Unmanned aerial vehicle path planning method based on improved sparrow algorithm

Abstract

The invention provides an unmanned aerial vehicle three-dimensional path planning method based on an improved sparrow algorithm, and the method comprises the steps: carrying out the initialization of a population through fusing Tent chaotic mapping and Sine chaotic mapping, thereby improving the diversity of the population, and improving the search range of the population; secondly, introducing a butterfly optimization algorithm, updating the position of a discoverer in a sparrow search algorithm, and enhancing the search capability of the algorithm for a globally optimal solution; then, position updating is carried out on the participant by using adaptive T distribution, the adaptability and flexibility of the algorithm are increased, K-means clustering is introduced to select the alert, and the path planning capability of the unmanned aerial vehicle in a complex environment is enhanced; finally, in order to prevent the algorithm from falling into local optimum, optimal individual Gaussian disturbance variation is introduced, and the algorithm is prevented from falling into local optimum in the search process. A simulation experiment result shows that compared with a traditional sparrow algorithm, the improved sparrow algorithm can find a safe and feasible path, and has relatively stable optimization capacity.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional path planning for unmanned aerial vehicles, and in particular to a three-dimensional path planning method for unmanned aerial vehicles based on an improved sparrow algorithm. Background Art

[0002] UAV path planning is a key area in the development of modern UAV technology, such as disaster relief, logistics distribution and agricultural monitoring. With the widespread application of UAVs, it is increasingly important to optimize the path from the starting point to the target point to improve efficiency and safety. Common path planning algorithms are divided into classical algorithms and intelligent bionic algorithms. Classical algorithms include A* algorithm, artificial potential field method and fast search random tree algorithm, and intelligent optimization algorithms include gray wolf algorithm, genetic algorithm, particle swarm algorithm, etc. Among them, the sparrow algorithm is an algorithm inspired by the foraging behavior of sparrows. Because of its strong global search ability, high adaptability and easy scalability, it is suitable for UAV path planning, especially in static or semi-dynamic environments. However, it is still easy to fall into the problem of local optimality and early convergence, and the sparrow algorithm needs to be improved. Summary of the invention

[0003] Purpose of the invention: The purpose of the present invention is, firstly, to integrate Tent chaotic mapping and Sine chaotic mapping to initialize the population, so as to improve the diversity and search range of the population; secondly, to introduce the butterfly optimization algorithm and update the position of the discoverer in the sparrow search algorithm, so as to enhance the algorithm's search ability for the global optimal solution; then, to update the position of the joiner using adaptive T distribution, so as to increase the adaptability and flexibility of the algorithm, and to introduce K-means clustering to select the sentinels, so as to enhance the path planning ability of the UAV in a complex environment; finally, in order to prevent the algorithm from falling into a local optimum, the Gaussian perturbation variation of the optimal individual is introduced to prevent it from falling into a local optimum during the search process.

[0004] Technical solution: The present invention provides a three-dimensional path planning method for a UAV based on an improved sparrow algorithm, which specifically includes the following steps:

[0005] (1): Initialize the algorithm parameters, including the sparrow population size (pop), the proportion of discoverers (PD), the proportion of joiners (SD), the maximum number of iterations (maxgen), and boundary constraints.

[0006] (2): Using the two chaotic mapping methods of Tent and Sine, the sparrow population is initialized to generate a more diverse initial population.

[0007] (3): Calculate the fitness value of each sparrow individual and sort them from high to low according to fitness.

[0008] (4): According to the proportion of discoverers, individuals with higher fitness are selected as discoverers and their location information is updated; the remaining individuals are selected as joiners and their locations are updated.

[0009] (5): Randomly select some individuals in the population as sentinels according to a certain ratio and update the sentinel positions.

[0010] (6): Detect whether the algorithm is stuck in a stagnant state. If so, adjust the population position to escape the local optimum.

[0011] (7): When the maximum number of iterations is reached, the iteration stops and the three-dimensional optimal path of the UAV is obtained.

[0012] Furthermore, the implementation process of step (1) is as follows:

[0013] In the process of initializing the parameters of the improved sparrow algorithm path planning method, the corresponding parameters include the sparrow population size (pop), the proportion of discoverers (PD), the proportion of joiners (SD), the maximum number of iterations (maxgen) and the boundary constraints.

[0014] Furthermore, the implementation process of step (2) is as follows:

[0015] The population is initialized by combining Tent chaotic mapping with Sine chaotic mapping. Tent mapping is a simple nonlinear chaotic mapping with good chaotic perturbation characteristics and more uniform probability density. The expression is as follows:

[0016] (1)

[0017] in The sequence of values ​​generated by the mapping, is a parameter (usually set to 0.5).

[0018] Sine mapping is ergodic and sensitive to initial values, so it is often used to generate chaotic sequences. The formula is:

[0019] (2)

[0020] in is the value of the nth iteration, and the initial value is usually randomly selected between (0, 1).

[0021] When the population is initialized, the Tent map and the Sine map are used alternately to generate a new chaotic sequence:

[0022] (3)

[0023] The mapping method to be used is determined based on the index values ​​of the individual and dimension. For odd-numbered index items in the sequence, Tent mapping is used, and for even-numbered index items, Sine mapping is used.

[0024] Furthermore, the implementation process of step (3) is as follows:

[0025] Calculate the fitness value of each individual and sort them from high to low. Select the individual with the best fitness as the discoverer and the remaining individuals as the sentinels.

[0026] Furthermore, the implementation process of step (4) is as follows:

[0027] During the foraging process, sparrows are divided into two roles: producers and foragers. Producers are mainly responsible for finding food resources and have strong exploration capabilities; foragers obtain food from producers and adjust their foraging strategies through imitation and learning. The performance of sparrows is evaluated and sorted according to their energy values. The top 20% are discoverers and the rest are joiners. Discoverers are responsible for global search, and their position updates should increase the diversity of the population to avoid premature convergence and falling into local optimality. Therefore, the butterfly optimization strategy is introduced and applied to the discoverer in the sparrow search algorithm to enhance the global search capability. The butterfly optimization algorithm (BOA) simulates the migration behavior of butterflies and balances global search and local search through the position update strategies of leaders and followers. The butterfly optimization strategy formula is as follows:

[0028] (4)

[0029] in, Time step length, t is the attenuation factor and t is the number of iterations.

[0030] The fundamental update formula of the detector position after fusion improvement is:

[0031] (5)

[0032] If the random number R is less than the threshold ST, then To update the location, the new location is updated by sending the current best location Add a Zoom from current position Point to the best position vector, and this vector will decay exponentially with time t; if the random number R is greater than the threshold ST, then To update the location, the new location is obtained by Update by adding a random length vector L scaled by Q.

[0033] In the sparrow search algorithm, the follower is mainly responsible for local exploration around the producer, thereby increasing the diversity and exploration depth of the population. However, it is easy to fall into local optimality and lacks global exploration ability. Therefore, the adaptive T distribution is introduced to update the position of the follower. The adaptive T distribution can dynamically adjust the sampling distribution during the search process, so that the individual gradually tends to the optimal solution. The adaptive T distribution gradually narrows the search range by controlling the degrees of freedom or standard deviation of the T distribution, thereby improving the exploration ability of the algorithm and enhancing the local search ability of the algorithm to avoid falling into the local optimal. Utilizing the sampling characteristics of the adaptive T distribution, the standard deviation is gradually reduced to control the movement range of the individual in the search space. As the number of iterations increases, the value of the standard deviation will gradually decrease, which will shrink the search range of the later individuals and help converge to the optimal solution. The update formula of the adaptive T distribution is as follows:

[0034] (6)

[0035] in, Make the position of the optimal solution in the current population, is a scaling factor that decreases with the number of iterations and is used to control the search range in each generation. randn is a standard normal distribution random number generated to simulate the sampling effect of the T distribution. is defined as:

[0036] (7)

[0037] Where M is the total number of iterations and t is the current number of iterations. As the number of iterations increases, The value of decreases from large to close to 0, so that the algorithm has a larger search range in the early stage, and in the later convergence process, the exploration range is narrowed, and the optimal solution can be found more accurately.

[0038] Furthermore, the implementation process of step (5) is as follows:

[0039] The sentinels in the sparrow algorithm are responsible for detecting individuals in the population and moving to new locations when the environment changes or a danger signal appears, thereby maintaining the vitality and diversity of the population. The sentinels in the sparrow algorithm have shortcomings such as a single update strategy, poor environmental adaptability, and strong randomness. Therefore, K-means clustering is selected to improve its search efficiency. K-means clustering is a commonly used clustering algorithm. , where each data point x i There are dim features. The goal of the algorithm is to minimize the sum of the squared distances of all data points to their respective cluster centers, that is,

[0040] (8)

[0041] in, is the center of the kth cluster, is the set of data points belonging to the kth cluster, represents the Euclidean distance.

[0042] K clusters are obtained through K-means clustering, and a representative point is selected in each cluster as a scout to further explore the population.

[0043] Furthermore, the implementation process of step (6) is as follows:

[0044] The Gaussian perturbation of the optimal individual is used to prevent the algorithm from converging prematurely into the local optimum during the optimization process, thereby improving the diversity of the search and the global search capability. When the algorithm stagnates, the current global optimal solution x best Apply Gaussian perturbation, and the optimal solution update formula after perturbation is as follows:

[0045] (9)

[0046] in represents a random vector generated by a standard normal distribution with a mean of 0 and a variance of 1. is the perturbation intensity. Improve the global exploration ability of the algorithm.

[0047] Furthermore, the implementation process of step (7) is as follows:

[0048] When the maximum number of iterations is reached, the iteration stops and the optimal three-dimensional path of the UAV is obtained.

[0049] Beneficial effects: Compared with the prior art, the present invention has the following beneficial effects: The present invention integrates the Tent chaotic map and the Sine chaotic map in the traditional sparrow algorithm to initialize the population, thereby improving the diversity and search range of the population; secondly, the butterfly optimization algorithm is introduced and the position of the discoverer in the sparrow search algorithm is updated to enhance the algorithm's ability to search for the global optimal solution; then, the position of the joiner is updated using the adaptive T distribution to increase the adaptability and flexibility of the algorithm, and K-means clustering is introduced to select the sentinel to enhance the path planning ability of the UAV in a complex environment; finally, in order to prevent the algorithm from falling into the local optimum, the optimal individual Gaussian perturbation variation is introduced to prevent it from falling into the local optimum during the search process. A three-dimensional path planning method for UAVs based on the improved sparrow algorithm is provided; it is beneficial to reduce the possibility of local optimum, improve the efficiency of search, shorten the search path, and thus ensure that the UAV obtains the optimal path in a shorter time. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Flowchart for improving the sparrow algorithm;

[0051] Figure 2 The optimal path for improving the sparrow algorithm in the present invention;

[0052] Figure 3 The best individual fitness change trend of the improved sparrow algorithm in the present invention;

[0053] Figure 4 It is the optimal path of the traditional sparrow algorithm;

[0054] Figure 5 The changing trend of the best individual fitness of the traditional sparrow algorithm; DETAILED DESCRIPTION

[0055] The present invention is further described in detail below with reference to the accompanying drawings. The present invention provides a three-dimensional path planning method for a UAV based on an improved sparrow algorithm, such as Figure 1 As shown, the following steps are included:

[0056] Step 1: Initialize the algorithm parameters, including the sparrow population size (pop), the proportion of discoverers (PD), the proportion of joiners (SD), the maximum number of iterations (maxgen), and boundary constraints.

[0057] Step 2: Initialize the population. Tent mapping is a simple nonlinear chaotic mapping with good chaotic perturbation characteristics and more uniform probability density. The expression is as follows:

[0058] (10)

[0059] in The sequence of values ​​generated by the mapping, is a parameter (usually set to 0.5).

[0060] Sine mapping is ergodic and sensitive to initial values, so it is often used to generate chaotic sequences. The formula is:

[0061] (11)

[0062] in is the value of the nth iteration, and the initial value is usually randomly selected between (0, 1).

[0063] When the population is initialized, the Tent map and the Sine map are used alternately to generate a new chaotic sequence:

[0064] (12)

[0065] The mapping method to be used is determined based on the index values ​​of the individual and dimension. For odd-numbered index items in the sequence, Tent mapping is used, and for even-numbered index items, Sine mapping is used.

[0066] Step 3: Calculate the fitness value of each sparrow individual and sort them from high to low according to fitness.

[0067] Step 4: According to the proportion of discoverers, select individuals with higher fitness as discoverers and update their location information; the remaining individuals are joiners and their locations are updated. The discoverers are responsible for global search, and their location updates should increase the diversity of the population to avoid premature convergence and fall into local optimality, so the butterfly optimization strategy is introduced. , and apply it to the finder in the sparrow search algorithm to enhance the global search capability. The butterfly optimization algorithm (BOA) simulates the migration behavior of butterflies and balances global search and local search through the position update strategy of leaders and followers. The butterfly optimization strategy formula is as follows:

[0068] (13)

[0069] in, Time step length, t is the attenuation factor and t is the number of iterations.

[0070] The fundamental update formula of the detector position after fusion improvement is:

[0071] (14)

[0072] If the random number R is less than the threshold ST, then To update the location, the new location is updated by sending the current best location Add a Zoom from current position Point to the best position vector, and this vector will decay exponentially with time t; if the random number R is greater than the threshold ST, then To update the location, the new location is obtained by Update by adding a random length vector L scaled by Q.

[0073] In the sparrow search algorithm, the follower is mainly responsible for local exploration around the producer. The adaptive T distribution can dynamically adjust the sampling distribution during the search process, so that the individual gradually tends to the optimal solution. The adaptive T distribution controls the degrees of freedom or standard deviation of the T distribution to gradually narrow the search range, thereby improving the algorithm's exploration ability and enhancing the algorithm's local search ability to avoid falling into the local optimum. By utilizing the sampling characteristics of the adaptive T distribution, the standard deviation is gradually reduced to control the movement range of the individual in the search space. As the number of iterations increases, the value of the standard deviation will gradually decrease, shrinking the search range of the later individuals, which is conducive to convergence to the optimal solution. The update formula of the adaptive T distribution is as follows:

[0074] (15)

[0075] in, Make the position of the optimal solution in the current population, is a scaling factor that decreases with the number of iterations and is used to control the search range in each generation. randn is a standard normal distribution random number generated to simulate the sampling effect of the T distribution. is defined as:

[0076] (16)

[0077] Where M is the total number of iterations and t is the current number of iterations. As the number of iterations increases, The value of decreases from large to close to 0, so that the algorithm has a larger search range in the early stage, and in the later convergence process, the exploration range is narrowed, and the optimal solution can be found more accurately.

[0078] Step 5: Select 10%-20% of the individuals in the sparrow algorithm as sentinels. The sentinels in the sparrow algorithm are responsible for detecting individuals in the population and moving to new locations when the environment changes or danger signals appear, thereby maintaining the vitality and diversity of the population. The sentinels in the sparrow algorithm have shortcomings such as a single update strategy, poor environmental adaptability, and strong randomness. Therefore, K-means clustering is selected to improve its search efficiency. K-means clustering is a commonly used clustering algorithm. , where each data point x i There are dim features. The goal of the algorithm is to minimize the sum of the squared distances of all data points to their respective cluster centers, that is,

[0079] (17)

[0080] in, is the center of the kth cluster, is the set of data points belonging to the kth cluster, represents the Euclidean distance.

[0081] K clusters are obtained through K-means clustering, and a representative point is selected in each cluster as a scout to further explore the population.

[0082] Step 6: The traditional sparrow algorithm is prone to fall into the local optimum, so the Gaussian perturbation of the optimal individual is used to prevent the algorithm from converging too early and falling into the local optimum during the optimization process, thereby improving the diversity of the search and the global search capability. When the algorithm stagnates, the current global optimal solution x best Apply Gaussian perturbation, and the optimal solution update formula after perturbation is as follows:

[0083] (18)

[0084] in represents a random vector generated by a standard normal distribution with a mean of 0 and a variance of 1. is the perturbation intensity. Improve the global exploration ability of the algorithm.

[0085] Step 7: When the maximum number of iterations is reached, the iteration stops and the three-dimensional optimal path of the UAV is obtained.

[0086] To verify the correctness and effectiveness of the present invention, this paper simulated and compared the traditional sparrow algorithm and the improved sparrow algorithm using Matlab 2023b simulation software under the Window11 operating system of AMD Ryzen 9 7945HX with RadeonGraphics. The algorithm parameters are set as follows: the proportion of discoverers is set to 20%, the remaining sparrow individuals are followers, the number of vigilant individuals is 10%, the population size is 200, and the number of iterations is 500.

[0087] The optimal path of the simulation results of the improved sparrow algorithm and the traditional sparrow algorithm proposed in the present invention in the three-dimensional environment model is as follows: Figure 2 and Figure 4 As shown in Figure 3 and Figure 5 As shown. Analysis and comparison of simulation results show that the improved sparrow algorithm outperforms the traditional sparrow algorithm in terms of path cost, threat cost and height cost, and has fewer path turns, better path smoothness and faster convergence speed. Therefore, the path obtained by the improved sparrow algorithm is closer to the actual path and meets the actual needs of path planning of drones in a three-dimensional environment. The planning results of the traditional sparrow algorithm and the improved sparrow algorithm are compared in the form of a table, as shown in Table 1:

[0088] Table 1 Comparison between the present invention and the traditional sparrow algorithm

[0089] algorithm Path cost Threat Cost High cost Corner cost Optimal fitness value Traditional Mahjong Algorithm 1271.33 5.4303 135.789 97.8064 7817.79 Improved sparrow algorithm (this invention) 1204.47 20.4267 45.2478 0 6495.24

[0090] Through comparative simulation, it can be concluded that the improved sparrow algorithm proposed in the present invention can effectively improve the convergence speed of the algorithm, and the optimization effect is better than that of the traditional sparrow algorithm.

Claims

1. A three-dimensional path planning method for unmanned aerial vehicles based on an improved sparrow algorithm, characterized in that: The following steps are involved: (1): Initialize algorithm parameters, including the sparrow population size (pop), discoverer ratio (PD), joiner ratio (SD), maximum number of iterations (maxgen), and boundary constraints; (2): Using the Tent and Sine chaotic mapping methods, the sparrow population is initialized to generate a more diverse initial population; (3): Calculate the fitness value of each sparrow individual and sort them from high to low according to fitness; (4): According to the proportion of discoverers, select individuals with higher fitness as discoverers and update their location information; The remaining individuals act as joiners and update their positions; (5): Randomly select some individuals in the population as sentinels according to a certain ratio and update the sentinel positions; (6): Detect whether the algorithm is stuck in a stagnant state. If so, adjust the population position to escape from the local optimum; (7): When the maximum number of iterations is reached, the iteration stops and the three-dimensional optimal path of the UAV is obtained.

2. The method for three-dimensional path planning of an unmanned aerial vehicle based on an improved sparrow algorithm according to claim 1, characterized in that: The implementation process of step (1) is as follows: In the process of initializing the parameters of the improved sparrow algorithm path planning method, the corresponding parameters include the sparrow population size (pop), the proportion of discoverers (PD), the proportion of joiners (SD), the maximum number of iterations (maxgen) and the boundary constraints.

3. The method for three-dimensional path planning of an unmanned aerial vehicle based on an improved sparrow algorithm according to claim 1, characterized in that: The implementation process of step (2) is as follows: The population is initialized by combining Tent chaotic mapping with Sine chaotic mapping. Tent mapping is a simple nonlinear chaotic mapping with good chaotic perturbation characteristics and more uniform probability density. The expression is as follows: （1） in The sequence of values ​​generated by the mapping, is a parameter (usually set to 0.5); Sine mapping is ergodic and sensitive to initial values, so it is often used to generate chaotic sequences. The formula is: （2） in is the value of the nth iteration, and the initial value is usually randomly selected between (0, 1); When the population is initialized, the Tent map and the Sine map are used alternately to generate a new chaotic sequence: （3） The mapping method to be used is determined based on the index values ​​of the individual and dimension. For odd-numbered index items in the sequence, Tent mapping is used, and for even-numbered index items, Sine mapping is used.

4. The method for three-dimensional path planning of an unmanned aerial vehicle based on an improved sparrow algorithm according to claim 1, characterized in that: The implementation process of step (3) is as follows: Calculate the fitness value of each individual and sort them from high to low. Select the individual with the best fitness as the discoverer and the remaining individuals as the sentinels.

5. According to the method for three-dimensional path planning of unmanned aerial vehicle based on improved sparrow algorithm as claimed in claim 1, the implementation process of step (4) is as follows: During the foraging process, sparrows are divided into two roles: producers and foragers. Producers are mainly responsible for finding food resources and have strong exploration capabilities; foragers obtain food from producers and adjust their foraging strategies through imitation and learning. The performance of sparrows is evaluated and sorted according to their energy values. The top 20% are discoverers and the rest are joiners. Discoverers are responsible for global search, and their position updates should increase the diversity of the population to avoid premature convergence and falling into local optimality. Therefore, the butterfly optimization strategy is introduced and applied to the discoverer in the sparrow search algorithm to enhance the global search capability. The butterfly optimization algorithm (BOA) simulates the migration behavior of butterflies and balances global search and local search through the position update strategies of leaders and followers. The butterfly optimization strategy formula is as follows: （4） in, Time step length, Attenuation factor at time t, number of iterations at time t; The fundamental update formula of the detector position after fusion improvement is: （5） If the random number R is less than the threshold ST, then To update the location, the new location is updated by sending the current best location Add a Zoom from current position Point to the best position vector, and this vector will decay exponentially with time t; if the random number R is greater than the threshold ST, then To update the location, the new location is obtained by Add a random length vector L scaled by Q to update; In the sparrow search algorithm, followers are mainly responsible for local exploration around producers, thereby increasing the diversity and exploration depth of the population. However, they are prone to falling into local optimality and insufficient global exploration capabilities. Therefore, adaptive T distribution is introduced to update the position of followers. Adaptive T distribution can dynamically adjust the sampling distribution during the search process, so that individuals gradually tend to the optimal solution. Adaptive T distribution controls the degrees of freedom or standard deviation of T distribution to gradually narrow the search range, thereby improving the algorithm's exploration ability and enhancing the algorithm's local search ability to avoid falling into the local optimum. By utilizing the sampling characteristics of adaptive T distribution, the standard deviation is gradually reduced to control the movement range of individuals in the search space. As the number of iterations increases, the value of the standard deviation will gradually decrease, shrinking the search range of individuals in the later stage, which is conducive to convergence to the optimal solution. The update formula of adaptive T distribution is as follows: （6） in, Make the position of the optimal solution in the current population, is a scaling factor that decreases with the number of iterations and is used to control the search range in each generation. randn is a standard normal distribution random number generated to simulate the sampling effect of the T distribution. is defined as: （7） Where M is the total number of iterations and t is the current number of iterations. As the number of iterations increases, The value of decreases from large to close to 0, so that the algorithm has a larger search range in the early stage, and in the later convergence process, the exploration range is narrowed, and the optimal solution can be found more accurately.

6. The method for three-dimensional path planning of an unmanned aerial vehicle based on an improved sparrow algorithm according to claim 1, characterized in that: The implementation process of step (5) is as follows: The sentinels in the sparrow algorithm are responsible for detecting individuals in the population and moving to new locations when the environment changes or a danger signal appears, thereby maintaining the vitality and diversity of the population. The sentinels in the sparrow algorithm have shortcomings such as a single update strategy, poor environmental adaptability, and strong randomness. Therefore, K-means clustering is selected to improve its search efficiency. K-means clustering is a commonly used clustering algorithm. , where each data point x i There are dim features. The goal of the algorithm is to minimize the sum of the squared distances of all data points to their respective cluster centers, that is, （8） in, is the center of the kth cluster, is the set of data points belonging to the kth cluster, represents the Euclidean distance; K clusters are obtained through K-means clustering, and a representative point is selected in each cluster as a scout to further explore the population.

7. The method for three-dimensional path planning of an unmanned aerial vehicle based on an improved sparrow algorithm according to claim 1, characterized in that: The implementation process of step (6) is as follows: The Gaussian perturbation of the optimal individual is used to prevent the algorithm from converging prematurely into the local optimum during the optimization process, thereby improving the diversity of the search and the global search capability. When the algorithm stagnates, the current global optimal solution x best Apply Gaussian perturbation, and the optimal solution update formula after perturbation is as follows: （9） in represents a random vector generated by a standard normal distribution with a mean of 0 and a variance of 1. is the disturbance intensity. Improve the global exploration capability of the algorithm.

8. The method for three-dimensional path planning of an unmanned aerial vehicle based on an improved sparrow algorithm according to claim 1, characterized in that: The implementation process of step (7) is as follows: when the maximum number of iterations is reached, the iteration stops, and the three-dimensional optimal path of the UAV is obtained.

Images