Rescue unmanned aerial vehicle path planning method and system based on improved seathead whale migration algorithm

Abstract

The invention discloses a rescue unmanned aerial vehicle path planning method and system based on an improved seathead whale migration algorithm, and the method comprises the steps: collecting the current three-dimensional coordinates of a rescue unmanned aerial vehicle and the coordinates of a destination, and constructing a target function through an optimal distance; the three-dimensional positions of the rescue unmanned aerial vehicles are substituted into the target function, and based on a cetella migration algorithm, each cetella population generates a group of corresponding rescue unmanned aerial vehicles with the minimum load power and the minimum flight cost and the shortest supply distance; improving a seat head whale migration algorithm; constructing a fitness function, substituting the fitness function into the current three-dimensional coordinates of the rescue unmanned aerial vehicle, and obtaining the optimal three-dimensional position of the rescue unmanned aerial vehicle by using an improved seathead whale migration algorithm; forming an optimal operation path of the rescue unmanned aerial vehicle according to the optimal three-dimensional position, and adjusting the direction of the rescue unmanned aerial vehicle based on the optimal path; the search efficiency can be improved, the operation cost can be reduced, and the method is suitable for complex environments and cost-limited search scenes.

Description

Technical Field

[0001] This invention relates to a path planning method and system for rescue drones, and more particularly to a path planning method and system for rescue drones based on an improved humpback whale migration algorithm, belonging to the fields of energy-saving control and intelligent transportation technology. Background Technology

[0002] Against the backdrop of global climate change and frequent natural disasters, the importance of rescue drones is increasingly prominent. With their advantages of rapid response, flexibility, and efficient operation, they can quickly arrive at and execute missions in complex and dangerous disaster sites, providing crucial support for rescue efforts. Whether rapidly transporting medical supplies, conducting disaster reconnaissance, or assisting in search and rescue operations, rescue drones effectively improve rescue efficiency, save valuable time, and play a vital role in saving lives and minimizing losses, becoming an indispensable component of modern emergency rescue systems.

[0003] Intelligent algorithms play a crucial role in modern society. They can process and analyze massive amounts of data, extracting valuable information and providing strong support for decision-making. Heuristic algorithms, by simulating the intelligent behavior of biological groups, can efficiently handle large-scale and complex optimization problems, especially when facing high-dimensional, nonlinear, and dynamically changing systems, exhibiting powerful global search capabilities and adaptability. Their parallel computing characteristics can fully utilize modern computing resources, significantly improving solution efficiency; at the same time, the flexibility of these algorithms allows them to easily adapt to different problem constraints and objective functions, making them widely used in machine learning, path planning, resource allocation, and other fields. They provide efficient and reliable optimization solutions for solving practical problems, becoming an important tool in the field of modern intelligent computing.

[0004] The current limitations of rescue drone technology include the inability to quickly adjust its path when facing complex terrain or dynamic obstacles due to algorithmic limitations. For example, in narrow areas or environments with dense obstacles, the drone's planning results may be inaccurate or fail due to insufficient algorithmic computation efficiency or its inability to adapt to environmental changes in a timely manner. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to provide a path planning method and system for rescue drones based on an improved humpback whale migration algorithm that can improve pathfinding efficiency.

[0006] Technical solution: The present invention provides a path planning method for rescue drones based on an improved humpback whale migration algorithm, comprising:

[0007] (1) Collect the current three-dimensional coordinates of the rescue drone and the coordinates of the destination, and construct the objective function with the optimal distance;

[0008] (2) Substitute the three-dimensional position of the rescue drone into the objective function, and based on the humpback whale migration algorithm, generate a set of rescue drones with the minimum load power and the minimum flight cost and the shortest resupply distance for each humpback whale population.

[0009] (3) Improve the humpback whale migration algorithm, the improvements include: introducing Halton sequences when generating the initial population, replacing the leader whale guidance position update strategy in the original humpback whale migration algorithm with the zebra foraging strategy, and adding the Gauss-Cauchy perturbation mutation strategy in the later stage of the humpback whale migration algorithm iteration.

[0010] (4) Construct a fitness function and substitute it with the current three-dimensional coordinates of the rescue drone. Use the improved humpback whale migration algorithm to obtain the optimal three-dimensional position of the rescue drone.

[0011] (5) The optimal operating path of the rescue drone is formed based on the optimal three-dimensional position, and the direction of the rescue drone is adjusted based on the optimal path.

[0012] Further, the construction of the objective function with the optimal distance in step (1) includes:

[0013] The formula for the distance between adjacent nodes is introduced to enable the rescue drone to select the nearest node D(c,v,k) within the area. The calculation formula is as follows:

[0014]

[0015] Among them, (x c ,y v ,z k (x) represents the three-dimensional coordinates of the currently available node; c+1 ,y v+1 ,z k+1 () represents the three-dimensional coordinates of the next available node being searched by the current node;

[0016] A target point distance factor is introduced, which enables the UAV to select the node Q(c,v,k) that is closer to the target point when searching for a path. The calculation formula is as follows:

[0017]

[0018] Among them, (x last ,y last ,z last (x) represents the three-dimensional coordinates of the target point. a+1 ,y a+1 ,z a+1 () represents the three-dimensional coordinates of the next available node; the objective function is constructed using the optimal distance as follows:

[0019] minZ=ln(D(c,v,k)+Q(c,v,k)).

[0020] Furthermore, step (3) of introducing the Halton sequence when generating the initial population includes:

[0021] Assuming the search space is three-dimensional, three prime numbers are selected as the base, corresponding to the three dimensions respectively. Each dimension is repeatedly divided and transitioned to values ​​based on the base, thereby forming some non-repeating and uniform points. The mathematical model of the division process is shown in the following equation:

[0022]

[0023] In the formula, n represents the ordinal number of the Halton sequence; j is the power of the cardinality, and m i The exponent represents the highest power of the cardinality, i represents the current dimension, and p represents the exponent. i The cardinality of the Halton sequence is a prime number greater than or equal to 2; a j ∈{0,1,2,…,p1-1} are constant variables; H(n) is a defined sequence function; H(n) is a three-dimensional uniform Halton sequence.

[0024] Further, step (4) includes:

[0025] (41) Initialize the algorithm parameters, set the maximum boundary, minimum boundary, population size N, dimension Dim and maximum number of iterations max_iter for the humpback whale optimization algorithm search;

[0026] (42) Substitute the current 3D coordinates of the rescue drone into the fitness function to obtain the fitness value. The position of each humpback whale represents the 3D position of a rescue drone. Construct an optimized fitness function based on the least squares method to calculate the fitness value. The function formula is:

[0027]

[0028] In the formula, J is the fitness function, t_des is the expected value, and t_act is the current value;

[0029] (43) The humpback whale population begins to simulate migration and sorts the individuals in the population according to their fitness values. The individual with the highest fitness is selected as the leader humpback whale and the current position of the entire migrating humpback whale population is determined.

[0030] (44) Use the simulated zebra search process to update the location of the search agent population;

[0031] (45) Simulate the behavior of the lead whale in discovering and searching for the destination during migration and add a Gauss-Cauchy perturbation mutation strategy to obtain the optimal search agent position in the current population;

[0032] (46) Determine if the maximum number of iterations has been reached. If so, stop the optimization and output the optimal path. Otherwise, return to (43) and continue the optimization.

[0033] Furthermore, step (43) involves the humpback whale population beginning to simulate migration, including:

[0034] W f =L+rand(1,D)⊙(Ub-Lb),f=1,2,…,N pop

[0035] Among them, W f represents the position of the f-th migrating humpback whale in the population, Lb represents the lower bound of the search space, Ub represents the upper bound of the search space, rand(1,D) represents the vector that generates random numbers in the interval [0,1] of dimension D, and ⊙ represents the Hadamard product of the two vectors.

[0036] The process of selecting the most fit individual as the leader humpback whale and determining the current location of the entire migrating humpback whale pod includes:

[0037]

[0038] In the formula, W g The gth migrating humpback whale, N L The number of humpback whales representing experienced leaders, W Mean For the current N L The average position of the leader humpback whale.

[0039] Further, step (44) includes:

[0040] After a leader humpback whale is selected in the population, it guides the other humpback whales to its position within the population; the position update formula is as follows:

[0041]

[0042] Among them, W f This refers to the location of the humpback whale in the fth migration. To determine the location of the humpback whale in the fth migration after the update, PZ f The humpback whale is the leader of each population. For the leader humpback whale after its position is updated, F is the objective function value. h It corresponds to W f The objective function value, yes The objective function value is r, which is a random number between [0, 1], and I∈{1,2}. The higher the value of I, the greater the population change.

[0043] Furthermore, step (45) includes simulating the behavior of the lead whale discovering and searching for its destination during migration:

[0044]

[0045] Among them, W h This indicates the location of the humpback whale in the h-th migration. For the leader humpback whale after its position is updated, r1 and r2 are vectors of random numbers from the interval [0, 1] with dimension D, L represents the starting position vector of the leader humpback whale, U represents the destination position vector, and UL is the relative direction vector;

[0046] Step (45) further includes incorporating a Gaussian-Cauchy perturbation mutation strategy, and performing the mutation formula for the optimal individual as follows:

[0047] H b (t)=W b (t)*(1+μ1*Gauss(σ)+μ2*cauchy(σ))

[0048] μ1 = t / max_iter

[0049] μ2 = 1 - t / max_iter

[0050] Where t represents the current iteration number, max_iter represents the maximum iteration number, and W b (t) represents the optimal position of individual W in the t-th iteration, H b (t) represents the optimal position in the t-th iteration, W b (t) represents the position after the Gaussian-Cauchy mixed perturbation, where Gauss(σ) is the Gaussian mutation operator and cauchy(σ) is the Cauchy mutation operator.

[0051] Based on the same inventive concept, this invention also provides a path planning system for rescue drones based on an improved humpback whale migration algorithm, comprising:

[0052] The data acquisition module is used to collect the current three-dimensional coordinates of the rescue drone and the coordinates of the destination, and to construct the objective function with the optimal distance.

[0053] The preprocessing module is used to substitute the three-dimensional position of the rescue drone into the objective function. Based on the humpback whale migration algorithm, it generates a set of rescue drones with the minimum load power, the minimum flight cost, and the shortest resupply distance for each humpback whale population.

[0054] An improvement module is used to improve the humpback whale migration algorithm. The improvements include: introducing Halton sequences when generating the initial population, replacing the leader whale guidance position update strategy in the original humpback whale migration algorithm with a zebra foraging strategy, and adding a Gauss-Cauchy perturbation mutation strategy in the later stage of the humpback whale migration algorithm iteration.

[0055] The iterative module is used to construct the fitness function and, by substituting the current three-dimensional coordinates of the rescue drone, obtain the optimal three-dimensional position of the rescue drone using the improved humpback whale migration algorithm.

[0056] The pathfinding module is used to compose the optimal operating path for the rescue drone based on the optimal three-dimensional position, and to adjust the direction of the rescue drone based on the optimal path.

[0057] Based on the same inventive concept, the present invention also provides a computing device, comprising: one or more processors, one or more memories, and one or more programs, the programs being stored in the memory and configured to be executed by the processor, the programs being loaded onto the processor to implement the steps of the rescue drone path planning method based on the improved humpback whale migration algorithm as described in any of the preceding claims.

[0058] Based on the same inventive concept, the present invention also provides a storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, cause the processor to perform the steps of the rescue drone path planning method based on the improved humpback whale migration algorithm as described in any of the preceding claims.

[0059] Beneficial Effects: Compared with existing technologies, this invention has the following significant advantages: 1. This invention uses the shortest transportation distance as the objective function. Based on optimizing the objective function, it uses the humpback whale migration algorithm to optimize the pathfinding path, improving pathfinding efficiency and saving operating costs; 2. This invention first introduces Halton sequences, which, by increasing the diversity of the initial population, helps avoid getting trapped in local optima. Secondly, it replaces the lead whale-guided position update strategy in the original humpback whale migration algorithm with a zebra foraging strategy, avoiding the drawbacks of high randomness and excessively long steps. Finally, a Gaussian-Cauchy perturbation mutation strategy is added in the later stages of the humpback whale migration algorithm iteration, giving the algorithm better local exploration capabilities and accelerating the convergence speed. Attached Figure Description

[0060] Figure 1 This is a flowchart of a method according to an embodiment of the present invention;

[0061] Figure 2 A comparison diagram of Halton sequences is introduced for the initialization of embodiments of the present invention;

[0062] Figure 3This is a comparison of the humpback whale migration algorithm WMA (before improvement) and the humpback whale migration algorithm IWMA (after improvement) in the CE2005 test set F1 of this invention.

[0063] Figure 4 This is a comparison of the humpback whale migration algorithm WMA (before improvement) and the humpback whale migration algorithm IWMA (after improvement) in the CE2005 test set F2 of this invention.

[0064] Figure 5 This is a comparison of the humpback whale migration algorithm WMA (before improvement) and the humpback whale migration algorithm IWMA (after improvement) in the CE2005 test set F3 of this invention.

[0065] Figure 6 This is a comparison of the humpback whale migration algorithm WMA (before improvement) and the humpback whale migration algorithm IWMA (after improvement) in the CE2005 test set F4 of this invention.

[0066] Figure 7 This is a comparison of the humpback whale migration algorithm WMA (before improvement) and the humpback whale migration algorithm IWMA (after improvement) in the CE2005 test set F5 of this invention.

[0067] Figure 8 This is a comparison of the humpback whale migration algorithm WMA (before improvement) and the humpback whale migration algorithm IWMA (after improvement) in the CE2005 test set F6 of this invention.

[0068] Figure 9 This is a comparison of the original humpback whale migration algorithm WMA and the improved humpback whale migration algorithm IWMA in the CE2005 test set F7 of this invention. Detailed Implementation

[0069] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0070] As attached Figure 1 As shown in this embodiment, the rescue drone path planning method based on the improved humpback whale migration algorithm includes:

[0071] (1) Collect the current three-dimensional coordinates of the rescue drone and the coordinates of the destination, and construct the objective function with the optimal distance;

[0072] (2) Substitute the three-dimensional position of the rescue drone into the objective function, and based on the humpback whale migration algorithm, generate a set of rescue drones with the minimum load power and the minimum flight cost and the shortest resupply distance for each humpback whale population.

[0073] (3) Improve the humpback whale migration algorithm to obtain the improved humpback whale migration algorithm IWMA. The improvements include: introducing Halton sequence when generating the initial population, replacing the leader whale guidance position update strategy in the original humpback whale migration algorithm with zebra foraging strategy, and adding Gauss-Cauchy perturbation mutation strategy in the later stage of humpback whale migration algorithm iteration.

[0074] (4) Construct a fitness function and substitute it with the current three-dimensional coordinates of the rescue drone. Use the improved humpback whale migration algorithm to obtain the optimal three-dimensional position of the rescue drone.

[0075] (5) The optimal operating path of the rescue drone is formed based on the optimal three-dimensional position, and the direction of the rescue drone is adjusted based on the optimal path.

[0076] Specifically, in step (1), the objective function is constructed using the optimal distance, including:

[0077] The formula for determining the distance between adjacent nodes is introduced to allow rescue drones to select the nearest node within the area. The calculation formula is as follows:

[0078]

[0079] Among them, (x c ,y v ,z k (x) represents the three-dimensional coordinates of the currently available node; c+1 ,y v+1 ,z k+1 () represents the three-dimensional coordinates of the next available node being searched by the current node;

[0080] A target point distance factor is introduced to enable the UAV to select nodes closer to the target point when searching for a path. The calculation formula is as follows:

[0081]

[0082] Among them, (x last ,y last ,z last (x) represents the three-dimensional coordinates of the target point. a+1 ,y a+1 ,z a+1 () represents the three-dimensional coordinates of the next available node; construct the objective function using the optimal distance:

[0083] minZ=ln(D(c,v,k)+Q(c,v,k)).

[0084] Step (3) involves introducing the Halton sequence when generating the initial population, such as... Figure 2 As shown, it includes:

[0085] Assuming the search space is three-dimensional, three prime numbers are selected as the base, corresponding to the three dimensions respectively. Each dimension is repeatedly divided and transitioned back to values ​​based on the base, thereby forming some non-repeating and uniform points. The mathematical model of the division process is shown in the following equation:

[0086]

[0087] In the formula, n represents the ordinal number of the Halton sequence; j is the power of the cardinality, and m i The exponent represents the highest power of the cardinality, i represents the current dimension, and p represents the exponent. i The cardinality of the Halton sequence is a prime number greater than or equal to 2; a j ∈{0,1,2,…,p1-1} are constant variables; H(n) is a defined sequence function; H(n) is a three-dimensional uniform Halton sequence.

[0088] The process of obtaining the optimal three-dimensional position of the rescue drone using the improved humpback whale migration algorithm in step (4) is as follows:

[0089] (41) Algorithm parameter initialization: Set the maximum boundary ub, minimum boundary 1b, population size N, dimension dim, and maximum number of iterations max_iter for the humpback whale optimization algorithm search.

[0090] (42) Substitute the current 3D coordinates of the rescue drone into the fitness function to obtain the fitness value. The position of each humpback whale represents the 3D position of a rescue drone. Construct an optimized fitness function based on the least squares method to calculate the fitness value. The function formula is:

[0091]

[0092] In the formula, J is the fitness function, t_des is the expected value, and t_act is the current value;

[0093] (43) The humpback whale population begins to simulate migration, and individuals in the population are ranked according to their fitness values. The individual with the highest fitness is selected as the leader humpback whale, and the current location of the entire migrating humpback whale population is determined. The simulated migration of the humpback whale population includes:

[0094] W f =L+rand(1,D)⊙(Ub-Lb),f=1,2,…,N pop

[0095] Among them, W frepresents the position of the f-th migrating humpback whale in the population, Lb represents the lower bound of the search space, Ub represents the upper bound of the search space, rand(1,D) represents the vector that generates random numbers in the interval [0,1] of dimension D, and ⊙ represents the Hadamard product of the two vectors.

[0096] The process of selecting the most fit individual as the leader humpback whale and determining the current location of the entire migrating humpback whale pod includes:

[0097]

[0098] In the formula, W g The gth migrating humpback whale, N L The number of humpback whales representing experienced leaders, W Mean For the current N L The average position of the leader humpback whale.

[0099] (44) Update the location of the search agent population by simulating the zebra search process:

[0100] After a lead zebra is selected in the population, it guides the other zebras to its position within the population. The position update formula is as follows:

[0101]

[0102] Among them, W f This refers to the location of the humpback whale in the fth migration. To determine the location of the humpback whale in the fth migration after the update, PZ f The humpback whale is the leader of each population. For the leader humpback whale after its position is updated, F is the objective function value. h It corresponds to W f The objective function value, yes The objective function value is r, which is a random number between [0, 1], and I∈{1,2}. The higher the value of I, the greater the population change.

[0103] (45) Simulate the behavior of the lead whale in discovering and searching for its destination during migration and incorporate a Gaussian-Cauchy perturbation mutation strategy to obtain the optimal search agent location in the current population:

[0104]

[0105] Among them, W h This indicates the location of the humpback whale in the h-th migration. For the leader humpback whale after its position is updated, r1 and r2 are vectors of random numbers from the interval [0, 1] with dimension D, L represents the starting position vector of the leader humpback whale, U represents the destination position vector, and UL is the relative direction vector;

[0106] Incorporating a Gaussian-Cauchy perturbation mutation strategy, the mutation formula for the optimal individual is as follows:

[0107] H b (t)=W b (t)*(1+μ1*Gauss(σ)+μ2*cauchy(σ))

[0108] μ1 = t / max_iter

[0109] μ2 = 1 - t / max_iter

[0110] Where t represents the current iteration number, max_iter represents the maximum iteration number, and W b (t) represents the optimal position of individual W in the t-th iteration, H b (t) represents the optimal position in the t-th iteration, W b (t) represents the position after the Gaussian-Cauchy mixed perturbation, where Gauss(σ) is the Gaussian mutation operator and cauchy(σ) is the Cauchy mutation operator.

[0111] (46) Determine if the maximum number of iterations has been reached. If so, stop the optimization and output the optimal path. Otherwise, return to (43) and continue the optimization.

[0112] Simulations of the test function iteration graphs were performed on the humpback whale migration algorithm WMA (before improvement) and the improved humpback whale migration algorithm IWMA (after improvement). The simulation comparison results for each test set are as follows: Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 , Figure 8 and Figure 9 As shown, the optimized humpback whale migration algorithm IWMA converges faster.

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

[0114] The data acquisition module is used to collect the current three-dimensional coordinates of the rescue drone and the coordinates of the destination, and to construct the objective function with the optimal distance.

[0115] The preprocessing module is used to substitute the three-dimensional position of the rescue drone into the objective function. Based on the humpback whale migration algorithm, it generates a set of rescue drones with the minimum load power, the minimum flight cost, and the shortest resupply distance for each humpback whale population.

[0116] An improvement module is used to improve the humpback whale migration algorithm. The improvements include: introducing Halton sequences when generating the initial population, replacing the leader whale guidance position update strategy in the original humpback whale migration algorithm with a zebra foraging strategy, and adding a Gauss-Cauchy perturbation mutation strategy in the later stage of the humpback whale migration algorithm iteration.

[0117] The iterative module is used to construct the fitness function and, by substituting the current three-dimensional coordinates of the rescue drone, obtain the optimal three-dimensional position of the rescue drone using the improved humpback whale migration algorithm.

[0118] The pathfinding module is used to compose the optimal operating path for the rescue drone based on the optimal three-dimensional position, and to adjust the direction of the rescue drone based on the optimal path.

[0119] 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, the programs being stored in the memory and configured to be executed by the processor, the programs being loaded onto the processor to implement the steps of the rescue drone path planning method based on the improved humpback whale migration algorithm according to any of the preceding claims.

[0120] Based on the same inventive concept, this embodiment also provides a storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, cause the processor to perform the steps of the rescue drone path planning method based on the improved humpback whale migration algorithm described above.

[0121] To address the issues of local optima traps, slow convergence, and high randomness in the humpback whale migration algorithm, this invention first introduces Halton sequences to increase population diversity. Secondly, it replaces the lead whale-guided position update strategy in the original humpback whale migration algorithm with a zebra foraging strategy to avoid excessive randomness. Finally, a Gaussian-Cauchy perturbation mutation strategy is added in the later stages of the humpback whale migration algorithm iteration to enhance its local exploration capabilities and accelerate convergence.

[0122] This invention innovatively proposes an improved humpback whale migration algorithm, which has the advantages of fast and efficient path selection, improving not only the accuracy and efficiency of pathfinding but also reducing operating costs, thus meeting the search needs of complex environments and cost-constrained environments.

Claims

1. A method for rescue drone path planning based on improved humpback whale migration algorithm, characterized in that, The method comprises the following steps: (1) collecting the current three-dimensional coordinates of the rescue unmanned aerial vehicle and the coordinates of the destination, and constructing a target function with the optimal distance; (2) substituting the three-dimensional position of the rescue unmanned aerial vehicle into the target function, and enabling each humpback whale population to generate a corresponding set of rescue unmanned aerial vehicles with the minimum load power and the minimum flight cost and the shortest supply distance based on the humpback whale migration algorithm; (3) improving the humpback whale migration algorithm, wherein the improvement comprises: introducing a Halton sequence when generating an initial population, replacing the leader whale guiding position updating strategy in the original humpback whale migration algorithm with a zebra foraging strategy, and adding a Gaussian-Cauchy disturbance variation strategy in the later stage of the humpback whale migration algorithm iteration; (4) constructing a fitness function and substituting the current three-dimensional coordinates of the rescue unmanned aerial vehicle into the fitness function, and obtaining the optimal three-dimensional position of the rescue unmanned aerial vehicle by using the improved humpback whale migration algorithm; (5) forming an optimal operation path of the rescue unmanned aerial vehicle according to the optimal three-dimensional position, and adjusting the direction of the rescue unmanned aerial vehicle based on the optimal path.

2. The rescue drone path planning method based on improved sei whale migration algorithm according to claim 1, characterized in that, The step (1) comprises the following steps of constructing the target function with the optimal distance: introducing a neighboring node distance formula to enable the rescue unmanned aerial vehicle to select a node D(c, v, k) with a relatively short distance in the region, and the calculation formula is as follows: Wherein, (x c ,y v ,z k ) represents the three-dimensional coordinates of the current available node; (x c+1 ,y v+1 ,z k+1 ) represents the three-dimensional coordinates of the next available node found by the current node; introducing a target point distance factor to enable the unmanned aerial vehicle to select a node Q(c, v, k) with a relatively short distance to the target point when searching for a path, and the calculation formula is as follows: Wherein, (x last ,y last ,z last ) represents the target point three-dimensional coordinates, (x a+1 ,y a+1 ,z a+1 ) represents the next available node three-dimensional coordinates;With the optimal distance to build the target function as follows: minZ = ln(D(c, v, k) + Q(c, v, k)).

3. The rescue drone path planning method based on improved sei whale migration algorithm according to claim 1, characterized in that, The step (3) comprises the following steps of introducing the Halton sequence when generating the initial population: assuming that the search space is three-dimensional, selecting three prime numbers as the bases corresponding to the three dimensions, and respectively cutting and migrating the values of (0, 1) according to the bases to form some non-repeated and uniform points, and the cutting process mathematical model is as shown in the following formula: where n represents the ordinal number of the Halton sequence; j is the power of the base, m i represents the index of the highest power of the base, i represents the current dimension, p i represents the base of the Halton sequence, and takes a prime number greater than or equal to 2; a j ∈{0,1,2,…,p1-1} is a constant variable; is a defined sequence function; H(n) is a three-dimensional uniform Halton sequence.

4. The rescue drone path planning method based on improved sei whale migration algorithm according to claim 1, characterized in that, The step (4) comprises the following steps: (41) initializing algorithm parameters, setting the maximum boundary, the minimum boundary, the population size N, the dimension Dim and the maximum iteration number max-iter of the humpback whale optimization algorithm search; (42) substituting the current three-dimensional coordinates of the rescue unmanned aerial vehicle into the fitness function to obtain a fitness value, the position of each humpback whale representing a three-dimensional position of a rescue unmanned aerial vehicle, constructing an optimized fitness function based on the least square method for calculating the fitness value, and the function formula is as follows: In the formula, J is the fitness function, t-des is the expected value, and t-act is the current value; (43) starting the simulation migration movement of the humpback whale population, and sorting the individuals in the population according to the fitness value, selecting the individual with the highest fitness value as the leader humpback whale and determining the current position of the entire migration humpback whale population; (44) updating the position of the search agent population by using the simulation zebra foraging process; (45) simulating the behavior of the leader humpback whale finding and searching for the destination movement in the migration process and adding the Gaussian-Cauchy disturbance variation strategy to obtain the optimal search agent position in the current population; (46) judging whether the maximum iteration number is reached, if yes, stopping the optimization and outputting the optimal path, otherwise, returning to (43) to continue the optimization.

5. The rescue drone path planning method based on improved sei whale migration algorithm according to claim 4, characterized in that, The step (43) comprises the following steps of starting the simulation migration movement of the humpback whale population: W f = L + rand(1, D) 0 (Ub - Lb), f = 1, 2, …, N pop where W f represents the position of the fth migrating humpback whale in the population, Lb represents the lower bound of the search space given, Ub represents the upper bound of the search space given, rand(1, D) represents a vector of random numbers in the interval [0, 1] of dimension D, and represents the Hadamard product of two vectors; The individual with the highest fitness is selected as the leader humpback whale and the current position of the entire migration humpback whale group is determined, including: where W g The number of migrating humpback whales, N L The number of experienced leader humpback whales, W Mean The current N L The average position of leader humpback whales.

6. The rescue drone path planning method based on improved sei whale migration algorithm according to claim 4, characterized in that, The step (44) includes: After the leader humpback whale is selected in the population, the leader humpback whale guides other humpback whales to its position in the population; the position update formula is as follows: wherein W f is the position of the humpback whale of the fth migration, is the position of the humpback whale of the fth migration after updating, PZ f is the leader humpback whale of each population, is the leader humpback whale after updating the position, F is the value of the objective function, F h is the value of the objective function corresponding to W f is the value of the objective function of , r is a random number between 0 and 1, I ∈ {1, 2}, the higher the value of I, the greater the change of the population.​ 7. The rescue drone path planning method based on improved sei whale migration algorithm according to claim 4, characterized in that, The step (45) includes simulating the behavior of the leader humpback whale discovering and searching for the destination movement during migration: where W h is the position of the h-th migrating humpback whale, is the leader humpback whale after updating the position, r1 and r2 are vectors of random numbers from the interval [0, 1] with dimension D, L represents the position vector of the leader humpback whale departure, U denotes the position vector of the destination, U-L is the relative direction vector; The step (45) further includes adding a Gaussian-Cosine disturbance mutation strategy to mutate the optimal individual, and the formula is as follows: H b (t) = W b (t) * (1 + μ1*Gauss(σ) + μ2*cauchy(σ)) μ1=t / max_iter μ2=1-t / max_iter where t represents the current iteration number, max-iter represents the maximum iteration number, W b (t) is the optimal position of individual W in the tth iteration, H b (t) is the optimal position in the tth iteration, W b (t) is the position after the Gaussian-Cauchy hybrid disturbance, Gauss(σ) is the Gaussian variation operator, and cauchy(σ) is the Cauchy variation operator.

8. A rescue drone path planning system based on an improved humpback whale migration algorithm, characterized in that, Including: The collection module is configured to collect the current three-dimensional coordinates of the rescue unmanned aerial vehicle and the coordinates of the destination to construct a target function with an optimal distance; The preprocessing module is configured to substitute the three-dimensional position of the rescue unmanned aerial vehicle into the target function, and based on the humpback whale migration algorithm, each humpback whale population generates a set of rescue unmanned aerial vehicles with the minimum load power and the least flight cost and the shortest supply distance corresponding to the target function; The improvement module is configured to improve the humpback whale migration algorithm, and the improvement includes: introducing a Halton sequence when generating an initial population, replacing the leader humpback whale position update strategy in the original humpback whale migration algorithm with a zebra foraging strategy, and adding a Gaussian-Cosine disturbance mutation strategy in the later stage of the humpback whale migration algorithm iteration; The iteration module is configured to construct a fitness function and substitute the current three-dimensional coordinates of the rescue unmanned aerial vehicle into the fitness function, and obtain the optimal three-dimensional position of the rescue unmanned aerial vehicle by using the improved humpback whale migration algorithm; The pathfinding module is configured to form an optimal operation path of the rescue unmanned aerial vehicle according to the optimal three-dimensional position, and adjust the direction of the rescue unmanned aerial vehicle based on the optimal path.

9. A computing device, comprising: One or more processors, one or more memories, and one or more programs, the programs being stored in the memories and configured to be executed by the processors, when the programs are loaded into the processors, the programs implement the steps of the rescue unmanned aerial vehicle path planning method based on the improved humpback whale migration algorithm according to any one of claims 1 to 7. The storage medium stores a computer program, and the computer program includes program instructions, which, when executed by a processor, cause the processor to perform the steps of the rescue unmanned aerial vehicle path planning method based on the improved humpback whale migration algorithm according to any one of claims 1 to 7.

10. A storage medium, characterized by ​

Images