Unmanned aerial vehicle dynamic path planning method based on crow search algorithm

Abstract

The embodiment of the invention discloses an unmanned aerial vehicle dynamic path planning method based on a crow search algorithm. A good path planning effect can be obtained in a real environment. According to the method, a crow search algorithm is applied to path planning of the unmanned aerial vehicle, a fitness function is adjusted according to a currently detected dynamic event in each round of iteration, and various behavior strategies for a crow population are designed, so that dynamic changes of a task scene can be quickly responded, and the path planning accuracy of the unmanned aerial vehicle is improved. And a target path planning scheme set meeting a plurality of optimization targets is provided in time.

Description

Technical Field

[0001] The present application relates to the field of UAV technology, and in particular to a UAV dynamic path planning method based on a crow search algorithm. Background Art

[0002] Path planning is one of the key technologies for controlling drones to complete tasks efficiently. It is used to plan the optimal path from the starting point to the target point for the drone to minimize the threat cost and maximize the mission completion effect.

[0003] Current path planning methods mainly focus on solving static path planning problems. However, in a real environment with high uncertainty, various dynamic events (such as mobile threat sources, moving targets, and deteriorating weather conditions) will make it difficult for drones to reach the target point as planned to perform tasks. In other words, current path planning methods are difficult to achieve good path planning results in a real environment. Summary of the invention

[0004] The purpose of the embodiment of the present application is to provide a dynamic path planning method for an unmanned aerial vehicle based on a crow search algorithm, which can achieve better path planning results in a real environment.

[0005] In a first aspect, an embodiment of the present application provides a method for dynamic path planning of a drone based on a crow search algorithm, the method comprising: Initialize subpopulations for multiple optimization objectives respectively, each of the subpopulations includes N randomly initialized crows, the positions of the crows are used to characterize the path planning scheme of the UAV, and N is a positive integer; Iteratively updating the sub-archives corresponding to the plurality of sub-populations based on the crow search algorithm, and in each iteration, adjusting the fitness function according to the currently detected dynamic events, and updating the position of each crow based on the diverse behavior strategies, wherein the sub-archives are used to store non-dominated solutions in the corresponding sub-populations; The main archive is updated according to the sub-archives obtained in each round of iteration, and the target path planning solution set of the UAV is determined according to each non-dominated solution in the main archive obtained in the last update, and the main archive is used to store the non-dominated solutions in all sub-populations; The diverse behavior strategies include: In response to the observed behavior of the crows, randomly selecting the crow that each crow needs to follow from the sub-archives corresponding to each crow and the crows associated with each main archive; According to the attention behavior of crows, different perception probabilities are determined for different crows, and, for a first crow that perceives being followed by other crows, the first crow is set to lead the other crows following it to a random location outside the set radius of the food hiding location it memorizes; In response to the fatigue behavior of crows, a flight length that is negatively correlated with the memory time is determined for each crow, and the memory time is used to represent: the time interval for updating the food hiding position memorized by each crow based on the new position of each crow.

[0006] A second aspect of an embodiment of the present application provides a UAV dynamic path planning device based on a crow search algorithm, the device comprising: An initialization module, used to initialize subpopulations for multiple optimization objectives respectively, each of the subpopulations includes N randomly initialized crows, the positions of the crows are used to characterize the path planning scheme of the UAV, and N is a positive integer; A first processing module is used for iteratively updating the sub-archives corresponding to the plurality of sub-populations based on the crow search algorithm, and in each round of iteration, adjusting the fitness function according to the currently detected dynamic events, and updating the position of each crow based on the diverse behavior strategies, wherein the sub-archives are used to store non-dominated solutions in the corresponding sub-populations; A second processing module is used to update the main archive according to the sub-archives obtained in each round of iteration, and determine the target path planning solution set of the UAV according to each non-dominated solution in the main archive obtained in the last update, and the main archive is used to store the non-dominated solutions in all sub-populations; The diverse behavior strategies include: In response to the observed behavior of the crows, randomly selecting the crow that each crow needs to follow from the sub-archives corresponding to each crow and the crows associated with each main archive; According to the attention behavior of crows, different perception probabilities are determined for different crows, and, for a first crow that perceives being followed by other crows, the first crow is set to lead the other crows following it to a random location outside the set radius of the food hiding location it memorizes; In response to the fatigue behavior of crows, a flight length that is negatively correlated with the memory time is determined for each crow, and the memory time is used to represent: the time interval for updating the food hiding position memorized by each crow based on the new position of each crow.

[0007] A third aspect of an embodiment of the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the unmanned aerial vehicle dynamic path planning method based on the crow search algorithm as described in the first aspect.

[0008] In a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the steps of the unmanned aerial vehicle dynamic path planning method based on the crow search algorithm as described in the first aspect are implemented.

[0009] In a fifth aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for dynamic path planning of a drone based on a crow search algorithm as described in the first aspect are implemented.

[0010] It can be seen from the above technical solutions that the present application applies the crow search algorithm to the path planning method of the UAV, and adjusts the fitness function according to the currently detected dynamic events in each round of iteration, so that the crow search algorithm can obtain the optimal path planning scheme that meets the current scene, thereby realizing dynamic path planning for the UAV, and the present application also designs a variety of behavior strategies for the crow population to enhance the search optimization performance of the crow search algorithm, thereby effectively improving the real-time and effectiveness of the crow search algorithm in solving the dynamic path planning problem of the UAV, so that the UAV dynamic path planning method provided by the present application can quickly respond to the dynamic changes of the mission scene, and timely provide a variety of feasible paths that meet the said multiple optimization objectives (i.e., a set of target path planning schemes), so as to obtain better path planning effects in a real environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 A schematic diagram of a flow chart of a dynamic multi-objective optimization evolutionary algorithm provided in an embodiment of the present application; Figure 2 A flowchart of an implementation method of a UAV dynamic path planning method based on a crow search algorithm provided in an embodiment of the present application; Figure 3 A perception probability provided in the embodiment of the present application AP Schematic diagram of the effect of parameters; Figure 4 A flight length provided in an embodiment of the present application fl Schematic diagram of the effect of parameters; Figure 5 A schematic diagram of a splitting option provided in an embodiment of the present application; Figure 6 A schematic diagram of an artificial potential field provided in an embodiment of the present application; Figure 7 A schematic diagram of a threat zone provided in an embodiment of the present application; Figure 8 A schematic diagram of dynamic path planning of a drone in a rescue and search mission scenario provided in an embodiment of the present application; Fig. 9 A schematic diagram of the structure of a UAV dynamic path planning device based on a crow search algorithm provided in an embodiment of the present application; Fig.10 A schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0014] With the in-depth development and widespread application of integrated circuits and unmanned control technologies, unmanned systems represented by unmanned aerial vehicle platforms (i.e., drones) are profoundly changing the production and lifestyle of human society, and single unmanned systems have become common in various fields.

[0015] In the civilian field, the use of drones for oilfield inspections, fixed-point photography, traffic inspections and other related applications has achieved initial results; in the military field, the Reaper, Global Hawk, and Valkyrie drones, which are mainly used for wide-area reconnaissance and search, and the Puma, Black Jack, and Raven drones, which are mainly used for long-range detection and close-range operations, are all well-known. In the aforementioned applications, path planning is one of the key technologies for controlling drones to complete tasks efficiently. When planning the path of a drone, many constraints need to be considered (such as platform capabilities, threat sources, mission objectives, and collaborative relationships, etc.), so as to plan the optimal path from the starting point to the target point for the drone to minimize the cost and maximize the task completion effect.

[0016] Current path planning methods mainly focus on solving static path planning problems. However, in a real environment with high uncertainty, various dynamic events (such as mobile threat sources, moving targets, and deteriorating weather conditions) will make it difficult for drones to reach the target point as planned to perform tasks. That is, in real mission scenarios with high uncertainty, dynamic emergencies will reduce the feasibility of drones flying along the original path and increase the cost, which seriously limits the application of static path planning methods in actual scenarios. Therefore, the key is how to respond to changes in the scene to dynamically plan the path of the drone and update the optimal path in real time.

[0017] Based on the above analysis, in order to address the problem that current path planning methods are difficult to achieve good path planning effects in real environments, an embodiment of the present application provides a UAV dynamic path planning method based on a crow search algorithm, which can quickly respond to dynamic changes in mission scenarios and promptly provide a variety of feasible paths that meet the multiple optimization objectives, thereby achieving better path planning effects in real environments.

[0018] First of all, in order to facilitate understanding of the technical solutions provided by the present application, the main technical concepts involved in the embodiments of the present application are briefly described below.

[0019] 1. Dynamic Multi-Objective Problems (DMOP) The DMOP problem refers to a problem in which the problem conditions or objectives change dynamically over time and contain multiple optimization objectives. It is often used to model the dynamic task allocation problem of unmanned swarms, and often uses methods based on artificial potential fields, reinforcement learning methods, heuristic search methods, and evolutionary algorithm-based methods to solve the DMOP problem. For example, minimizing DMOP is defined as follows:

[0020] in, x It's space R n The decision vector in t is a time variable (i.e., time slot); t , F ( x , t ) is the evaluation decision vector x The objective function vector, M t is the optimization goal (i.e., the objective function f ), n g ( t )and n h ( t) are the inequality constraints and equality constraints The number of

[0021] 2. DMOP related definitions In time slot t , DMOP is transformed into static multi-objective optimization, on this basis, the following definitions are made: Definition 1: In the time slot t , given 2 solutions x and y , when the following constraints are met, it is called x Dominate y , expressed as :

[0022] Definition 2: In the time slot t , given the solution , if and only if there is no other solution that dominates it, that is, hour, x is the Pareto optimal solution; decision space Ω The set of all Pareto optimal solutions in is called the Pareto solution set (ParetoSet, PS), as shown in the following formula:

[0023] Definition 3: In the time slot t , PS t The corresponding set of target vectors is called the Pareto Front (PF), as shown below:

[0024] Based on the above definition, dynamic multi-objective optimization aims to track the dynamically changing PS while making the solution set have good convergence and distribution. In this regard, the dynamic multi-objective optimization evolutionary algorithm handles the DMOP problem through the static multi-objective optimization evolutionary algorithm, change detection and change response. The flow chart of the dynamic multi-objective optimization evolutionary algorithm is shown in Figure 1 shown.

[0025] 3. Evolutionary Algorithms Evolutionary algorithm is a method that simulates the biological behavior or natural phenomenon of survival of the fittest in nature to iteratively optimize the solution to a problem. Evolutionary algorithms have been deeply studied and widely used in many fields. Among them, the Crow Search Algorithm (CSA) is a novel evolutionary algorithm proposed in recent years. It has the advantages of few parameters, easy implementation, and strong optimization performance, and has been applied in many fields.

[0026] 4. Crow Search Algorithm CSA is inspired by the intelligent behavior of crows: crows look for hidden food in other birds and try to steal it. When they detect a thief, they mislead the thief to protect their own food.

[0027] Assume that d Dimensional space supply N crows search, and the position of each crow is represented by an n-dimensional vector Each crow will remember the location of its hidden food. ,Once the crow finds a better location, it moves the food to this new location.,In the iterative process, the crow moves in space to find a better location.

[0028] Assume that in the iteration τ ,crow i Decided to follow the crow j To steal its food, there are two situations: Case 1: Crow j Didn't notice the crow i Tracking it, then, the crow i Will be to the crow j The memorized food storage position (hereinafter referred to as the memory position) is moved closer and its position is changed from Updated to ,in:

[0029] in, is a random number uniformly distributed between 0 and 1, fl Represents crow i Flight Length, Represents crow j memory location.

[0030] Case 2: Crow j Notice the crow i is tracking it, then the crow j Will guide the crows i Fly to a random location in the search space, thus preventing its food from being stolen.

[0031] Based on the above two situations, the crow i The position update formula can be expressed as:

[0032] in, r j is a random number uniformly distributed between 0 and 1, AP Represents the awareness probability of the crow. Based on the above, the pseudo code of CSA is as follows: Input: population size N, maximum number of iterations T, perception probability AP ; Output: The memory location of the optimal crow; 1. Initialize N crows; 2. Update the memory position of each crow; 3.while current iteration number t <T 4.for crow i=1: N 5. Randomly select a crow j; 6. Update the position of crow i according to the above position update formula; 7. end for 8. Check the feasibility of the crow moving to the new position. If not, keep the original position; 9. Evaluate the fitness of each crow in its new location; 10. Update the crow's memory location; 11.end while 12. Output the optimal solution based on the fitness of each crow at the food hiding location.

[0033] Among them, step 8 is used to check whether each crow can move to a new position. If the position is beyond the search space, the crow will remain in the original position; step 9 is specifically to evaluate the fitness of the individual crow at the current position according to a pre-defined fitness function; in step 10, if the fitness of the crow at the new position is better than its original memory position, then the crow will update the memory position.

[0034] See also Figure 2 As shown, it is an implementation flow chart of a method for dynamic path planning of a UAV based on a crow search algorithm provided in an embodiment of the present application. The method may include the following steps: Step S101: randomly initialize subpopulations for multiple optimization objectives, each of which contains N crows. The positions of the crows are used to characterize the path planning scheme of the drone, and N is a positive integer.

[0035] In the specific implementation, multiple optimization targets are determined according to the actual task scenario requirements, and a sub-population containing N crows is randomly initialized for each optimization target. During the random initialization process, the initial position of each randomly generated crow is determined as the initial food hiding position it memorizes.

[0036] Step S102: iteratively updating the sub-archives corresponding to the plurality of sub-populations based on the crow search algorithm, and in each iteration, adjusting the fitness function according to the currently detected dynamic events, and updating the position of each crow based on the diverse behavior strategy, wherein the sub-archive is used to store the non-dominated solutions in the corresponding sub-population; wherein the diverse behavior strategy includes: In response to the observed behavior of the crows, randomly selecting the crow that each crow needs to follow from the sub-archives corresponding to each crow and the crows associated with each main archive; According to the attention behavior of crows, different perception probabilities are determined for different crows, and, for a first crow that perceives being followed by other crows, the first crow is set to lead the other crows following it to a random location outside the set radius of the food hiding location it memorizes; In response to the fatigue behavior of crows, a flight length that is negatively correlated with the memory time is determined for each crow, and the memory time is used to represent: the time interval for updating the food hiding position memorized by each crow based on the new position of each crow.

[0037] It is understandable that the path planning problem of drones is a nondeterministic polynomial time (NP) hard problem. It is difficult to find an exact solution to the problem in polynomial time. The dynamic characteristics added on this basis will make the problem more complicated. To solve this problem, this application models it as a DMOP problem, and considering that the evolutionary algorithm has the advantages of low computational complexity, high globality, and strong applicability, this application selects the evolutionary algorithm (specifically the crow search algorithm) to solve the dynamic path planning problem of drones.

[0038] In specific implementation, when the drone executes the crow search algorithm to iteratively update the sub-archives corresponding to each of the multiple sub-populations, the fitness function (which can be constructed based on one or more objective functions) will be adjusted according to the currently detected dynamic events. For example, when a new threat area is detected (such as a threat area related to the collision threat of a moving flying object or a threat area related to the threat of sudden weather), the objective function related to the threat cost in the fitness function is updated; for another example, when a new target point is detected, the objective function related to the path length in the fitness function is updated; thereby, the crow search algorithm can obtain the optimal crow memory position (i.e., the preferred path planning solution) that meets the current scenario according to the adjusted fitness function, thereby realizing dynamic path planning for the drone.

[0039] In order to enhance the search optimization capability of the crow search algorithm, this application also designs a variety of behavior strategies based on the complex behavior of crows.

[0040] Specifically, with respect to the observation behavior of crows, the present application designs each crow to randomly select (e.g., based on roulette selection) the crow that each crow needs to follow from its corresponding sub-archive and the crows associated with the main archive, thereby simulating the behavior that crows will observe other crows and are more likely to follow those crows that hide food in better locations.

[0041] As a possible implementation manner, randomly selecting a crow that each crow needs to follow from the sub-archive corresponding to each crow and the crows associated with the main archive, includes: For each crow, when a random value between 0 and 1 is r o >0.5, select the crow that the crow needs to follow from the crows associated with the main archive; For each crow, when a random value between 0 and 1 is r o When ≤0.5, the crow that the crow needs to follow is selected from the crows associated with the sub-archive corresponding to the crow.

[0042] Regarding the attention behavior of crows, the present application determines different perception probabilities for different crows, such as randomly distributing their perception probabilities within a range so that the attention of different crows is different, and, for a first crow that perceives that it is being followed by other crows, sets the first crow to lead the other crows that are following it to a random location outside the set radius of the food hiding location it remembers.

[0043] In response to the fatigue behavior of crows, a flight length that is negatively correlated with the memory time is determined for each crow, and the memory time is used to characterize: the time interval for updating the food hiding location memorized by each crow based on the new location of each crow; thereby, the flight length of the crows is gradually reduced as they become tired until they find a better food hiding location.

[0044] Optionally, when the crow flies to a new unreachable location, the variables in the new location that are outside the search bounds are bounded before updating the crow's position.

[0045] It should be noted that the perceived probability AP and flight length fl It is the key parameter that controls the optimization tendency of CSA. Figure 3 The perceived probability shown AP Schematic diagram of the effect of parameters. Figure 3 middle r j is a random number, For Crowi location, For Crow j The memory location of AP The probability of crows searching around the memory location is reduced by , which forces CSA to explore the search space globally (which can be determined by the physical properties of the drone or divided artificially); in contrast, by reducing AP Values ​​of , CSA is more likely to perform local searches in the neighborhood of the current good solution.

[0046] Reference Figure 4 Flight length shown fl Schematic diagram of the effect of the parameters, which illustrates the situation when situation 2 in the above crow search algorithm occurs fl The parameters affect the effect. Figure 4 In, if fl is set to a value less than 1, the crow i The reachable position of the crow i In iteration τ +1 position ) is restricted to crows i In iteration τ Location and crows j In iteration τ Memory location On the dotted line between fl If the value is set to greater than 1, the crow i The search range will expand, and its next position on the dotted line may exceed .

[0047] Considering the perceived probability in CSA AP and flight length fl These two key parameters play an important role in its optimization performance, but they are originally static and not suitable for dynamic environments. Therefore, based on the complex behavior of crows, this application designs the two as dynamic parameters in a diverse behavior strategy to enhance the search optimization performance of the crow search algorithm, making it more suitable for solving the dynamic path planning problem of drones.

[0048] Step S103: Update the main archive according to the sub-archives obtained in each round of iteration, and determine the target path planning solution set of the UAV according to each non-dominated solution in the main archive obtained by the last update. The main archive is used to store non-dominated solutions in all sub-populations.

[0049] In the specific implementation, an archive is used to store the better solutions in the population, that is, the archive can be regarded as a Pareto solution set. Specifically, in each round of iteration, each sub-archive (that is, the Pareto solution set corresponding to each sub-population) is updated, and then all sub-archives (that is, the Pareto solution set corresponding to all sub-populations) are used to update the main archive (that is, the Pareto solution set corresponding to the entire group composed of all sub-populations). By decoding each non-dominated solution in the main archive obtained by the last update, the target path planning solution set of the UAV can be obtained, and the subsequent UAV can select a path planning solution from it according to the decision maker's preference and fly according to the path planning solution.

[0050] It can be seen from the above technical solutions that the present application applies the crow search algorithm to the path planning method of the UAV, and adjusts the fitness function according to the currently detected dynamic events in each round of iteration, so that the crow search algorithm can obtain the optimal path planning scheme that meets the current scene, thereby realizing dynamic path planning for the UAV, and the present application also designs a variety of behavior strategies for the crow population to enhance the search optimization performance of the crow search algorithm, thereby effectively improving the real-time and effectiveness of the crow search algorithm in solving the dynamic path planning problem of the UAV, so that the UAV dynamic path planning method provided by the present application can quickly respond to the dynamic changes of the mission scene, and timely provide a variety of feasible paths that meet the said multiple optimization objectives (i.e., a set of target path planning schemes), so as to obtain better path planning effects in a real environment.

[0051] As a possible implementation, j A crow (i.e. a crow j ) Determined by the following formula:

[0052] in, represents the preset perception probability, represents the first number in the digital sequence generated by the chaotic map Fuch map j A value can be calculated by the following formula:

[0053] in, represents the number in the sequence of numbers generated by the Fuch map j-1 value.

[0054] In the specific implementation, when a crow perceives that another crow is following it (corresponding to the random number In the case of a crow that is following it, it (the first crow) will lead the crow that is following it to a random location outside the radius of its memory location.

[0055] Optionally, in denoting the first crow as the first j crow and denote the other crows that track the first crow as i In the case of a crow, a random position outside the set radius of the food hiding position memorized by the first crow is determined by the following formula:

[0056] in, Indicates j crows are iterating τ Lieutenant General i The random location that the crows lead to is d The value of the dimension, and Used as the first i crows are iterating τ +1 position d The value of the dimension; Indicates j crows are iterating τ The first food storage location in the memory d The value of the dimension; and ub d They are d The lower and upper bounds of the decision variables (i.e., the encoding dimensions, such as the upper and lower bounds of the yaw angle change); R d Indicates d dimensional radius, and , Represents the radius coefficient, which can be set to 0.05 in practical applications.

[0057] As a possible implementation, the flight length of each crow is determined by the following formula:

[0058] in, Indicates i crows are iterating τ The flight length in Indicates the preset flight length, Indicates i The memory time corresponding to each crow.

[0059] As a possible implementation, the position of the crow includes: the yaw angle change value, pitch angle and length factor corresponding to each segment in the path planning scheme, each segment is composed of every two consecutive waypoints from the starting point to the target point in the path planning scheme, and the first waypoint from the starting point to the target point in the path planning scheme is the first waypoint from the starting point to the target point. j+1 waypoint N j+1 The three-dimensional coordinates of are determined by the following formula:

[0060] in,( x j+1 , y j+1 , z j+1 ) indicates the j +1 waypoint N j+1 The three-dimensional coordinates of x j , y j , z j ) indicates the j Waypoints N j The three-dimensional coordinates of θ j Indicates j Waypoints The yaw angle, , α j and Respectively indicated by j Waypoints N j and j +1 waypoint N j+1 The flight segments N j N j+1 The corresponding yaw angle change value, pitch angle and length factor, R Indicates the maximum range of the drone.

[0061] In this implementation, n waypoints from the starting point to the target point are used to form the UAV path (i.e., the path planning scheme), and considering the actual application scenario, the UAV yaw angle change value and pitch angle range are set to (-π / 2, π / 2), and the length factor coding is set to a value range of (0, 0.5). When the coding is converted into the actual flight segment length, it is . Since the n -1 waypoint to the end point (i.e., the target point) is determined by the yaw angle, pitch angle, and distance ratio (i.e., length factor), so the encoding length is 3( n -1). Assuming that the total number of waypoints in the path planning scheme is 5, the encoding example corresponding to the path planning scheme (i.e., the position of the crow) is shown in Table 1.

[0062] Table 1 3D path planning coding example

[0063] It is understandable that in each round of iteration, the new position of each crow will be decoded to obtain its fitness. If the traditional (x, y, z) encoding method is used in this process, the search space will include the range of each axis of the entire environment (i.e., the x-axis, y-axis, and z-axis), resulting in low search efficiency. To solve the above problems, this application introduces heading angle changes and pitch angles in the coding design to reflect certain physical properties of the drone and narrow the search range, and this application also introduces the distance ratio of the maximum range (i.e., the length factor) to further narrow the coding search range, thereby effectively improving the search efficiency.

[0064] As a possible implementation, the method further includes: In each round of iteration, for each crow, if the new position of the crow and the food hiding position it memorizes do not dominate each other, and the fitness of the new position of the crow is worse than that of the food hiding position it memorizes on the corresponding optimization target, the new position of the crow is not updated to the food hiding position it memorizes, but the new position of the crow is added as a non-dominated solution to the main archive and the sub-archives corresponding to the sub-populations other than the sub-population to which the crow belongs; When the amount of data stored in the sub-archive exceeds the maximum capacity of the sub-archive, a truncation technique based on congestion is used to reduce the amount of data stored in the sub-archive to retain the most diverse non-dominated solutions in the sub-archive.

[0065] In specific implementation, the present application introduces a multi-population co-evolution mechanism into CSA to improve its population diversity and search capabilities. In this mechanism, each subpopulation focuses on optimizing a given optimization target and maintains a sub-archive to store the non-dominated solutions in the subpopulation. Then, the solutions in the sub-archive are used to update the main archive, so that the main archive stores the non-dominated solutions in the entire group; when the new position of the crow does not dominate the memory position, but the fitness of the new position on the corresponding optimization target is worse, the new position will not be updated as the memory position, but will be immediately introduced into the sub-archive and main archive corresponding to other subpopulations.

[0066] When a sub-archive exceeds a preset maximum size (i.e., maximum capacity), a crowding-based truncation technique is used to retain the most diverse solutions in the sub-archive, which mainly includes the following steps: 1. Sort by target: Sort the individuals in the current non-dominated layer in ascending order according to the value of each objective function (such as path length, threat level under the path, etc.).

[0067] 2. Processing boundary points: Set the crowding distance part of the minimum and maximum values ​​of each objective function corresponding to the individual (i.e., boundary point) to infinity (or maximum value) to ensure that it is retained first.

[0068] 3. Calculation of intermediate individuals: For non-boundary individuals, calculate the difference (i.e. distance) between adjacent individuals on each objective function. ,Right now:

[0069] in, and Indicates i+ 1 individual and i -1 individual in the objective function The value on .

[0070] The total crowding distance is the sum of contributions from all M targets, namely: .

[0071] 4. Truncate the selection process: (1) Non-dominated sorting: Divide the population (e.g., subpopulation) into multiple non-dominated levels (e.g., the first level is the Pareto optimal solution).

[0072] (2) Fill the population layer by layer: Starting from the optimal layer, each layer is added to the new population in turn until a certain layer can no longer accommodate all the members.

[0073] (3) Crowding distance sorting: For layers that cannot be fully accommodated, sort them from large to small according to the crowding distance, and select individuals with large distances until the population is filled.

[0074] For example, suppose there are 10 individuals in a non-dominated layer and 5 of them need to be selected. Then, the crowding distance of each individual needs to be calculated first, and after sorting by crowding distance, the 5 individuals with the largest crowding distance are selected (the memory positions of these 5 individuals will also be used as sub-archives of the corresponding population). It can be understood that the boundary individuals (such as the minimum value of objective function 1) will be retained first because of the extremely large crowding distance, and then the sparser individuals around them will be selected for retention.

[0075] In this embodiment, the present application promotes the co-evolution of subpopulations by exchanging elite individuals between subpopulations (i.e., in each round of iteration, taking out the best and worst individuals under the corresponding optimization objective from each subpopulation and copying them to other subpopulations), and then uses a truncation technique based on crowding distance to retain the most diverse solutions in the subpopulations.

[0076] As a possible implementation, the method further includes: If a dynamic event is currently detected, then in a new round of iterations, the memory time of the N crows is mt >1, merge the food hiding location memorized by the second crow into the sub-archive corresponding to the sub-population to which it belongs; After merging the food hiding location memorized by the second crow into the sub-archive corresponding to the sub-population to which it belongs, approximating the target space of the plurality of sub-populations into one dimension, and dividing the target space by a dichotomy method to obtain a dividing line; For non-dominated solutions located in the set area where the dividing line is located, generating a repulsive force response to the dynamic event in the threat area in the artificial potential field environment to adjust the non-dominated solutions; For non-dominated solutions that are not located in the set area where the segmentation line is located, the non-dominated solutions are adjusted by normal perturbation variation.

[0077] In this implementation, considering that when a dynamic event occurs, the food hiding location memorized by the crow may no longer be a Pareto solution or even an infeasible solution, in a new round of iterations, the present application does not directly use the crow's memory location, but selects the old solution based on the split selection method and memory reuse strategy as the initial solution in the new environment. The process specifically includes: First, remember the time mt The current positions and memory positions of the crows with position >1 are different, so the memory positions of these crows (i.e., the second crow) are merged into the corresponding sub-population to achieve the reuse of the crow's memory position.

[0078] Then, the split selection method is used to select half of the old solutions with better diversity in the subpopulation. Specifically: refer to Figure 5 The schematic diagram of split selection shown in the figure shows that since each subpopulation focuses on its corresponding optimization goal, the target space of the subpopulation can be approximated as one-dimensional; then, the target space is divided by bisection, and the solution closest to the dividing line is selected. It can be understood that this method only calculates one-dimensional distance, so the time overhead is lower than other selection methods based on Euclidean distance.

[0079] It should be noted that Figure 5 For example, the distribution of each solution in the sub-archive in the entire target space is relatively wide, and it can be determined that its diversity is good, that is, it can provide better solutions under multiple preferences (for example, only considering the objective function f 2, you can choose the edge solution in the lower right corner, which can provide the best objective function f 2 corresponds to the optimization target result); if a large number of solutions are clustered in a certain range, it means that the diversity of these solutions is poor, that is, the solutions provided by these solutions are convergent and difficult to be used for different decision preferences For all selected solutions (i.e., non-dominated solutions located in the set area where the dividing line is located), the repulsive response dynamic events of the threat zone in the artificial potential field generation environment are used to adjust the non-dominated solutions (such as adjusting the yaw angle change value and pitch angle for the affected waypoints), thereby obtaining a feasible solution that can avoid threats in the new environment. It should be noted that the artificial potential field (APF) method simulates the concept of potential field in physics, and guides the drone to move to the target point by constructing a virtual gravitational potential field and a repulsive potential field, while avoiding collisions with obstacles in the environment. The schematic diagram of the artificial potential field is shown in the figure below. Figure 6 As shown, Figure 6 middle F Indicates combined force, F att Represents gravity, F rep , F rep1 and F rep2 Indicates repulsion.

[0080] For the unselected solutions (i.e., the non-dominated solutions that are not located in the set area where the segmentation line is located), the encoding values ​​are subjected to normal perturbation mutation to enhance the algorithm's ability to adapt to new environments, such as , where s represents the normal disturbance variation, Indicates the yaw angle change value.

[0081] As a possible implementation, the method further includes: In each iteration, for each crow, when determining to update the food hiding location memorized by the crow, the memory time corresponding to the crow is set. mt is 1; In each iteration, for each crow, when it is determined that the food hiding location memorized by the crow is not to be updated, the memory time corresponding to the crow is set. mt = mt +1; In each iteration, the corresponding memory time is used mt =1 to update the corresponding sub-archive and the main archive.

[0082] In the specific implementation, in order to speed up the search process of the entire group, this application also introduces memory time and a mutation strategy based on memory time. Specifically, each crow in the group saves a memory time parameter associated with its memory. mt , during the iteration, each time the memory position is updated, mt Set to 1, otherwise mt = mt +1. Therefore, the present application introduces the memory time as an indication of updating the archive, so that after each position update step, only mt =1 crows to update the sub-archives and main archives corresponding to the relevant sub-populations, thereby reducing the workload of archive updates.

[0083] As a possible implementation, the multiple optimization objectives at least include minimizing the threat cost of the drone and minimizing the path length of the drone. When the threat zone is regarded as a cylinder and the drones that can cross the threat zone are regarded as two concentric cylinders with different heights, the expressions of the multiple optimization objectives are as follows:

[0084] in, represents the path length of the UAV to be minimized, represents the minimization of the threat cost of drones, N j Indicates the first j Waypoints, N j+1 Indicates the first j +1 waypoint, ( x j , y j , z j )express N j The three-dimensional coordinates of x j+1 , y j+1 , z j+1 )express N j+1 The three-dimensional coordinates of n Indicates the number of waypoints, K represents the set of threat zones, T k Indicates k The threat value of each threat zone, and Represent the heights of the outer and inner cylinders of the threat area, and Respectively represent k The radius of the inner and outer cylinders in the threat zone, d k Indicates flight segment N j N j+1 With kThe plane distance between the center points of the threat areas, α and β is the threat coefficient, and 0< α < β <1, the schematic diagram of the threat area is shown in Figure 7 shown.

[0085] Optionally, the threat value of the threat zone can be determined based on influencing factors associated with different threat types. For example, the threat value of a mid-air collision can be determined based on influencing factors such as the size, type, and speed of the opposing aircraft; the threat value of severe weather can be determined mainly based on influencing factors such as the weather type and degree. For example, a higher threat value can be assigned to large thunderstorms.

[0086] For example, taking the mission scenario of a drone performing rescue search as an example (the mission scenarios of the embodiments of the present application are not limited to this), in this mission scenario, the drone needs to plan a reasonable maneuvering path for itself in the presence of a threat area, so as to maneuver to the target point at a lower cost.

[0087] Reference Figure 8 The schematic diagram of the dynamic path planning of the UAV in the rescue and search mission scenario shown is that the UAV needs to avoid threat areas 1 to 6, go to target points A and B to perform rescue and search missions, and return to the starting point after completing the mission.

[0088] Considering three-dimensional path planning, the threat area is regarded as a cylinder, some of which can be crossed by adjusting the flight altitude. For example, the threat levels of the drone are different when flying at different altitudes in the No. 2 mountain area and the No. 5 lightning weather threat area. These threat areas (that is, the threat areas that can be crossed) can be regarded as two concentric cylinders with different heights.

[0089] The dynamic events in this mission scenario include the collision threat of mobile flying object No. 3, the sudden weather threat No. 5, and the newly added target point B. These dynamic events will change the problem function, resulting in changes in the fitness (and default degree) of individuals in CSA, making the optimal path no longer meet the requirements. Therefore, it is necessary to remodel the path planning problem of the mission scenario, such as updating the expression of the optimization objective, and adjusting the fitness function according to the updated expression of the optimization objective, so that the crow search algorithm can obtain the optimal path planning solution that meets the current scenario.

[0090] Optionally, the multiple optimization objectives further include: a violation degree of the crow individual, the violation degree is used to measure the degree to which the crow individual violates a constraint condition, and the constraint condition includes at least one of the following: Range constraints: ; Maximum Threat Constraint: ; Maximum and minimum height constraints: ; in, represents the path length of the UAV, N j Indicates the first j Waypoints, N j+1 Indicates the first j +1 waypoint, Indicates the maximum range of the drone. represents the threat cost of drones, T k Indicates k The threat value of each threat zone, U Indicates the maximum threat cost that drones can bear, Indicates j The z-axis coordinate of the waypoint, and They represent the minimum and maximum flight altitudes of the drone, n represents the number of waypoints, K Represents a collection of threat zones.

[0091] In the specific implementation, the violation degree is used to reflect the degree to which the drone violates the above constraints. For example, for the range constraint, its corresponding violation degree can be expressed as max{0, }, other inequality constraints such as the maximum threat constraint and the maximum and minimum height constraints can also be determined in a similar manner to determine the corresponding violation degree, and the violation degrees corresponding to multiple constraints can be obtained by weighted sum and other methods.

[0092] Optionally, in order to reduce the impact of inconsistent dimensions, each default degree can be normalized, that is: norm =(v i -v max) / (v min -v max) , where v max and v min Represent the maximum and minimum values ​​of the default degree corresponding to the sub-population, v i represents the default degree of individuals in the subpopulation, v norm Represents the normalized default degree.

[0093] In this embodiment, the present application regards the default degree of individual crows as an additional optimization target to establish a sub-population, so as to continuously optimize the default degree during the iteration process, thereby prompting the entire population to search and optimize towards the non-default space.

[0094] For example, the pseudo code of a three-dimensional dynamic path planning method for a drone provided in this application is as follows: Input: population size N, maximum number of iterations T, UAV 3D dynamic path planning problem model M; Output: a set of optimal path planning solutions (i.e., a set of target path planning solutions); 1. According to the problem model M, randomly initialize a sub-population containing N crows for each optimization goal; 2. while t <T 3. Evaluate each crow in all sub-populations, update each sub-archive and use the sub-archive to update the main archive; 4. If dynamic changes occur 5. Apply dynamic response strategies based on artificial potential fields; 6.end if 7. Update the memory position and reduce the population size to N through non-dominated sorting; 8. Apply the multi-population mechanism, and each sub-population copies the optimal and worst solutions under the corresponding optimization objective to other sub-populations; 9.for each subpopulation 10.for crow i=1: N 11. Update location based on diverse behavior strategies; 12. end for 13. end for 14. Check the feasibility of the crow moving to the new position. If not, keep the original position. 15.end while 16. Update each sub-archive and main archive, and output the optimal path planning solution set in the main archive.

[0095] Based on the above embodiments, the present application proposes a UAV dynamic path planning method based on the crow search algorithm for the dynamic path planning problem of UAVs, designs appropriate individual coding for UAV path planning; introduces a multi-population collaboration mechanism to improve the population diversity of CSA, thereby improving the convergence and diversity of the solution set generated by the method, and enables the method to effectively handle constraints; designs diverse behavior strategies for the crow population to enhance the search optimization performance of the method; adopts a dynamic response strategy based on an artificial potential field, so that the method can quickly respond to dynamic events, avoid sudden threats in the environment in time, and provide a better path planning solution in the new environment.

[0096] Through the above improvements, the real-time and effectiveness of the crow search algorithm in solving the dynamic path planning of UAVs can be effectively improved, so that the method can quickly respond to dynamic emergencies in the scene and provide feasible paths with lower expected costs in a timely manner.

[0097] It should be noted that, for the method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited by the described order of actions, because according to the embodiments of the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present application.

[0098] The present application also provides a UAV dynamic path planning device based on the crow search algorithm. Fig. 9 As shown, the device comprises: An initialization module, used to initialize subpopulations for multiple optimization objectives respectively, each of the subpopulations includes N randomly initialized crows, the positions of the crows are used to characterize the path planning scheme of the UAV, and N is a positive integer; A first processing module is used for iteratively updating the sub-archives corresponding to the plurality of sub-populations based on the crow search algorithm, and in each round of iteration, adjusting the fitness function according to the currently detected dynamic events, and updating the position of each crow based on the diverse behavior strategies, wherein the sub-archives are used to store non-dominated solutions in the corresponding sub-populations; A second processing module is used to update the main archive according to the sub-archives obtained in each round of iteration, and determine the target path planning solution set of the UAV according to each non-dominated solution in the main archive obtained in the last update, and the main archive is used to store the non-dominated solutions in all sub-populations; The diverse behavior strategies include: In response to the observed behavior of the crows, randomly selecting the crow that each crow needs to follow from the sub-archives corresponding to each crow and the crows associated with each main archive; According to the attention behavior of crows, different perception probabilities are determined for different crows, and, for a first crow that perceives being followed by other crows, the first crow is set to lead the other crows following it to a random location outside the set radius of the food hiding location it memorizes; In response to the fatigue behavior of crows, a flight length that is negatively correlated with the memory time is determined for each crow, and the memory time is used to represent: the time interval for updating the food hiding position memorized by each crow based on the new position of each crow.

[0099] Optionally, j The probability of a crow sensing Determined by the following formula:

[0100] in, represents the preset perception probability, represents the first number in the digital sequence generated by the chaotic map Fuch map j values; In denoting the first crow as j crow and denote the other crows that track the first crow as i In the case of a crow, a random position outside the set radius of the food hiding position memorized by the first crow is determined by the following formula:

[0101] in, Indicates j crows are iterating τ Lieutenant General i The random location that the crows lead to is d The value of the dimension, and Used as the first i crows are iterating τ +1 position d The value of the dimension; Indicates the j crows are iterating τ The first food storage location in the memory d The value of the dimension; and ub d They are d lower and upper bounds for the decision variables; R d Indicates d dimensional radius, and , Represents the radius coefficient.

[0102] Optionally, the flight length of each of the crows is determined by the following formula:

[0103] in, Indicates i crows are iterating τ The flight length in Indicates the preset flight length, Indicates i The memory time corresponding to each crow.

[0104] Optionally, the position of the crow includes: the yaw angle change value, pitch angle and length factor corresponding to each segment in the path planning scheme, each segment is respectively composed of every two consecutive waypoints from the starting point to the target point in the path planning scheme, and the first waypoint from the starting point to the target point in the path planning scheme is respectively composed of the first waypoint from the starting point to the target point in the path planning scheme.j +1 waypoint N j+1 The three-dimensional coordinates of are determined by the following formula:

[0105] in,( x j+1 , y j+1 , z j+1 ) indicates the j +1 waypoint N j+1 The three-dimensional coordinates of x j , y j , z j ) indicates the j Waypoints N j The three-dimensional coordinates of θ j Indicates j Waypoints The yaw angle, , α j and Respectively indicated by j Waypoints and j +1 waypoint The flight segments The corresponding yaw angle change value, pitch angle and length factor, Indicates the maximum range of the drone.

[0106] Optionally, the device further comprises: The third processing module is used for, in each round of iteration, for each crow, if the new position of the crow and the food hiding position it memorizes do not dominate each other and the fitness of the new position of the crow is worse than that of the food hiding position it memorizes on the corresponding optimization target, not updating the new position of the crow to the food hiding position it memorizes, but adding the new position of the crow as a non-dominated solution to the main archive and the sub-archives corresponding to the sub-populations other than the sub-population to which the crow belongs; The fourth processing module is used to reduce the amount of data stored in the sub-archive by using a congestion-based truncation technique when the amount of data stored in the sub-archive exceeds the maximum capacity of the sub-archive, so as to retain the most diverse non-dominated solutions in the sub-archive.

[0107] Optionally, the device further comprises: The fifth processing module is used for, if a dynamic event is currently detected, in a new round of iteration, memorizing the time of the N crows. mt >1, merge the food hiding location memorized by the second crow into the sub-archive corresponding to the sub-population to which it belongs; a sixth processing module, for approximating the target spaces of the plurality of subpopulations into one dimension after merging the food hiding locations memorized by the second crow into the sub-archive corresponding to the subpopulation to which it belongs, and dividing the target spaces by a dichotomy method to obtain a dividing line; A seventh processing module, for adjusting the non-dominated solutions located in the set area where the dividing line is located, based on the repulsive force response of the threat area in the artificial potential field generation environment to the dynamic event, the non-dominated solutions; The eighth processing module is used to adjust the non-dominated solutions that are not located in the set area where the segmentation line is located by normal perturbation variation.

[0108] Optionally, the device further comprises: The first setting module is used to set the memory time corresponding to each crow when determining to update the food hiding location memorized by the crow in each iteration. mt is 1; The second setting module is used to set the memory time corresponding to each crow in each iteration when it is determined that the food hiding location memorized by the crow is not to be updated. mt = mt +1; Archive update module, used to use the corresponding memory time in each iteration mt =1 to update the corresponding sub-archive and the main archive.

[0109] Optionally, randomly selecting a crow that each crow needs to follow from the sub-archive corresponding to each crow and the crows associated with the main archive, includes: For each crow, when a random value between 0 and 1 is r o >0.5, select the crow that the crow needs to follow from the crows associated with the main archive; For each crow, when a random value between 0 and 1 is r o When ≤0.5, the crow that the crow needs to follow is selected from the crows associated with the sub-archive corresponding to the crow.

[0110] Optionally, the multiple optimization objectives at least include minimizing the threat cost of the drone and minimizing the path length of the drone. When the threat zone is regarded as a cylinder and the drones that can cross the threat zone are regarded as two concentric cylinders with different heights, the expressions of the multiple optimization objectives are as follows:

[0111] in, represents the path length of the UAV to be minimized, represents the minimization of the threat cost of drones, N j Indicates the first j Waypoints, N j+1 Indicates the first j +1 waypoint, ( x j , y j , z j )express N j The three-dimensional coordinates of x j+1 , y j+1 , z j+1 )express N j+1 The three-dimensional coordinates of n Indicates the number of waypoints, K represents the set of threat zones, T k Indicates k The threat value of each threat zone, represents the height of the cylinder outside the threat area, and Respectively represent k The radius of the inner and outer cylinders in the threat zone, d k Indicates flight segment N j N j+1 With k The plane distance between the center points of the threat areas, α and β is the threat coefficient, and 0< α < β <1.

[0112] Optionally, the multiple optimization objectives further include: a violation degree of the crow individual, the violation degree is used to measure the degree to which the crow individual violates a constraint condition, and the constraint condition includes at least one of the following: Range constraints: ; Maximum Threat Constraint: ; Maximum and minimum height constraints: ; in, represents the path length of the UAV, N j Indicates the first j Waypoints, N j+1 Indicates the first j +1 waypoint, Indicates the maximum range of the drone. represents the threat cost of drones, T k Indicates k The threat value of each threat zone, Indicates the maximum threat cost that drones can bear, Indicates j The z-axis coordinate of the waypoint, and They represent the minimum and maximum flight altitudes of the drone, n represents the number of waypoints, K Represents a collection of threat zones.

[0113] It can be seen from the above technical solutions that the present application applies the crow search algorithm to the path planning of the UAV, and adjusts the fitness function according to the currently detected dynamic events in each round of iteration, so that the crow search algorithm can obtain the optimal path planning scheme that meets the current scene, thereby realizing dynamic path planning for the UAV, and the present application also designs a variety of behavior strategies for the crow population to enhance the search optimization performance of the crow search algorithm, thereby effectively improving the real-time and effectiveness of the crow search algorithm in solving the dynamic path planning problem of the UAV, so that the UAV dynamic path planning scheme provided by the present application can quickly respond to the dynamic changes of the mission scene, and timely provide a variety of feasible paths that meet the multiple optimization goals (i.e., a set of target path planning schemes), so as to obtain better path planning effects in a real environment.

[0114] The present application also provides an electronic device, referring to Fig.10 , Fig.10 Schematic diagram of an electronic device according to an embodiment of the present application. Fig.10As shown, the electronic device 100 includes: a memory 110 and a processor 120. The memory 110 and the processor 120 are connected via a bus communication. A computer program is stored in the memory 110. The computer program can be run on the processor 120 to implement the steps in the dynamic path planning method for a drone based on a crow search algorithm disclosed in an embodiment of the present application.

[0115] An embodiment of the present application also provides a computer-readable storage medium on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the method for dynamic path planning of a drone based on a crow search algorithm as disclosed in an embodiment of the present application is implemented.

[0116] The embodiment of the present application also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the unmanned aerial vehicle dynamic path planning method based on the crow search algorithm as disclosed in the embodiment of the present application.

[0117] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0118] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0119] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, systems, devices, storage media, and program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0120] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0121] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0122] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0123] The above is a detailed introduction to a UAV dynamic path planning method based on a crow search algorithm provided by the present application. This article uses specific examples to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for dynamic path planning of unmanned aerial vehicles based on crow search algorithm, characterized in that: The method comprises: Randomly initialize subpopulations for multiple optimization objectives, each of which contains N crows, and the positions of the crows are used to characterize the path planning scheme of the UAV, where N is a positive integer; Iteratively updating the sub-archives corresponding to the plurality of sub-populations based on the crow search algorithm, and in each iteration, adjusting the fitness function according to the currently detected dynamic events, and updating the position of each crow based on the diverse behavior strategies, wherein the sub-archives are used to store non-dominated solutions in the corresponding sub-populations; The main archive is updated according to the sub-archives obtained in each round of iteration, and the target path planning solution set of the UAV is determined according to each non-dominated solution in the main archive obtained in the last update, and the main archive is used to store the non-dominated solutions in all sub-populations; The diverse behavior strategies include: In response to the observed behavior of the crows, randomly selecting the crow that each crow needs to follow from the sub-archives corresponding to each crow and the crows associated with each main archive; According to the attention behavior of crows, different perception probabilities are determined for different crows, and, for a first crow that perceives being followed by other crows, the first crow is set to lead the other crows following it to a random location outside the set radius of the food hiding location it memorizes; In response to the fatigue behavior of crows, a flight length that is negatively correlated with the memory time is determined for each crow, and the memory time is used to represent: the time interval for updating the food hiding position memorized by each crow based on the new position of each crow.

2. The method according to claim 1, characterized in that No. j The probability of a crow sensing Determined by the following formula: in, represents the preset perception probability, represents the first number in the digital sequence generated by the chaotic map Fuch map j Values; In denoting the first crow as j crow and denote the other crows that track the first crow as i In the case of a crow, a random position outside the set radius of the food hiding position memorized by the first crow is determined by the following formula: in, Indicates j crows are iterating τ Lieutenant General i The random location that the crows lead to is d The value of the dimension, and Used as the first i crows are iterating τ +1 position d The value of the dimension; Indicates j crows are iterating τ The first food storage location in the memory d The value of the dimension; and ub d They are d lower and upper bounds for the decision variables; R d Indicates d dimensional radius, and , Represents the radius coefficient.

3. The method according to claim 1, characterized in that The flight length of each of said crows is determined by the following formula: in, Indicates i crows are iterating τ The flight length in Indicates the preset flight length, Indicates i The memory time corresponding to each crow.

4. The method according to claim 1, characterized in that The position of the crow includes: the yaw angle change value, pitch angle and length factor corresponding to each flight segment in the path planning scheme, each flight segment is composed of every two consecutive waypoints from the starting point to the target point in the path planning scheme, and the first waypoint from the starting point to the target point in the path planning scheme is composed of the first waypoint from the starting point to the target point in the path planning scheme. j +1 waypoint N j+1 The three-dimensional coordinates of are determined by the following formula: in,( x j+1 , y j+1 , z j+1 ) indicates the j +1 waypoint N j+1 The three-dimensional coordinates of x j , y j , z j ) indicates the j Waypoints N j The three-dimensional coordinates of θ j Indicates j Waypoints The yaw angle, , α j and Respectively indicated by j Waypoints N j and j +1 waypoint N j+1 The flight segments N j N j+1 The corresponding yaw angle change value, pitch angle and length factor, R Indicates the maximum range of the drone.

5. The method according to claim 1, characterized in that The method further comprises: In each round of iteration, for each crow, if the new position of the crow and the food hiding position it memorizes do not dominate each other, and the fitness of the new position of the crow is worse than that of the food hiding position it memorizes on the corresponding optimization target, the new position of the crow is not updated to the food hiding position it memorizes, but the new position of the crow is added as a non-dominated solution to the main archive and the sub-archives corresponding to the sub-populations other than the sub-population to which the crow belongs; When the amount of data stored in the sub-archive exceeds the maximum capacity of the sub-archive, a truncation technique based on congestion is used to reduce the amount of data stored in the sub-archive to retain the most diverse non-dominated solutions in the sub-archive.

6. The method according to claim 1, characterized in that The method further comprises: If a dynamic event is currently detected, then in a new round of iterations, the memory time of the N crows is mt >1, merge the food hiding location memorized by the second crow into the sub-archive corresponding to the sub-population to which it belongs; After merging the food hiding location memorized by the second crow into the sub-archive corresponding to the sub-population to which it belongs, approximating the target space of the plurality of sub-populations into one dimension, and dividing the target space by a dichotomy method to obtain a dividing line; For non-dominated solutions located in the set area where the dividing line is located, generating a repulsive force response to the dynamic event in the threat area in the artificial potential field environment to adjust the non-dominated solutions; For non-dominated solutions that are not located in the set area where the segmentation line is located, the non-dominated solutions are adjusted by normal perturbation variation.

7. The method according to claim 1, characterized in that The method further comprises: In each iteration, for each crow, when determining to update the food hiding location memorized by the crow, the memory time corresponding to the crow is set. mt is 1; In each iteration, for each crow, when it is determined that the food hiding location memorized by the crow is not to be updated, the memory time corresponding to the crow is set. mt = mt +1; In each iteration, the corresponding memory time is used mt =1 to update the corresponding sub-archive and the main archive.

8. The method according to any one of claims 1 to 7, characterized in that: The step of randomly selecting a crow that each crow needs to follow from the sub-archive corresponding to each crow and the crows associated with each main archive comprises: For each crow, when a random value between 0 and 1 is r o >0.5, select the crow that the crow needs to follow from the crows associated with the main archive; For each crow, when a random value between 0 and 1 is r o When ≤0.5, the crow that the crow needs to follow is selected from the crows associated with the sub-archive corresponding to the crow.

9. The method according to any one of claims 1 to 7, characterized in that: The multiple optimization objectives at least include minimizing the threat cost of the drone and minimizing the path length of the drone. When the threat zone is regarded as a cylinder and the drones that can cross the threat zone are regarded as two concentric cylinders with different heights, the expressions of the multiple optimization objectives are as follows: in, represents the path length of the UAV to be minimized, represents the minimization of the threat cost of drones, N j Indicates the first j Waypoints, N j+1 Indicates the first j +1 waypoint, ( x j , y j , z j )express N j The three-dimensional coordinates of x j+1 , y j+1 , z j+1 )express N j+1 The three-dimensional coordinates of n Indicates the number of waypoints, K represents the set of threat zones, T k Indicates k The threat value of each threat zone, Indicates k The height of the cylinder outside the threat zone, and Respectively represent k The radius of the inner and outer cylinders in the threat zone, d k Indicates flight segment N j N j+1 With k The plane distance between the center points of the threat areas, α and β is the threat coefficient, and 0< α < β <1.

10. The method according to claim 9, characterized in that The multiple optimization objectives also include: the violation degree of the crow individual, the violation degree is used to measure the degree to which the crow individual violates the constraint condition, and the constraint condition includes at least one of the following: Range constraints: ; Maximum Threat Constraint: ; Maximum and minimum height constraints: ; in, represents the path length of the UAV, N j Indicates the first j Waypoints, N j+1 Indicates the first j +1 waypoint, Indicates the maximum range of the drone. represents the threat cost of drones, T k Indicates k The threat value of each threat zone, U Indicates the maximum threat cost that drones can bear, Indicates j The z-axis coordinate of the waypoint, and They represent the minimum and maximum flight altitudes of the drone, n represents the number of waypoints, K Represents a collection of threat zones.

Images