Multi-unmanned aerial vehicle path collaborative planning method based on improved sparrow search algorithm

Abstract

The invention relates to a multi-unmanned aerial vehicle path collaborative planning method based on an improved sparrow search algorithm. The method comprises the following steps: establishing a space-time dynamic model and a three-dimensional environment model of multiple unmanned aerial vehicles; respectively constructing cost functions of energy consumption, path length, time error and obstacle avoidance capability of the unmanned aerial vehicle, and endowing the constructed cost functions with different weights to construct a target function; and taking the constructed target function as a fitness function of an improved sparrow search algorithm, adding a time difference penalty term into the target function, and realizing three-dimensional path planning of multiple unmanned aerial vehicles from different starting point positions to the same target point at the same time. Compared with the prior art, the method has the advantages that the multiple unmanned aerial vehicles can reach the same target point from different starting point positions at the same time, and the capacity of port safety defense for resisting invasion targets and executing complex tasks is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle path planning, and more particularly to a multi-unmanned aerial vehicle path collaborative planning method based on an improved sparrow search algorithm. Background Art

[0002] As the gathering point and hub of water and land transportation, ports not only play an important role in logistics transportation, but also are the key nodes connecting global trade. With the continuous development of global trade, as an important hub of international trade, the security of ports has received more and more extensive attention. And with the development of technology, port operations tend to be automated and intelligent. As an important part of the unmanned system, the multi-unmanned aerial vehicle system has been widely used in tasks such as port inspection, monitoring and emergency response due to its advantages such as high efficiency, real-time performance and the ability to cope with various complex and changeable environments and weather conditions.

[0003] Since various security risks are usually encountered in the port environment, security measures are required to be able to respond quickly and effectively to sudden events. The cooperation of the multi-unmanned aerial vehicle system can perform real-time monitoring and data collection in the port environment, thereby enhancing the security defense ability. When an intrusion target is detected, since a single unmanned aerial vehicle may encounter problems such as equipment damage and insufficient energy during the execution of tasks, and the cooperation of the multi-unmanned aerial vehicle system will improve the task success rate and efficiency, multiple unmanned aerial vehicles located at different positions in the port environment will be called to coordinate the multi-unmanned aerial vehicles to reach the target point simultaneously to resist the intrusion target. However, due to the complex and changeable port environment and ensuring that multiple unmanned aerial vehicles reach the target point simultaneously, higher requirements will be put forward for the path planning of unmanned aerial vehicles. Therefore, designing a three-dimensional path planning algorithm for multiple unmanned aerial vehicles with different starting positions to reach simultaneously under multiple constraints has important practical significance for improving the security and the ability to perform complex tasks in the port security scenario.

[0004] After retrieval, Chinese Invention Patent Application Publication No. CN112880688A discloses a method for three-dimensional flight path planning of unmanned aerial vehicles based on a chaotic adaptive sparrow search algorithm, including the following steps: establishing a flight environment model according to the flight environment; establishing a flight cost function of the unmanned aerial vehicle to evaluate the flight path performance of the unmanned aerial vehicle; improving the sparrow search algorithm by using a chaotic initialization population strategy, an adaptive weight strategy, and a Cauchy-Gaussian hybrid mutation strategy, and proposing a chaotic adaptive sparrow search algorithm; using the chaotic adaptive sparrow search algorithm to plan the flight path of the unmanned aerial vehicle in a three-dimensional environment to obtain the optimal solution of the unmanned aerial vehicle flight path planning and obtain the planning result; it is proposed that the improved sparrow algorithm has obvious advantages in solving quality, its chaotic strategy and adaptive strategy enable the algorithm to have a fast convergence speed and excellent convergence accuracy, and the mutation strategy enables the algorithm to have a strong ability to jump out of the local optimum, and thus an excellent flight route of the unmanned aerial vehicle can be obtained quickly. This existing patent application has the problem that it cannot realize the path planning for multiple unmanned aerial vehicles to reach the same target position simultaneously.

[0005] How to realize the path planning of multiple unmanned aerial vehicles at different starting positions so that they can reach the same target position simultaneously has become a technical problem to be solved. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for collaborative path planning of multiple unmanned aerial vehicles based on an improved sparrow search algorithm to overcome the defects of the above-mentioned existing technologies.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] According to one aspect of the present invention, there is provided a method for collaborative path planning of multiple unmanned aerial vehicles based on an improved sparrow search algorithm, the method comprising:

[0009] Establishing a spatio-temporal dynamics model and a three-dimensional environment model of multiple unmanned aerial vehicles;

[0010] Respectively constructing cost functions for the energy consumption, path length, time error, and obstacle avoidance ability of the unmanned aerial vehicle, and assigning different weight values to the constructed cost functions to construct an objective function;

[0011] Taking the constructed objective function as the fitness function of the improved sparrow search algorithm, and adding a time difference penalty term to the objective function to realize the three-dimensional path planning for multiple unmanned aerial vehicles to reach the same target point from different starting positions simultaneously.

[0012] Preferably, the objective function is specifically:

[0013] F = w1.F1 + w2.T diff + w3.P obstacle + w4.E,

[0014] Wherein, w1, w2, w3 and w4 are all weight coefficients, F is the objective function, F1 is the cost function of the path length, T diff is the cost function of the time error, P obstacle is the cost function of the obstacle avoidance ability, E is the cost function of the energy consumption, and the weight coefficient w2 of the time error cost function is greater than other weight coefficients.

[0015] More preferably, the cost function of the path length is the accumulation of the Euclidean distances between all adjacent path points on the three-dimensional space path;

[0016] The cost function of the time error is specifically:

[0017]

[0018] Wherein, T i is the arrival time of the i-th drone, T max is the maximum value of the arrival times of all drones, o is the first drone, and O is the number of drones;

[0019] The cost function of the energy consumption is specifically:

[0020] E = k1.F1,

[0021] Wherein, E represents the energy consumption of the drone during flight; k1 is the energy consumption coefficient, which is a fixed constant.

[0022] More preferably, the cost function of the obstacle avoidance ability is specifically:

[0023]

[0024] Wherein, d is the distance between the drone and the obstacle, D safe is the safe distance between the drone and the obstacle, D alert is the critical distance between the drone and the obstacle, P obstacle is the cost function of the drone's obstacle avoidance ability, o is the first drone, O is the number of drones, N1 is the number of path points, n1 is the starting point of the path, and k is the penalty intensity.

[0025] Preferably, the improved sparrow search algorithm includes:

[0026] Using the Latin hypercube sampling technique for population initialization;

[0027] Introducing a phased control compensation strategy and a non-linear attenuation factor in the position update of the discoverer, and dynamically adjusting the dynamic step size at different stages of the algorithm;

[0028] Introduce a chaotic cosine variation factor in the position update of the followers to enhance the search range of the followers for unknown areas.

[0029] More preferably, the position update of the improved discoverer is specifically:

[0030]

[0031] In the above formula, R2 and ST are the safety value and the warning value respectively; Q is a normally distributed random number; is the position of the updated discoverer; X d is the current position of the discoverer; X b is the optimal position; μ is the non-linear factor; β(t) is the dynamic step size; ε(t) is the random perturbation;

[0032] When the safety value is less than the warning value, the sparrows perform a global search; when the safety value is greater than or equal to the warning value, the sparrows will perform a random walk in a normal distribution.

[0033] More preferably, different values are set for the random perturbation and the dynamic step size according to the search stage, where the search stage divides the entire search stage into the early stage, the middle stage, and the late stage according to the ratio of the current iteration number to the maximum iteration number, specifically:

[0034]

[0035] Among them, ε(t) is the random perturbation; μ is the non-linear attenuation factor, T is the maximum iteration number; t is the current iteration number, and U is the uniform distribution function;

[0036]

[0037] Among them, nitial_step_size is the step size in the early stage; mid_step_size is the step size in the middle stage; final_step_size is the step size in the late stage; T is the maximum iteration number; t is the current iteration number; γ is a hyperparameter, that is, the rate of controlling the step size attenuation in the initial stage of the sparrow search algorithm.

[0038] More preferably, the position update of the followers includes:

[0039] Define a chaotic cosine variation factor, which is expressed as:

[0040]

[0041] Among them, δ is the strength of the chaotic factor, t is the current iteration number, and T is the maximum iteration number;

[0042] The chaotic cosine variation factor is added to the position update of the followers, and the improved formula for the position update of the followers is as follows:

[0043]

[0044] In the above formula, Q is a random number of normal distribution; x ′ w,j is the global optimal position; x i ′ ,j is the current position of the follower; is the position of the discoverer; L is the learning factor; A + is a random number of normal distribution; k is a random integer used to generate periodic changes; n is the number of sparrow populations; i is the index of the sparrow individual in the whole population; t is the current iteration number; j represents the dimension index of the sparrow.

[0045] Preferably, the time difference penalty term is specifically:

[0046]

[0047] In the formula, T i is the arrival time of each i-th unmanned aerial vehicle, T max is the maximum value of the arrival times of all unmanned aerial vehicles, N is the number of unmanned aerial vehicles, T diff1 is the time difference penalty term.

[0048] Preferably, the establishment process of the three-dimensional environment model includes: simulating the port environment model in the way of grid map, setting the obstacles that the unmanned aerial vehicle cannot pass through as regular three-dimensional cube models with different shapes, setting no-fly zones according to the size and position of the obstacles, and defining the projection coordinates of the xOy plane of the O i th obstacle as the radius of the obstacle is the height is If the number of no-fly zones is N, the set of created no-fly zones is as follows:

[0049]

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] (1) After constructing the spatio-temporal dynamics model and the three-dimensional environment model of the unmanned aerial vehicle (UAV), the objective function is constructed based on the cost functions of path length, obstacle avoidance ability, energy consumption, and time error respectively. The path planning problem is transformed into an optimization problem with multiple constraints. The objective function is used as the fitness function of the improved sparrow search algorithm, and a time difference penalty term is added to the objective function to achieve the three-dimensional path planning of multiple UAVs arriving at the same target point simultaneously from different starting positions, thereby enhancing the port's security defense against intrusion targets and its ability to execute complex tasks.

[0052] (2) The cost function of the present invention is constructed based on the cost functions of path length, obstacle avoidance ability, energy consumption, and time error, and then the objective function is constructed by weighting multiple cost functions. Considering the simultaneous arrival of multiple UAVs, the weight coefficient of the time error cost function is designed to be greater than other weight coefficients to minimize the time error of multiple UAVs arriving at the target point.

[0053] (3) The present invention improves the sparrow search algorithm. The Latin hypercube sampling method is used to initialize the population. Compared with random initialization, the individuals of the population can be more evenly distributed, which can significantly improve the diversity of the population, enhance the global search ability, and improve the stability and reliability of the randomly initialized population in the original sparrow search algorithm when solving the three-dimensional path planning of multiple UAVs; a phased compensation control strategy and a non-linear factor are added to the position update of the discoverer; in order to reduce the probability of falling into local optimum in the position update of the follower, a chaotic cosine factor is introduced to dynamically adjust the dynamic step size and random perturbation, meeting the search ability of the algorithm at different stages and improving the ability to avoid falling into local optimum, enhancing the global search ability during the UAV path planning, so as to quickly plan a high-quality flight path for the UAV in a complex three-dimensional environment.

[0054] (4) By adding a time difference penalty term to the objective function, the present invention ensures that the UAVs arrive at the target point at as similar times as possible, which is used to improve the collaborative security defense ability of multiple UAVs. Description of the Drawings

[0055] Figure 1 It is a framework and process schematic diagram of the path planning method in the present invention;

[0056] Figure 2(a) is a histogram of the uniformity of the distribution of the population initialized by Latin hypercube sampling in the X-axis in the present invention;

[0057] Figure 2(b) is a histogram of the uniformity of the distribution of the population initialized by Latin hypercube sampling in the Y-axis in the present invention;

[0058] Figure 2(c) is a histogram of the uniformity of the distribution of the population initialized by Latin hypercube sampling in the Z-axis in the present invention;

[0059] Figure 3 This is the fitness comparison chart between the improved sparrow search algorithm and the unimproved sparrow search algorithm in the present invention;

[0060] Figure 4 This is a schematic diagram showing the relationship between the distance of each UAV to the target point and the number of iterations;

[0061] Figure 5 This is a histogram of the time step for each UAV to reach the target point. Detailed implementation manners

[0062] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0063] The UAVs (Unmanned Aerial Vehicles) mentioned in this application are multiple UAVs.

[0064] This embodiment relates to a multi-UAV path collaborative planning method based on an improved sparrow search algorithm, such as Figure 1 , including:

[0065] S1. Establish a spatio-temporal dynamics model and a three-dimensional environment model of multiple UAVs (UAVs);

[0066] S2. Improve the cost function, construct the cost function from four aspects: the energy consumption of the UAV, the path length, the time error, and collision avoidance, assign different weight values to the constructed cost function, and formulate the objective function;

[0067] S3. Improve the sparrow search algorithm;

[0068] S4: Use the objective function formulated in S2 as the fitness function of the improved sparrow search algorithm, add a time difference penalty term to the objective function, and use the improved sparrow search algorithm to realize the three-dimensional path planning for multiple UAVs to reach the same target point simultaneously.

[0069] The specific steps for establishing the spatio-temporal dynamics model and the three-dimensional environment model of UAVs in S1 are as follows:

[0070] S11. Establish the spatio-temporal dynamics model of the UAV:

[0071] In this application, the UAV is equivalent to a mass point model with mass, and the spatio-temporal dynamics model of the UAV is: P new = P0 + V i .T, where P newis the current position of the UAV, P0 is the position of the UAV at the initial moment, V i is the speed of the i-th UAV, and T is the movement time of the UAV.

[0072] S12. Establish a three-dimensional environment model: Use the grid map method to simulate the port environment model. Since the types of obstacles in the port environment are complex, set the obstacles that the UAV cannot pass through as regular three-dimensional cube models with different shapes. To avoid collisions, set no-fly zones according to the size and position of the obstacles, and define the projection coordinates of the xOy plane of the O i th obstacle as The radius of the obstacle is For non-cylindrical obstacles, its radius is equivalent to the distance from the projection coordinate point to the farthest side length, and the height is The number of no-fly zones is N, and the set of no-fly zones created based on this is as follows:

[0073]

[0074] In this embodiment, the types of obstacles are constructed as two types: cylinders and cubes, and the positions and sizes of the two types of obstacles are randomly generated. Further, the types of obstacles in this application can be other shapes except cylinders and cubes.

[0075] The starting positions of multiple UAVs are randomly generated using the Latin sampling method, and the starting positions of each UAV are different. The three-dimensional coordinates of the end position are (50, 50, 50).

[0076] In S2, transform the cost function, formulate the objective function, and transform the path planning problem of multiple UAVs arriving simultaneously into a multi-constraint optimization problem. The specific operation method is as follows:

[0077] S21. According to actual needs, during the flight of the UAV, to save energy consumption, it is necessary to select the shortest path for flight. Let the flight path point of the UAV be P (i,j) =(x (i,j) , y (i,j) , z (i,j) ), which represents the position of the j-th path point in the i-th flight path in the three-dimensional environment. The expression of the path length is:

[0078]

[0079] In the above formula, F1 represents the cost function of the path length, that is, the accumulation of the Euclidean distances between adjacent path points, and N1 is the number of path points.

[0080] S22. Since the shorter the planned path of the UAV during mission execution, the relatively smaller its energy consumption, the energy consumption is directly proportional to the path length. The expression for constructing the cost function of energy consumption is as follows:

[0081] E = k1.F1,

[0082] where E represents the energy consumption of the UAV during flight; k1 is the energy consumption coefficient, which is a fixed constant.

[0083] S23. The obstacle avoidance ability of the UAV refers to the ability of the UAV to avoid the no - fly zone set in S1 during flight. To prevent the UAV from colliding, a safety distance D safe and a critical distance D alert are set. Then the cost function of the UAV's obstacle avoidance ability can be expressed as:

[0084]

[0085] where d is the distance between the UAV and the obstacle, D safe is the safety distance between the UAV and the obstacle, D alert is the critical distance between the UAV and the obstacle, P obstacle is the cost function of the UAV's obstacle avoidance ability, o is the first UAV, O is the number of UAVs, N1 is the number of path points, n1 is the starting point of the path, and k is a constant representing the penalty intensity.

[0086] S24. For the multi - UAV system to be able to execute a certain determined task collaboratively, at this time, UAVs located at different positions need to reach the target point simultaneously. Therefore, considering the time error constraint, the cost function is constructed as:

[0087]

[0088] where T i is the arrival time of the i - th UAV, T max is the maximum value of the arrival times of all UAVs, o is the first UAV, and O is the number of UAVs.

[0089] S25. When constructing the cost function, consider the four aspects of the UAV's path length, energy consumption, obstacle avoidance ability, and time difference penalty term. Assign different weight values to the above four cost functions. This application focuses on achieving the problem of simultaneous arrival of multiple UAVs. Therefore, in the design of the objective function, the weight coefficient of the time error cost function is greater than other weight coefficients, and an objective function with multi - objective constraints is established. The mathematical expression is as follows:

[0090] F = w1.F1+w2.T diff +w3.P obstacle +w4.E,

[0091] In the formula, w1, w2, w3, and w4 are the weight coefficients of their respective corresponding cost functions, and F is the expression of the objective function.

[0092] The improved sparrow search algorithm in S3 includes:

[0093] S31. Use the Latin hypercube sampling technique for population initialization;

[0094] S32. Introduce a phased control compensation strategy and a non-linear attenuation factor into the position update formula of the discoverer, adjust the step size at different stages of the algorithm, and balance the global and local search capabilities;

[0095] S33. Introduce a chaotic cosine variation factor into the position update of the followers, strengthen the search range of the followers for unknown regions, and reduce the probability of falling into local optima.

[0096] To improve the problem of uneven initial population distribution of the sparrow search algorithm, the Latin hypercube sampling technique is used for the initialization of the sparrow population. Compared with random initialization, Latin cube sampling can make the population individuals more evenly distributed. Therefore, the initialization of the sparrow search algorithm is implemented by using the Latin cube sampling method. The histogram of the uniformity of the population initialized by using Latin cube sampling in the three axes is as Figures 2(a) to 2(c) shown. The specific operation steps of S31 are as follows:

[0097] S311. First, determine the movement range of the UAV in three-dimensional space as follows: the movement range of the X-axis is: [X min , X max , the movement range of the Y-axis is: [Y min , Y max , and the movement range of the Z-axis is: [Z min , Z max ;

[0098] S312. Determine the sample number according to the number of UAVs performing the task: Let the number of the initial population of the initial sparrow algorithm be M. Divide each dimension (X-axis, Y-axis, and Z-axis dimensions) in the three-dimensional space into M three-dimensional coordinate points. Randomly select a value in each interval, and ensure that each interval is only selected once. Then randomly combine the sample points of the three dimensions together to form M three-dimensional coordinate points. The initialized population obtained in this way is used as the input of the sparrow algorithm.

[0099] In S32, to balance the global and local search capabilities of the sparrow search algorithm, a phased control compensation strategy and a non-linear attenuation factor are introduced into the position update formula of the discoverer, and the dynamic step size is adjusted at different stages of the algorithm. The specific implementation steps are as follows:

[0100] The stage control compensation strategy and the non-linear factor are introduced into the position update formula of the discoverer. The improved position update formula of the discoverer is as follows:

[0101]

[0102] In the above formula, R2 and ST are the safety value and the warning value respectively; Q is a normally distributed random number; is the position of the discoverer after update; X d is the current position of the discoverer; X b is the optimal position; μ is the non-linear factor. When R2 < ST, the sparrows conduct a global search; when R2 ≥ ST, the sparrows will perform a random walk in a normal distribution; β(t) is the dynamic step size; ε(t) is the random perturbation, and different values are set for the random perturbation according to the early, middle, and late stages.

[0103] The stages involved in the dynamic step size β(t) include: encouraging search in the early stage, balancing global and local search in the middle stage, and focusing on local search in the late stage. The specific design steps are as follows:

[0104]

[0105] In the above formula, nitial_step_size is the step size in the early stage, which is generally set to 1; mid_step_size is the step size in the middle stage, which is generally set to half of the initial compensation; final_step_size is the step size in the late stage, which is generally set to a small value, such as 0.05; T is the maximum number of iterations; t is the current number of iterations; γ is a hyperparameter, which is the rate of controlling the step size decay in the initial stage of the sparrow search algorithm.

[0106] The specific representation form of the random perturbation ε(t) is:

[0107]

[0108] μ is the non-linear attenuation factor, and its calculation method is:

[0109] According to the set maximum number of iterations, different stages are divided: both the random perturbation and the dynamic step size divide the entire search stage of the discoverer of the sparrow search algorithm into the early, middle, and late stages according to the ratio of the current number of iterations to the maximum number of iterations. And the corresponding search step sizes and random perturbation amounts are set according to the search characteristics in different periods. When there are no predators around the foraging environment, the discoverer updates its position according to the above formula. However, when a sparrow in the population discovers a predator, at this time, the other sparrows in the population issue an alarm, and at this time, the sparrows will need to quickly fly to other safe places to forage.

[0110] In S33, to enhance the search range of followers for unknown areas and reduce the probability of falling into local optima, a chaotic cosine variation factor is introduced in the position update of followers. The specific operation steps are as follows:

[0111] S331. Define the chaotic cosine variation factor, which is expressed as:

[0112]

[0113] where, δ is the strength of the chaotic factor, t is the current iteration number, and T is the maximum iteration number.

[0114] S332. Add the chaotic cosine variation factor to the position update formula of followers. The improved position update formula of followers is:

[0115]

[0116] In the above formula, Q is a normally distributed random number; x ′ w,j is the global optimal position; x i ′ ,j is the current position of the follower; is the position of the discoverer; L is the learning factor; A + is a normally distributed random number; k is a random integer used to generate periodic changes; n is the number of sparrow populations; i is the index of the sparrow individual in the entire population; t is the current iteration number; j represents the dimension index of the sparrow.

[0117] When , the sparrows with lower fitness do not obtain food, so they will need more food.

[0118] Through the above steps, in the position update of followers, a chaotic cosine variation factor is added. Through adjustments in different stages, the search ability of followers for unknown areas will be enhanced, and the probability of falling into local optima will be reduced. The fitness comparison between the improved sparrow search algorithm and the unimproved sparrow search algorithm is as Figure 3 shown.

[0119] In S4, a time difference penalty term is added to the objective function, and the improved sparrow search algorithm is used to finally achieve the three-dimensional path planning for multiple UAVs to reach the same target point simultaneously. Specifically: a time difference penalty term is added to the fitness function, and its calculation method is: add the squares of the differences between the times when all UAVs reach the target and the maximum time to ensure that the UAVs reach the target point at as similar times as possible. The time penalty term T diff1 can be expressed as:

[0120]

[0121] Wherein, T i is the arrival time of the i-th drone, and T max is the maximum value of the arrival times of all drones, and N is the number of drones.

[0122] Figure 4 is the comparison of the distances of each drone to the target point as the number of iterations increases, Figure 5 is the time step histogram of the arrival times of each drone at the target point. Combining Figure 4 and Figure 5 , it can be seen that after multiple iterations, each drone reaches the target point simultaneously.

[0123] By adding a time difference penalty term to the objective function, it is realized to encourage the drones to reach the target point as simultaneously as possible, thereby achieving the goal of multi-drone coordinated path planning and finally completing the three-dimensional path planning for multi-drones to reach simultaneously.

[0124] The electronic device of the present invention includes a central processing unit (CPU), which can execute various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) or computer program instructions loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for device operation can also be stored. The CPU, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0125] Multiple components in the device are connected to the I / O interface, including: an input unit, such as a keyboard, a mouse, etc.; an output unit, such as various types of displays, speakers, etc.; a storage unit, such as a disk, an optical disc, etc.; and a communication unit, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit allows the device to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0126] The processing unit executes the various methods and processes described above. For example, in some embodiments, the method can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device via the ROM and / or the communication unit. When the computer program is loaded into the RAM and executed by the CPU, one or more steps of the method described above can be executed. Alternatively, in other embodiments, the CPU can be configured to execute the method by any other appropriate means (for example, by means of firmware).

[0127] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that can be used include: Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), Systems on a Chip (SOCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0128] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowchart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, executed partially on the machine and partially on a remote machine as an independent software package, or executed entirely on a remote machine or server.

[0129] In the context of the present invention, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0130] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A multi-UAV path collaborative planning method based on an improved sparrow search algorithm, characterized in that: The method comprises: Establish the spatiotemporal dynamics model and three-dimensional environment model of multiple UAVs; The cost functions of the drone's energy consumption, path length, time error, and obstacle avoidance capability are constructed respectively, and different weights are assigned to the constructed cost functions to construct the objective function; The constructed objective function is used as the fitness function of the improved sparrow search algorithm. A time difference penalty term is added to the objective function to realize three-dimensional path planning for multiple UAVs to reach the same target point simultaneously from different starting positions.

2. The multi-UAV path collaborative planning method based on the improved sparrow search algorithm according to claim 1 is characterized in that: The objective function is specifically: F=w1.F1+w2.T diff +w3.P obstacle +w4.E, Where w1, w2, w3 and w4 are weight coefficients, F is the objective function, F1 is the cost function of the path length, T diff is the cost function of the time error, P obstacle is the cost function of obstacle avoidance capability, E is the cost function of energy consumption, and the weight coefficient w2 of the time error cost function is greater than the other weight coefficients.

3. The multi-UAV path collaborative planning method based on the improved sparrow search algorithm according to claim 2 is characterized in that: The cost function of the path length is the accumulation of the Euclidean distances between all adjacent path points on the three-dimensional space path; The cost function of the time error is specifically: Where, T i is the arrival time of each i-th UAV, T max is the maximum arrival time of all drones, o is the first drone, and O is the number of drones; The cost function of the energy consumption is specifically: E = k1.F1, In the formula, E represents the energy consumption of the UAV during flight; k1 is the energy consumption coefficient, which is a fixed constant.

4. The multi-UAV path collaborative planning method based on the improved sparrow search algorithm according to claim 2 is characterized in that: The cost function of the obstacle avoidance capability is specifically: Where d is the distance between the drone and the obstacle, D safe is the safe distance between the drone and obstacles, D alert is the critical distance between the drone and the obstacle, P obstacle is the cost function of the obstacle avoidance capability of the UAV, o is the first UAV, O is the number of UAVs, N1 is the number of path points, n1 is the starting point of the path, and k is the penalty intensity.

5. The multi-UAV path collaborative planning method based on the improved sparrow search algorithm according to claim 1 is characterized in that: The improved sparrow search algorithm comprises: Use Latin hypercube sampling technique to initialize the population; Introduce a phased control compensation strategy and a nonlinear attenuation factor in the position update of the finder, and dynamically adjust the dynamic step size at different stages of the algorithm; A chaotic cosine variation factor is introduced in the follower's position update to enhance the follower's search range for unknown areas.

6. The multi-UAV path collaborative planning method based on the improved sparrow search algorithm according to claim 5 is characterized in that: The improved location update of the discoverer is as follows: In the above formula, R2 and ST are safety value and warning value respectively; Q is a normally distributed random number; is the location of the discoverer after the update; X d is the current location of the finder; X b is the optimal position; μ is the nonlinear factor; β(t) is the dynamic step size; ε(t) is the random disturbance; When the safety value is less than the warning value, the sparrow will conduct a global search; when the safety value is greater than or equal to the warning value, the sparrow will perform a random walk in a normally distributed manner.

7. The multi-UAV path collaborative planning method based on the improved sparrow search algorithm according to claim 6 is characterized in that: The random disturbance and the dynamic step size are set to different values ​​according to the search stage, wherein the search stage is divided into the early stage, the middle stage and the late stage according to the ratio of the current number of iterations to the maximum number of iterations, specifically: Among them, ε(t) is the random disturbance; μ is the nonlinear attenuation factor, T is the maximum number of iterations; t is the current number of iterations, and U is the uniform distribution function; Among them, initial_step_size is the step size of the early stage; mid_step_size is the step size of the mid-term stage; final_step_size is the step size of the late stage; T is the maximum number of iterations; t is the current number of iterations; γ is a hyperparameter, which controls the rate of step size decay in the initial stage of the sparrow search algorithm.

8. The multi-UAV path collaborative planning method based on the improved sparrow search algorithm according to claim 5 is characterized in that: Follower location updates include: Define the chaotic cosine change factor, which is expressed as: in, δ is the intensity of the chaos factor, t is the current number of iterations, and T is the maximum number of iterations; The chaotic cosine change factor is added to the follower's position update, and the improved follower position update formula is: In the above formula, Q is a normally distributed random number; x ′ w,j is the global optimal position; x i ′ ,j is the current follower's position; is the position of the discoverer; L is the learning factor; A + is a normally distributed random number; k is a random integer used to generate periodic changes; n is the number of sparrows in the population; i is the index of the sparrow individual in the entire population; t is the current iteration number; j represents the dimension index of the sparrow.

9. The multi-UAV path collaborative planning method based on the improved sparrow search algorithm according to claim 1 is characterized in that: The time difference penalty term is specifically: Where, T i is the arrival time of each i-th UAV, T max is the maximum arrival time of all drones, N is the number of drones, T diff1 is the time difference penalty term.

10. The multi-UAV path collaborative planning method based on the improved sparrow search algorithm according to claim 1 is characterized in that: The process of establishing the three-dimensional environment model includes: simulating the port environment model by using a grid map, setting obstacles that the drone cannot pass through as regular three-dimensional cube models of different shapes, setting a no-fly zone according to the size and position of the obstacles, defining the first i The projection coordinates of the obstacle on the xOy plane are The radius of the obstacle is Height is The number of no-fly zones is N, and the set of no-fly zones created is as follows:

Images