124] Where X is the whale population; X g is the g-th whale, g=1,2,…,G; x ghis the position of the g-th whale in the h-th dimension, h = 1, 2, …, d; G is the size of the whale population; d is the dimension of the solution;

[0125] Assume that the position of the whale closest to the prey is equivalent to the current optimal solution. The other whales update their positions according to the best whale position and gradually move closer to surround the prey. The mathematical model is shown in equations (3) and (4):

[0126]

[0127] Where t represents the current iteration number; It represents the optimal position that the whale population has experienced so far; Represents the current whale position; and is the coefficient vector, which is determined by equations (5) and (6) respectively; by adjusting and The value of can control the whale to reach any position near the best whale; among them, as the number of iterations t increases, Decrease linearly from 2 to 0; is a random vector in [0,1], where t max Indicates the maximum number of iterations;

[0128] Update the position of whale population members in the search space. Let P be a random number between [0,1]. When P≥0.5, whales capture food in a spiral approach. Its mathematical model is shown in formula (8). When P<0.5, whales shrink the encirclement. The formula for whales to surround prey in a spiral form is shown in (9):

[0129]

[0130] Where b is a constant coefficient used to limit the logarithmic spiral form; l represents a random number between [-1,1].

[0131]

[0132] In the formula It means that the whale randomly selects the location of an individual from the population as the target location;

[0133] In the prey search phase, the whale conducts global exploration. When |A| < 1 control parameter, it searches for the local optimal solution, and its mathematical model is formula (11); when |A| ≥ 1 control parameter, the algorithm conducts random search, and its mathematical model is formula (12);

[0134]

[0135] Calculate the fitness of each individual and record the optimal individual position;

[0136] Determine whether the termination condition is met, that is, whether the destination is reached. If the termination condition is not met, repeat this step; if the constraint condition is met, output the optimal operation path.

[0137] Furthermore, the calculation module introduces Chebyshev mapping and adaptive weights to improve the whale algorithm, including:

[0138] Introducing an adaptive weight factor, in the early stage of iteration, the whale uses a larger weight to update the position, improving the global search ability of the whale algorithm. In the later stage of iteration, the whale uses a smaller weight to update the position, improving the local optimization ability of the whale algorithm. The improved formula is as follows:

[0139]

[0140] The Chebyshev chaotic map is used to generate population positions and enhance the diversity of whale populations. Its mathematical expression is as follows:

[0141] x n+1 = cos(α arccos x n ),x n ∈[-1,1] (15)

[0142] Where x n is the value at the nth iteration, n is a positive integer, α is a parameter, and its value is an integer greater than 1.

[0143] Based on the same inventive concept, this embodiment also provides a computing device, including: one or more processors, one or more memories, and one or more programs, wherein the programs are stored in the memories and configured to be executed by the processors, and when the programs are loaded into the processors, the steps of the rescue drone pathfinding method based on the improved whale algorithm as described in any one of the above items are implemented.

[0144] Based on the same inventive concept, this embodiment also provides a storage medium, which stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a processor, the processor executes the steps of the rescue drone pathfinding method based on the improved whale algorithm according to any of the above items.

Claims

1. A rescue drone pathfinding method based on an improved whale algorithm, characterized in that: include: (1) Collect the current two-dimensional coordinates of the drone and set the coordinates of the person to be rescued; (2) Construct the objective function with the goal of minimum load power, minimum cost and shortest rescue distance; (3) Substitute the current two-dimensional position of the UAV into the objective function to evaluate the fitness, introduce the Chebyshev mapping and adaptive weight to improve the whale algorithm, and use the improved whale algorithm to obtain the optimal next two-dimensional position of the UAV; (4) The optimal operation path of the UAV is formed according to the optimal two-dimensional position, and the direction of the UAV is adjusted based on the optimal path.

2. The rescue drone path-finding method based on the improved whale algorithm according to claim 1 is characterized in that: The objective function formula in step (2) is as follows: Where Z is the cost, st is the constraint condition, constraint (a) is used to avoid forming a loop containing dangerous areas, and constraint (b) is used to ensure that the initial starting point and final point of the rescue drone are both the rescue center; N is the set of people to be rescued, N = {0, 1, ..., l}, K is the set of rescue drones, K = {1, 2, ..., m}; λ is used to balance the minimum cost and minimum power consumption, and is evenly distributed in [0, 1]; c d is the cost of launching a rescue drone; c t is the fuel consumption cost per unit distance of a single rescue drone; is the distance between the rescue drone k and the rescued person i and the rescued person j; is the driving variable of the rescue drone k from the rescued person i to the rescued person j.

3. The rescue drone path-finding method based on the improved whale algorithm according to claim 1 is characterized in that: The whale algorithm described in step (3) is implemented as follows: Initialization parameters, whale population, maximum number of iterations t max , the current location of the drone; the population matrix of the whale algorithm is as follows: Where X is the whale population; X g is the g-th whale, g=1,2,…,G; x g is the position of the g-th whale in the h-th dimension, h = 1, 2, …, d; G is the size of the whale population; d is the dimension of the solution; Assume that the position of the whale closest to the prey is equivalent to the current optimal solution. The other whales update their positions according to the best whale position and gradually move closer to surround the prey. The mathematical model is shown in equations (3) and (4): Where t represents the current iteration number; It represents the optimal position that the whale population has experienced so far; Represents the current whale position; and is the coefficient vector, which is determined by equations (5) and (6) respectively; by adjusting and The value of can control the whale to reach any position near the best whale; among them, as the number of iterations t increases, Decrease linearly from 2 to 0; is a random vector in [0,1], where t max Indicates the maximum number of iterations; Update the position of whale population members in the search space. Let P be a random number between [0,1]. When P≥0.5, whales capture food in a spiral approach. Its mathematical model is shown in formula (8). When P<0.5, whales shrink the encirclement. The formula for whales to surround prey in a spiral form is shown in (9): Where b is a constant coefficient used to limit the logarithmic spiral form; l represents a random number between [-1,1]; In the formula It means that the whale randomly selects the location of an individual from the population as the target location; In the prey search phase, the whale conducts global exploration. When |A| < 1 control parameter, it searches for the local optimal solution, and its mathematical model is formula (11); when |A| ≥ 1 control parameter, the algorithm conducts random search, and its mathematical model is formula (12); Calculate the fitness of each individual and record the optimal individual position; Determine whether the termination condition is met, that is, whether the destination is reached. If the termination condition is not met, repeat this step; if the constraint condition is met, output the optimal operation path.

4. The rescue drone path-finding method based on the improved whale algorithm according to claim 3 is characterized in that: Step (3) introduces Chebyshev mapping and adaptive weights to improve the whale algorithm, including: Introducing an adaptive weight factor, in the early stage of iteration, the whale uses a larger weight to update the position, improving the global search ability of the whale algorithm. In the later stage of iteration, the whale uses a smaller weight to update the position, improving the local optimization ability of the whale algorithm. The improved formula is as follows: The Chebyshev chaotic map is used to generate population positions and enhance the diversity of whale populations. Its mathematical expression is as follows: x n+1 =cos(αarccosx n ),x n ∈[-1,1] (15) Where x n is the value at the nth iteration, n is a positive integer, α is a parameter, and its value is an integer greater than 1.

5. A rescue drone pathfinding system based on an improved whale algorithm, characterized in that: include: The initialization module is used to collect the current two-dimensional coordinates of the drone and set the coordinates of the person to be rescued; The objective function building module is used to build the objective function with the minimum load power, the minimum cost and the shortest rescue distance as the goals; The calculation module is used to substitute the current two-dimensional position of the UAV into the objective function to evaluate the fitness, introduce Chebyshev mapping and adaptive weight to improve the whale algorithm, and use the improved whale algorithm to obtain the optimal next two-dimensional position of the UAV; The path-finding module is used to form the optimal operation path of the drone according to the optimal two-dimensional position and adjust the direction of the drone based on the optimal path.

6. The rescue drone path-finding system based on the improved whale algorithm according to claim 5 is characterized in that: The objective function formula of the objective function building module is as follows: Where Z is the cost, st is the constraint condition, constraint (a) is used to avoid forming a loop containing dangerous areas, and constraint (b) is used to ensure that the initial starting point and final point of the rescue drone are both the rescue center; N is the set of people to be rescued, N = {0, 1, ..., l}, K is the set of rescue drones, K = {1, 2, ..., m}; λ is used to balance the minimum cost and minimum power consumption, and is evenly distributed in [0, 1]; c d is the cost of launching a rescue drone; c t is the fuel consumption cost per unit distance of a single rescue drone; is the distance between the rescue drone k and the rescued person i and the rescued person j; is the driving variable of the rescue drone k from the rescued person i to the rescued person j.

7. The rescue drone path-finding system based on the improved whale algorithm according to claim 5 is characterized in that: The whale algorithm described in the calculation module is implemented as follows: Initialization parameters, whale population, maximum number of iterations t max , the current location of the drone; the population matrix of the whale algorithm is as follows: Where X is the whale population; X g is the g-th whale, g=1,2,…,G; x gh is the position of the g-th whale in the h-th dimension, h = 1, 2, …, d; G is the size of the whale population; d is the dimension of the solution; Assume that the position of the whale closest to the prey is equivalent to the current optimal solution. The other whales update their positions according to the best whale position and gradually move closer to surround the prey. The mathematical model is shown in equations (3) and (4): Where t represents the current iteration number; It represents the optimal position that the whale population has experienced so far; Represents the current whale position; and is the coefficient vector, which is determined by equations (5) and (6) respectively; by adjusting and The value of can control the whale to reach any position near the best whale; among them, as the number of iterations t increases, Decrease linearly from 2 to 0; is a random vector in [0,1], where t max Indicates the maximum number of iterations; Update the position of whale population members in the search space. Let P be a random number between [0,1]. When P≥0.5, whales capture food in a spiral approach. Its mathematical model is shown in formula (8). When P<0.5, whales shrink the encirclement. The formula for whales to surround prey in a spiral form is shown in (9): Where b is a constant coefficient used to limit the logarithmic spiral form; l represents a random number between [-1,1]; In the formula It means that the whale randomly selects the location of an individual from the population as the target location; In the prey search phase, the whale conducts global exploration. When |A| < 1 control parameter, it searches for the local optimal solution, and its mathematical model is formula (11); when |A| ≥ 1 control parameter, the algorithm conducts random search, and its mathematical model is formula (12); Calculate the fitness of each individual and record the optimal individual position; Determine whether the termination condition is met, that is, whether the destination is reached. If the termination condition is not met, repeat this step; if the constraint condition is met, output the optimal operation path.

8. The rescue drone path-finding system based on the improved whale algorithm according to claim 7 is characterized in that: The calculation module introduces Chebyshev mapping and adaptive weights to improve the whale algorithm, including: Introducing an adaptive weight factor, in the early stage of iteration, the whale uses a larger weight to update the position, improving the global search ability of the whale algorithm. In the later stage of iteration, the whale uses a smaller weight to update the position, improving the local optimization ability of the whale algorithm. The improved formula is as follows: The Chebyshev chaotic map is used to generate population positions and enhance the diversity of whale populations. Its mathematical expression is as follows: x n+1 =cos(αarccosx n ),x n ∈[-1,1] (15) Where x n is the value at the nth iteration, n is a positive integer, α is a parameter, and its value is an integer greater than 1.

9. A computing device, characterized in that include: One or more processors, one or more memories, and one or more programs, wherein the programs are stored in the memories and configured to be executed by the processors, and when the programs are loaded into the processors, the steps of the rescue drone pathfinding method based on the improved whale algorithm according to any one of claims 1 to 4 are implemented.

10. A storage medium storing a computer program, characterized in that: The computer program includes program instructions, which, when executed by a processor, cause the processor to perform the steps of the rescue drone path-finding method based on the improved whale algorithm according to any one of claims 1 to 4.

Images