Method for planning bird-repelling cruise route of flapping-wing unmanned aerial vehicle based on WSN thought

Abstract

The invention discloses a wing-flapping unmanned aerial vehicle bird repelling cruise route planning method based on a WSN idea, and belongs to the technical field of unmanned aerial vehicles. In order to solve the problems of low deterrence, short timeliness, small coverage range and the like of a traditional bird repelling means, sensor nodes are generated through an improved particle swarm algorithm, and efficient coverage of a target area is achieved; clustering and partitioning the sensor nodes by adopting an improved k-means algorithm, and dividing a detection area for each unmanned aerial vehicle; and a cruising path is optimized by applying a genetic algorithm, so that the unmanned aerial vehicle is ensured to ensure the biological deterrence effect, the labor cost is reduced, and the inspection efficiency and coverage are improved. The method is suitable for bird damage control in wide areas such as agriculture and forestry orchards, power transmission and distribution lines, high-voltage substations and airports, and has remarkable economic benefits and social benefits.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicles (UAVs), and in particular provides a bird-repelling cruise route planning method for a flapping-wing UAV based on the WSN concept. Background Art

[0002] The problem of bird damage has always been an issue that cannot be ignored in the process of production and development in my country. In agriculture and forestry, problems such as reduced grain production and reduced fruit quality caused by bird damage have seriously restricted the development of the agricultural and forestry economy; at the same time, in the field of power systems, the losses caused by short circuits in transmission and distribution lines caused by bird damage are also increasing year by year; in the field of transportation, bird damage at airports often causes huge economic losses to airlines and seriously threatens the lives of airport staff and passengers.

[0003] An unmanned aerial vehicle (UAV) is an unmanned aircraft controlled by a radio remote control and a self-contained programmable controller, or operated completely or intermittently autonomously by an onboard computer. With the continuous advancement of UAV technology and the development of related industries, UAVs are finding widespread application in various fields, including agricultural and forestry monitoring, geological exploration, power and telecommunications inspections, security monitoring, and medical rescue.

[0004] Traditional bird control methods, such as manual inspections and noise-driven repelling, suffer from shortcomings such as low deterrence, short effectiveness, and limited coverage. In contrast, using drones for bird control significantly reduces labor costs while offering higher inspection efficiency and wider coverage. Furthermore, compared to traditional rotary-wing drones, flapping-wing drones have an eagle-like appearance and a movement more similar to that of natural birds of prey, acting as a biological deterrent to birds and achieving greater bird repellency.

[0005] Using flapping-wing drones for bird control requires them to cruise within the mission area. Their flight paths should be optimized to ensure they can scan every point within the mission area during the mission, monitoring for bird movements. This involves planning routes that ensure full coverage of the entire area.

[0006] The coverage problem in wireless sensor networks (WSNs) is how to ensure that every target area in the network is covered by at least one sensor node for effective monitoring and data collection. This problem aligns perfectly with the goal of bird-repelling patrols. Applying WSN concepts to path planning for bird-repelling patrols by flapping-wing robots, assuming that sensor nodes have already covered the target area as much as possible, the flapping-wing robot only needs to traverse these nodes along a specific path to achieve full coverage, transforming the full-area coverage path planning problem into a traveling salesman problem.

[0007] At the same time, key areas for bird pest control are agricultural, forestry, orchards, power transmission and distribution lines, high-voltage substations, and airports. These areas cover a wide range, and using only a single flapping-wing drone for patrol inspections would take a relatively long time, requiring a high battery capacity. Furthermore, excessive single patrol times would also affect the biodeterrent effect of the drone. Therefore, when using flapping-wing drones for bird control in agricultural, forestry, orchards, power transmission and distribution lines, high-voltage substations, and airports, multiple devices should be deployed for coordinated patrols to improve both patrol efficiency and biodeterrent effectiveness. This involves the issue of coordinated patrols involving multiple robots. Summary of the Invention

[0008] In response to the above problems, the present invention provides a flapping-wing UAV bird-repelling cruise route planning method based on WSN concept, which is used to solve the problems of low coverage efficiency and insufficient optimization of cruise path planning in the existing flapping-wing UAV bird-repelling cruise route planning method.

[0009] To achieve the above object, the technical solution adopted by the present invention is:

[0010] A method for planning a bird-repelling cruise route for a flapping-wing UAV based on a wireless sensor network (WSN) concept includes the following steps:

[0011] a. Use an improved particle swarm algorithm to generate sensor nodes within the target area. Each particle represents the center coordinate of a potential coverage circle. The optimal position of the coverage circle is determined through iterative optimization. Inter-particle influencing variables are introduced to enhance the mutual learning ability of particles.

[0012] b. Use the improved k-means clustering algorithm to cluster and partition the generated sensor nodes. By optimizing the selection of the initial cluster centers, the nodes are divided into several clusters, each of which corresponds to a UAV detection area.

[0013] c. Genetic algorithm is used to generate traversal paths for sensor nodes within each cluster, simulating the natural evolution process. Path length and steering angle are optimized through crossover and mutation operations, taking into account the flight dynamics of the UAV.

[0014] d. Conduct simulation verification, set parameters such as the target area, number of drones, and detection radius, and verify the feasibility and effectiveness of the proposed method through simulation experiments;

[0015] e. Compare and analyze the effectiveness of the proposed method with the traditional ox-plowing method for coverage cruising, and evaluate indicators such as coverage rate, cruising distance, endurance requirements, and single cruising time to determine the applicability of the method in specific mission scenarios.

[0016] The beneficial effect of the above scheme is: an efficient flapping-wing UAV bird-repellent cruise route planning method is realized, which can ensure that the UAV can perform effective bird-repellent operations in the target area, while improving the cruise coverage and biological deterrent effect, and reducing manpower and time costs.

[0017] The detailed working principle of the above scheme is as follows: by combining an improved particle swarm optimization algorithm, a k-means clustering algorithm, and a genetic algorithm, sensor nodes are first generated to determine the optimal location of the coverage circle. These nodes are then clustered and partitioned, and finally, the optimal traversal path is generated for the nodes within each cluster. This process involves the conversion of full-area coverage path planning and the solution of multi-robot collaborative navigation.

[0018] As a further improvement of the above scheme, the particle velocity update formula in the improved particle swarm algorithm is:

[0019]

[0020] in, and Represent the particle velocity at the tth iteration and t+1th iteration respectively; is the inertia weight of the algorithm at the tth iteration; and They represent the learning factors of particles affected by individual optimal positions and group optimal positions respectively; Represents a random number between 0 and 1, :particle In time location; and They are the individual optimal position storage variable and the group optimal position storage variable.

[0021] The beneficial effects of the above improvements are: improving the efficiency and accuracy of the particle swarm algorithm in searching for the global optimal solution, and enhancing the algorithm's ability to search for the global and local optimal solutions by dynamically updating the particle velocity.

[0022] The detailed working principle behind these improvements is that the particle velocity update formula takes into account the particle's current velocity, individual optimal position, group optimal position, and random perturbation terms. This update mechanism enables particles to conduct deeper and more extensive exploration of the solution space.

[0023] As a further improvement of the above scheme, the improved particle swarm algorithm includes inertia weight The update formula is:

[0024]

[0025] in, is the inertia weight of the algorithm at the tth iteration; is the maximum value of the inertia weight, is the minimum value of the inertia weight, is the maximum number of iterations, is the current iteration number.

[0026] The beneficial effect of the above improvements is that by adaptively adjusting the inertia weight, the performance of the particle swarm optimization algorithm between global search and local search is balanced, thereby avoiding the algorithm from converging to a suboptimal solution prematurely.

[0027] The detailed working principle adopted by the above improvement is: the inertia weight decreases with the increase of the number of iterations. This linear decrease method can maintain a large global search capability in the early stage of the algorithm, and gradually enhance the search for the area near the optimal solution in the later stage, that is, the local search capability.

[0028] As a further improvement of the above scheme, the particle swarm algorithm introduces the inter-particle influence variable To enhance the mutual learning ability between particles, the calculation formula is:

[0029]

[0030] in is the historical optimal position of another randomly selected particle; For particles In time location; For the The historical optimal fitness function value of each particle; is the historical optimal fitness function value of the jth particle.

[0031] The beneficial effects of the above improvements are: enhancing the information sharing between particles in the particle swarm optimization algorithm, and improving the learning ability and convergence speed of the entire group by introducing inter-particle influencing variables.

[0032] The detailed working principle adopted by the above improvement is: the inter-particle influence variable takes into account the difference between the particle and its historical optimal position. In this way, each particle considers not only its own experience but also the successful experience of other particles when updating its own position, thereby enhancing the synergistic effect of swarm intelligence.

[0033] As a further improvement of the above scheme, in the sensor node generation step, the coverage rate of the coverage circle The calculation formula is:

[0034]

[0035] in, : the total number of covered circles; :For all covering circles From 1 to The sum of : The point set in the area to be cruised; : Boolean variable, if the point Located in the coverage circle Within the coverage area, is 1; otherwise, it is 0; :For all points in the area to be scanned of Perform the summation; Indicates a point The number of times it is repeatedly covered by multiple covering circles; : The area of ​​the area to be cruised; : The discreteness of the region.

[0036] The beneficial effect of the above improvements is that it provides an accurate coverage calculation method that can quantitatively evaluate the coverage effect of each coverage circle on the target area, providing an important decision-making basis for path planning.

[0037] The detailed working principle adopted by the above improvement is: by calculating the relationship between the coverage circle and each point in the cruise area, and considering the coverage of points in the area that does not need to be detected, the coverage effect of the coverage circle is comprehensively evaluated, and the coverage rate of the entire system is calculated based on this.

[0038] As a further improvement of the above scheme, the improved k-means clustering algorithm includes the evaluation function for cluster center selection The calculation formula is:

[0039]

[0040] in, : the number of clusters; : the total number of sensor nodes; : Sensor node location; It is cluster centers, Representation node To cluster center distance.

[0041] The beneficial effects of the above improvements are: improving the clustering effect of the k-means clustering algorithm, and improving the clustering efficiency and clustering quality by optimizing the selection of cluster centers.

[0042] The detailed working principle adopted by the above improvement is: the selection of initial cluster centers is optimized through an evaluation function based on the sum of the distances from each node to the potential cluster centers, thereby ensuring the representativeness of the cluster centers and the compactness of the clusters.

[0043] As a further improvement of the above scheme, the fitness function of the particle swarm algorithm is The calculation formula is:

[0044]

[0045] in, Indicates a point Is it covering the circle? Within the coverage area, Indicates a point The number of times the circle is covered is reduced by one; : indicates the penalty value; : is the area of ​​the target area to be continued; : The discreteness of the region.

[0046] The beneficial effects of the above improvements are: ensuring that the traversal path generated by the genetic algorithm is both shortest and adapts to the flight characteristics of the UAV, thereby optimizing the cruising efficiency and flight safety of the UAV.

[0047] The detailed working principle adopted by the above improvement is: the fitness function comprehensively considers the path length and the cosine value of the steering angle, guiding the genetic algorithm to ensure that the path is as short as possible while avoiding excessive path curvature, thereby adapting to the flight dynamics characteristics of the drone.

[0048] As a further improvement to the above solution, the UAV detection radius in the simulation verification step is The calculation formula is:

[0049]

[0050] in, is the flight altitude, is the sensor field of view angle.

[0051] The beneficial effects of the above improvements are: providing accurate detection range for the UAV, ensuring effective sensor coverage and bird-repelling effect.

[0052] The detailed working principle adopted by the above improvement is: based on the flight altitude of the drone and the angle of the sensor's field of view, the detection radius of the drone is calculated, thereby determining the actual range that the drone can monitor at the center point of each coverage circle.

[0053] As a further improvement to the above scheme, the genetic algorithm further includes a mechanism for adaptively adjusting the crossover probability and mutation probability to optimize the algorithm's search performance; this mechanism includes:

[0054] a. Define the initial value of the crossover probability as 0.9 and dynamically adjust the crossover probability based on the algorithm's iteration progress and the quality of the solution;

[0055] b. Define the initial value of the mutation probability as 0.05 and dynamically adjust the mutation probability according to the changes in population diversity during the algorithm search process;

[0056] c. Design an adaptive adjustment strategy to appropriately increase the mutation probability to maintain solution diversity when the population shows signs of convergence;

[0057] d. The adaptive adjustment strategy also includes appropriately increasing the crossover probability in the local search phase to promote the generation of excellent solutions.

[0058] The beneficial effects of the above improvements are: by adaptively adjusting the crossover probability and mutation probability, the search performance of the genetic algorithm is optimized, and the efficiency of the algorithm in solving problems and the quality of the solution are improved.

[0059] The detailed working principle adopted by the above improvement is: according to the iteration progress of the algorithm and the diversity of the population, the crossover probability and mutation probability are dynamically adjusted to balance global exploration and local development during the search process, avoid premature convergence, and promote the generation of excellent solutions.

[0060] As a further improvement to the above solution, the simulation verification step further includes the following steps:

[0061] a. Define the target area as a rectangular area of ​​a specific size and divide it into discrete grids. Each grid is used as a discrete point to simulate the UAV's cruising and monitoring.

[0062] b. Set the detection radius of the drone so that the drone can cover a certain range of surrounding grids at each discrete point;

[0063] c. Based on the number of drones and the layout of the target area, each drone is assigned a cruising grid area to achieve collaborative cruising;

[0064] d. Use simulation software to simulate the drone cruising over the grid along the generated traversal path, collect coverage data, and verify the effectiveness of the path planning;

[0065] e. Simulation verification also includes evaluation of key indicators such as the drone's cruise route coverage, cruise distance, and battery life to ensure the practicality and efficiency of the planned path.

[0066] The beneficial effect of the above improvements is that the practicability and effectiveness of the proposed bird-repelling cruise route planning method are ensured through simulation verification, providing a reliable basis for practical application.

[0067] The detailed working principle adopted by the above-mentioned improvement is as follows: simulating the UAV cruising along the planned path in a simulation environment, and evaluating the performance of the path planning through set parameters and collected data, including key indicators such as coverage, cruising distance and battery life, verifying the feasibility and efficiency of the method.

[0068] Beneficial effects of the present invention:

[0069] This paper proposes a WSN-based method for planning bird-repellent patrol routes using flapping-wing drones. This method addresses the path planning problem of multi-robot coordinated patrols for full regional coverage. Using drones for bird repellent control significantly reduces labor costs while achieving higher patrol efficiency and coverage. Furthermore, compared to traditional rotary-wing drones, flapping-wing drones have an eagle-like appearance and a motion more similar to that of natural birds of prey, acting as a bio-deterrent to birds and achieving a more effective bird repellent. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 Flowchart for clustering sensor nodes.

[0071] Figure 2 The resulting graph is the final node.

[0072] Figure 3 This is an intuitive diagram of the clustering results.

[0073] Figure 4 Schematic diagram of the results obtained by genetic algorithm.

[0074] Figure 5 This is the shortest path change curve of path 1 obtained by the genetic algorithm.

[0075] Figure 6 This is the fitness value change curve of path 1 obtained by the genetic algorithm.

[0076] Figure 7 This is the shortest path change curve of path 2 obtained by the genetic algorithm.

[0077] Figure 8 This is the fitness value change curve of path 2 obtained by the genetic algorithm.

[0078] Figure 9 This is the shortest path change curve of path 3 obtained by the genetic algorithm.

[0079] Figure 10 This is the fitness value change curve of path 3 obtained by the genetic algorithm.

[0080] Figure 11 This is the shortest path change curve of path 4 obtained by the genetic algorithm.

[0081] Figure 12 This is the fitness value change curve of path 4 obtained by the genetic algorithm.

[0082] Figure 13 A schematic diagram of the overall path. DETAILED DESCRIPTION

[0083] In order to enable those skilled in the art to better understand the technical solution, the present invention is described in detail below in conjunction with embodiments. The description in this section is only exemplary and explanatory and should not have any limiting effect on the scope of protection of the present invention.

[0084] The present invention provides a full-area coverage path planning method based on the WSN concept, an improved particle swarm algorithm to generate sensor nodes, an improved kmeans algorithm for clustering and partitioning, and a genetic algorithm modeling and coding to determine the node connection order. Specific technical solutions:

[0085] The improved particle swarm algorithm is used to generate sensor nodes in the target area to ensure that the sensing range covers the target area as much as possible.

[0086] The improved kmeans algorithm is used to cluster and partition the sensor nodes, and each zone corresponds to the detection area of ​​a drone.

[0087] The most suitable traversal path is generated for each sensor node in the cluster through genetic algorithm.

[0088] The following technical effects can be achieved:

[0089] The efficiency of bird-repellent patrols and the biological deterrent effect are improved.

[0090] The feasibility and efficiency of the method are demonstrated through simulation.

[0091] Compared with the traditional ox-plowing method, although the coverage rate has decreased, the cruising distance has been greatly shortened, the endurance requirements for the drone are lower, and the single cruising time is shorter.

[0092] The implementation method is:

[0093] Use simulation software to build the target area and set the detection radius of the drone and the discreteness of the area.

[0094] The covering circles are generated by the improved particle swarm optimization algorithm, and clustering is performed by the improved kmeans algorithm.

[0095] The cruise path is generated using genetic algorithm and verified by simulation.

[0096] Specifically, such as Figure 1-13 As shown, the technical solution provided by the present invention is implemented as follows: a method for planning a bird-repelling cruise route for a flapping-wing UAV based on the concept of wireless sensor network (WSN), the method comprising the following steps:

[0097] a. Use an improved particle swarm algorithm to generate sensor nodes within the target area. Each particle represents the center coordinate of a potential coverage circle. The optimal position of the coverage circle is determined through iterative optimization. Inter-particle influencing variables are introduced to enhance the mutual learning ability of particles.

[0098] b. Use the improved k-means clustering algorithm to cluster and partition the generated sensor nodes. By optimizing the selection of the initial cluster centers, the nodes are divided into several clusters, each of which corresponds to a UAV detection area.

[0099] c. Genetic algorithm is used to generate traversal paths for sensor nodes within each cluster, simulating the natural evolution process. Path length and steering angle are optimized through crossover and mutation operations, taking into account the flight dynamics of the UAV.

[0100] d. Conduct simulation verification, set parameters such as the target area, number of drones, and detection radius, and verify the feasibility and effectiveness of the proposed method through simulation experiments;

[0101] e. Compare and analyze the effectiveness of the proposed method with the traditional ox-plowing method for coverage cruising, and evaluate indicators such as coverage rate, cruising distance, endurance requirements, and single cruising time to determine the applicability of the method in specific mission scenarios.

[0102] As a preferred embodiment of the above embodiment, the particle velocity update formula in the improved particle swarm optimization algorithm is:

[0103]

[0104] in, and Represent the particle velocities at the tth iteration and t+1th iteration respectively; is the inertia weight of the algorithm at the tth iteration; and They represent the learning factors of particles affected by individual optimal positions and group optimal positions respectively; Represents a random number between 0 and 1, :particle In time location; and They are the individual optimal position storage variable and the group optimal position storage variable.

[0105] As a preferred embodiment of the above, the improved particle swarm optimization algorithm includes inertia weight The update formula is:

[0106]

[0107] in, : In time The inertia weight of is the maximum value of the inertia weight, is the minimum value of the inertia weight, is the maximum number of iterations, is the current iteration number.

[0108] As a preferred embodiment of the above embodiment, the particle swarm algorithm introduces the inter-particle influence variable The calculation formula is:

[0109]

[0110] in is the historical optimal position of another randomly selected particle; For particles In time location; For the The historical optimal fitness function value of each particle; is the historical optimal fitness function value of the jth particle.

[0111] As a preferred embodiment of the above embodiment, in the sensor node generation step, the coverage rate of the coverage circle The calculation formula is:

[0112]

[0113] in, : the total number of covered circles; : The point set in the area to be cruised; : represents the point set in the target area that does not need to be detected; Indicates a point Is it covering the circle? Within the coverage area, Indicates a point The number of times the circle is covered is reduced by one; : The area of ​​the area to be cruised; : The discreteness of the region.

[0114] As a preferred embodiment of the above, the improved k-means clustering algorithm includes an evaluation function for cluster center selection: The calculation formula is:

[0115]

[0116] in, : the number of clusters; : the total number of sensor nodes; : Sensor node location; It is cluster centers, Representation node To cluster center distance.

[0117] As a preferred embodiment of the above embodiment, the fitness function of the particle swarm algorithm is The calculation formula is:

[0118]

[0119] in, Indicates a point Is it covering the circle? Within the coverage area, Indicates a point The number of times the circle is covered is reduced by one; : indicates the penalty value; : is the area of ​​the target area to be continued; : The discreteness of the region.

[0120] As a preferred embodiment of the above embodiment, the UAV detection radius in the simulation verification step is The calculation formula is:

[0121]

[0122] in, is the flight altitude, is the sensor field of view angle.

[0123] As a preferred embodiment of the above embodiment, the genetic algorithm further includes a mechanism for adaptively adjusting the crossover probability and mutation probability to optimize the search performance of the algorithm; the mechanism includes:

[0124] a. Define the initial value of the crossover probability as 0.9 and dynamically adjust the crossover probability based on the algorithm's iteration progress and the quality of the solution;

[0125] b. Define the initial value of the mutation probability as 0.05 and dynamically adjust the mutation probability according to the changes in population diversity during the algorithm search process;

[0126] c. Design an adaptive adjustment strategy to appropriately increase the mutation probability to maintain solution diversity when the population shows signs of convergence;

[0127] d. The adaptive adjustment strategy also includes appropriately increasing the crossover probability in the local search phase to promote the generation of excellent solutions.

[0128] As a preferred embodiment of the above embodiment, the simulation verification step further includes the following steps:

[0129] a. Define the target area as a rectangular area of ​​a specific size and divide it into discrete grids. Each grid is used as a discrete point to simulate the UAV's cruising and monitoring.

[0130] b. Set the detection radius of the drone so that the drone can cover a certain range of surrounding grids at each discrete point;

[0131] c. Based on the number of drones and the layout of the target area, each drone is assigned a cruising grid area to achieve collaborative cruising;

[0132] d. Use simulation software to simulate the drone cruising over the grid along the generated traversal path, collect coverage data, and verify the effectiveness of the path planning;

[0133] e. Simulation verification also includes evaluation of key indicators such as the drone's cruise route coverage, cruise distance, and battery life to ensure the practicality and efficiency of the planned path.

[0134] The principles of the present invention are as follows:

[0135] 1.1 Improved particle swarm optimization algorithm:

[0136] The particle swarm optimization (PSO) is a swarm intelligence optimization algorithm based on the foraging behavior of bird flocks. Similar to a genetic algorithm, the algorithm initializes a set of solutions and iteratively optimizes their value, ultimately achieving the optimal solution to a set of problems. As a meta-heuristic optimization algorithm, the particle swarm optimization algorithm offers advantages such as fast convergence and ease of adjustment. This paper uses the particle swarm optimization algorithm to determine the optimal position of the coverage circle.

[0137] For the basic particle swarm algorithm, particles adjust their speed and position according to their individual best position and group best position during the iteration process, which is expressed in the following formula:

[0138]

[0139]

[0140] in, and Represent the particle velocities at the tth iteration and t+1th iteration respectively; and They represent the positions of the particle at the tth iteration and the t+1th iteration respectively; w is the inertia weight of the particle motion; and They represent the learning factors of particles affected by individual optimal positions and group optimal positions respectively; Represents a random number between 0 and 1; and They are the individual optimal position storage variable and the group optimal position storage variable.

[0141] The larger the inertia weight w, the stronger the global search capability of the algorithm, and the relatively weaker the local optimization capability. In order to ensure the global search capability in the early stage of the algorithm iteration and the more accurate local optimization capability in the later stage of the iteration, w should decrease according to a certain rule during the iteration process. Therefore, w should be updated according to the following formula:

[0142]

[0143] in, is the inertia weight of the algorithm at the tth iteration; is the maximum value of the inertia weight, is the minimum value of the inertia weight, is the maximum number of iterations, is the current iteration number.

[0144] In view of the problem that the traditional particle swarm algorithm is prone to fall into local optimality, the following improvements can be made: introducing inter-particle influence variables , enhance the mutual learning ability between particles, where The calculation formula is as follows:

[0145]

[0146] in, is the historical optimal position of another randomly selected particle; For the The historical optimal fitness function value of each particle.

[0147] In summary, we can get the improved particle swarm algorithm formula as follows:

[0148]

[0149]

[0150] 1.2 Establishment of coverage model:

[0151] Assume that the area of ​​the target monitoring area is It consists of two parts: the area to be patrolled and the area not to be patrolled. A covering circle, The estimated values ​​are as follows:

[0152]

[0153] in, is the compensation coefficient, which is related to the intersection area of ​​the coverage circle; is the area of ​​the target area to be continued, is the area of ​​the covered circle. , assuming the flying altitude of the flapping-wing UAV is The detection sensor installed has a field of view angle of , the detection radius of the flapping-wing UAV can be obtained , thereby setting the area of ​​the covering circle .

[0154] Establish a rectangular coordinate system and discretize the points in the target area into discrete points, whose degree of discreteness is D=S / mn. Let the point set Represents the set of all discrete points in the target area, point set is the set of centers of the circles covering the target area, point set It is the set of points in the target area that do not need to be detected. is the set of points in the target area to be scanned. , , ', at the same time, for any point in the target area , whose coordinates are .

[0155] In the particle swarm algorithm, each particle represents a set of random solutions to the problem. For the coverage circle generation problem, the position information of each particle is defined as the coordinates of the center of the generated coverage circle. It can be seen that the dimension of the particle is n. The coverage circle needs to cover the area of ​​the target area to be scanned as much as possible. Here, a Boolean variable is introduced , the formula is as follows:

[0156]

[0157] It can be seen that Represents the point Is ∈E' on the covering circle? Within the coverage range. The larger the value of , the more points the covering circle covers in the target area. The total number of points covered by the covering circle is:

[0158]

[0159] At the same time, considering that the two covering circles may cover the same point pi at the same time, another variable gi is introduced into the formula as follows:

[0160]

[0161] At this time, the total number of points covered by the covering circle is:

[0162]

[0163] From this, we can get the coverage rate C of the coverage circle, which is as follows:

[0164]

[0165] But at the same time, in order to avoid the coverage circle covering the area that does not need to be detected, resulting in a waste of observation resources, the coverage circle should be avoided as much as possible on the point pi∈E. Therefore, the variable P is introduced as the penalty value, and its expression formula is as follows, where is the compensation coefficient:

[0166]

[0167] In summary, the fitness function of the particle swarm algorithm can be obtained as follows:

[0168]

[0169] 2. Clustering sensor nodes:

[0170] Assume use If a flapping-wing robot patrols and monitors the target area, the generated coverage circle needs to be divided into m areas. This paper divides the nodes into A large region.

[0171] 2.1 k-means clustering:

[0172] The k-means clustering algorithm is an unsupervised learning algorithm widely used in the fields of data mining and machine learning. Its basic idea is to divide the samples in the data set into k clusters so that the distance between samples in the same cluster is as small as possible, while the distance between samples in different clusters is as large as possible.

[0173] The standard k-means clustering method is as follows: first, randomly select k initial cluster centers, calculate the Euclidean distance of all points to the cluster center, and assign each point to the cluster with the smallest Euclidean distance to it; calculate the average value of all points in each cluster to obtain k new cluster centers, calculate the Euclidean distance of all points to the cluster center again and assign all points to the nearest cluster again, and repeat this step until the selected cluster center no longer changes. The specific process is as follows Figure 1 shown.

[0174] a) Improved k-means clustering method:

[0175] For standard k-means clustering, it is more sensitive to the selection of the initial cluster center, which has a great influence on the convergence results of the subsequent algorithm. Based on this, it is necessary to make some improvements to k-means clustering. By introducing an evaluation function Optimize the selection of initial cluster centers by The expression is as follows:

[0176]

[0177] Where, Used to represent each cluster center in each cluster; Representation node To cluster center The distance is guaranteed by the optimization algorithm. The result with the smallest value is selected as the initial cluster center, which can improve the clustering effect of the k-means algorithm.

[0178] 3. Generation of cruise path:

[0179] 3.1 Genetic Algorithm Modeling:

[0180] A genetic algorithm (GA) is an algorithm inspired by the evolutionary laws of organisms in nature. It searches for optimal solutions by simulating the natural evolutionary process. Using mathematical methods and computer simulations, this algorithm transforms the problem-solving process into a process similar to the crossover and mutation of chromosome genes in biological evolution. GAs offer the advantages of strong global search capabilities, wide applicability, and high flexibility. Based on these advantages, this paper uses a GA to solve the problem of sequentially connecting nodes within each cluster.

[0181] In order to effectively perform genetic operations, the chromosomes in the algorithm must first be encoded. As a chromosome in the algorithm, each node is considered as a gene unit. For example, the sequence = Represents a slave node The connection order of , such an ordered sequence is used as the chromosome in the genetic algorithm for subsequent genetic operations.

[0182] The crossover operator in a genetic algorithm is represented as follows: Given two parents, P1 and P2, two crossover points are randomly selected on the parent. Both parents are retained, and two offspring, F1 and F2, are obtained in the middle segment between these two crossover points. At this point, except for the nodes in these two middle segments, the remaining nodes in the offspring are undetermined. For the undetermined cities in F1, the cities in P2 are first listed in relative order starting from the second crossover point. The cities that appear in the middle segment of F1 are then deleted from this sequence. Finally, the remaining subsequence of P2 is inserted into the corresponding positions in F1 starting from the second crossover point. F2 is obtained in the same manner.

[0183] The mutation operator in the genetic algorithm is represented by randomly selecting two nodes in the ordered sequence and exchanging their positions.

[0184] In addition, considering the flight dynamics of the UAV, its cruising path should be constrained by its steering angle. To ensure the flight safety of the UAV, the curvature of its flight path should be restricted. At the same time, considering the energy consumption of the UAV, the length of the generated path should be as small as possible. Therefore, the fitness function of the genetic algorithm should comprehensively consider the length of the path and the steering problem.

[0185] The fitness function is the quotient of the path length and the cosine of the steering angle as follows:

[0186]

[0187] Based on this, encoding is performed to solve the connection order problem of the covering circle, and then the shortest path length of the cruise cluster area can be obtained.

[0188] Since the standard genetic algorithm is prone to premature convergence and is easily trapped in local convergence problems, the genetic algorithm is improved by introducing multi-population competition to increase population diversity in order to optimize the local convergence problem of the genetic algorithm.

[0189] 4. Simulation verification:

[0190] 4.1 Set up a scene for verification:

[0191] To verify the feasibility of the proposed method, a simulation was performed. Assuming the target area is a 100x100 rectangular area, within which there is an irregular quadrilateral area that does not need to be scanned, four drones were launched from the center of the target area to perform a patrol scan of the entire area. The drones' detection radius was set to 8.92, and the area discreteness was set to 1.

[0192] The improved particle swarm algorithm is used to generate the coverage circle for the above area, where the number of particles is set to 100, the number of iterations is 100, the weight w range is limited to 0.4~0.8, and the learning factor is set to , The uniform value is 1.5, the speed v range is limited to the interval of -2~2, and the compensation coefficient Take 2.5; the final node generation result is as follows Figure 1-2 shown.

[0193] The coverage rate calculated based on the node generation results is 90.76%. It can be seen that the node coverage rate generated by the algorithm is high and meets the basic coverage requirements.

[0194] The number of cluster centers of the improved k-means algorithm is set to 4, and the clustering results are intuitively shown in the figure below. Figure 3 As shown, the points of the same color represent the centers of the circles in the same cluster, and the clustering silhouette coefficient is 0.65107. It can be seen that the clustering effect is relatively ideal.

[0195] In the genetic algorithm, the crossover probability is 0.9, the mutation probability is 0.05, and the number of iterations is 1000. The results are as follows: Figure 4-13 shown.

[0196] 4.2 Comparison with traditional ox-plowing method of covering cruise:

[0197] Compared with the traditional ox-plowing method, the method proposed in this article has a lower area coverage rate, but a significantly shortened cruising distance. This method has lower requirements for the drone's endurance and a shorter single cruising time. It is more suitable for task scenarios such as bird-scaring inspections that require high-frequency drone cruising.

[0198] 5. Summary:

[0199] This paper proposes a full-area coverage path planning method based on the idea of ​​wireless sensor networks. By improving the particle swarm algorithm, sensor nodes with a detection range that covers the target area as much as possible are generated. The nodes are clustered and partitioned using the improved k-means algorithm. Genetic algorithm modeling and coding are used to determine the node connection order. The cruising paths of multiple UAVs are generated and simulated and verified according to specific scenarios. Compared with the traditional ox-plowing inspection method, this method is more suitable for task scenarios such as bird-scaring inspections that require high-frequency cruising. Specific embodiment:

[0201] 1. Implementation environment settings:

[0202] The target monitoring area is set as a rectangular area of ​​100×100 square meters, in which there is an irregular quadrilateral area that does not need to be scanned. 4 drones are sent to conduct a patrol scan of the entire area. The drones fly at an altitude of Meters, sensor field of view angle .

[0203] 2. Application of improved particle swarm optimization algorithm:

[0204] Initialize the particle swarm: Generate 100 particles, each particle represents a potential cruising path in the target area. The position of each particle and speed They are all randomly initialized.

[0205] Determine the parameters:

[0206] Inertia Weight Initial value , the final value .

[0207] Learning Factor .

[0208] The speed range is limited to -2 to 2.

[0209] Iterative process: 100 iterations are performed to optimize the position and velocity of the particles.

[0210] Update inertia weight: At each iteration In the equation, the inertia weight is linearly reduced according to the following formula :

[0211]

[0212] Update particle velocity and position: For each particle , in each iteration , update its velocity using the following formula :

[0213]

[0214] in, is a random number in the interval [0,1], It is a particle The individual optimal position of It is The global optimal position of the iteration.

[0215] Introducing Inter-Particle Influence: Updating Particles Location When the inter-particle influence term is added :

[0216]

[0217] in, is based on the effect of other particles on the particle The vector obtained by calculating the influence of .

[0218] Calculate the influence between particles: For each particle , randomly select another particle , calculated according to the following formula :

[0219]

[0220] in is the historical optimal position of another randomly selected particle; For particles In time location; For the The historical optimal fitness function value of each particle; is the historical optimal fitness function value of the jth particle.

[0221] Fitness evaluation: The position of each particle is evaluated using a fitness function that takes into account the coverage, efficiency, and other relevant factors of the cruise path.

[0222] Update individual and global optimality: If the particle In iteration Position in With better fitness, update its individual optimal and the global optimum .

[0223] Termination condition: When the maximum number of iterations is reached or other termination conditions are met, the algorithm ends and outputs the global optimal solution As the cruising path of the drone.

[0224] Through this improvement, the particle swarm optimization algorithm not only considers the search history and global information of individual particles, but also increases the diversity and adaptability of the search through the mutual influence between particles, which helps to find a better cruising path.

[0225] 3. Inertia weight update:

[0226] In iteration When the inertia weight Calculated as:

[0227]

[0228]

[0229] 4. Calculation of inter-particle influence variables:

[0230] Assume that the historical optimal position of a randomly selected particle is is (5,5), particle Current Location is (3,4), performance index , ,but:

[0231]

[0232] 5. Calculation of coverage circle coverage rate:

[0233] 1). Determine the area to be cruised :

[0234] Total area of ​​target monitoring area square meters.

[0235] Assuming that the area that does not need to be scanned accounts for 4% of the total area, the area of ​​the waiting patrol area is square meters.

[0236] 2). Calculate the area of ​​a single covering circle :

[0237] Detection radius of the drone ,in rice, .

[0238] rice.

[0239] Area of ​​a single covering circle square meters.

[0240] 3). Calculate the total number of required coverage circles :

[0241] To achieve 96% coverage, we set covering circles, so that near .

[0242] .

[0243] 4). Round off the number of covered circles:

[0244] Since the number of covering circles must be an integer, we take A coverage circle was used to ensure that the coverage area exceeded 96%.

[0245] 5). Calculate coverage :

[0246] Assume that each covering circle can be covered independently area, without overlap or omission.

[0247] Coverage The calculation formula is:

[0248]

[0249] because (each point represents 1 square meter), which simplifies to:

[0250]

[0251] Substitution and Value:

[0252]

[0253] 6. Improved application of k-means clustering algorithm:

[0254] Set the number of clusters , total number of sensor nodes , each node To cluster center The average distance , then the evaluation function for:

[0255]

[0256] 7. Application of Genetic Algorithm Genetic Algorithm Parameter Setting:

[0257] Crossover probability: 0.9

[0258] Mutation probability: 0.05

[0259] Iterations: 1000

[0260] Fitness function Calculate, assume , (no repeated coverage), penalty value It is a value determined based on the number of points covering the invalid area or other rules, which is tentatively set here as ,but:

[0261]

[0262] 8. Simulation verification:

[0263] Drone detection radius calculate:

[0264]

[0265] Data tables and technical effect proof

[0266]

[0267] 10. Result analysis:

[0268] Simulations have shown that the proposed method achieves a coverage rate of 96% and a detection radius of 17.32 meters. The simulated cruising range is set at 500 meters, and when the drone cruises at a speed of 5 meters per second, the cruising time is 100 seconds. Compared with traditional methods, the proposed method offers significant advantages in improving coverage efficiency, reducing cruising time, and enhancing drone endurance.

[0269] It should be noted that, in this article, the terms: include, contain and any other variations are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements that are not explicitly listed, or also includes elements that are inherent to such process, method, article or device. Specific examples are used herein to illustrate the principles and implementation methods of the technical solution of the present invention. The above examples are only used to help understand the method of the present invention and its core idea. The above is only a preferred embodiment of the present invention. It should be pointed out that due to the limitations of textual expression, there are objectively infinite specific structures. For ordinary technicians in this technical field, without departing from the principles of the present invention, several improvements, modifications or changes can be made, and the above technical features can be combined in an appropriate manner; these improvements, modifications, changes or combinations, or the direct application of the concept and technical solution of the present invention to other occasions without improvement, should be regarded as the scope of protection of the present invention.

Claims

1. A bird-scaring cruise route planning method for flapping-wing UAV based on WSN concept, characterized by: The method comprises the following steps: a. Use an improved particle swarm algorithm to generate sensor nodes within the target area. Each particle represents the center coordinate of a potential coverage circle. The optimal position of the coverage circle is determined through iterative optimization. Inter-particle influencing variables are introduced to enhance the mutual learning ability of particles. b. Use the improved k-means clustering algorithm to cluster and partition the generated sensor nodes. By optimizing the selection of the initial cluster centers, the nodes are divided into several clusters, each of which corresponds to a UAV detection area. c. Genetic algorithm is used to generate traversal paths for sensor nodes within each cluster, simulating the natural evolution process. Path length and steering angle are optimized through crossover and mutation operations, taking into account the flight dynamics of the UAV. d. Conduct simulation verification, set the target area, number of drones, and detection radius parameters, and verify the feasibility and effectiveness of the proposed method through simulation experiments; e. Compare and analyze the effectiveness of the proposed method with the traditional ox-plowing method for coverage cruising, and evaluate the coverage rate, cruising distance, endurance requirements, and single cruising time indicators to determine the applicability of the method in specific mission scenarios.

2. The method according to claim 1, characterized in that The particle velocity update formula in the improved particle swarm optimization algorithm is: in, and Represent the particle velocity at the tth iteration and t+1th iteration respectively; is the inertia weight of the algorithm at the tth iteration; and They represent the learning factors of particles affected by individual optimal positions and group optimal positions respectively; Represents a random number between 0 and 1, :particle In time location; and They are the individual optimal position storage variable and the group optimal position storage variable.

3. The method according to claim 1, characterized in that Improved particle swarm optimization algorithm including inertia weight The update formula is: in, is the inertia weight of the algorithm at the tth iteration; is the maximum value of the inertia weight, is the minimum value of the inertia weight, is the maximum number of iterations, is the current iteration number.

4. The method according to claim 1, wherein Introducing inter-particle influence variables into particle swarm optimization Enhance the mutual learning ability between particles, The calculation formula is: in is the historical optimal position of another randomly selected particle; For particles In time location; For the The historical optimal fitness function value of each particle; is the historical optimal fitness function value of the jth particle.

5. The method according to claim 1, wherein In the sensor node generation step, the coverage rate of the coverage circle The calculation formula is: in, : the total number of covered circles; :For all covering circles From 1 to The sum of : The point set in the area to be cruised; : Boolean variable, if the point Located in the coverage circle Within the coverage area, is 1; otherwise, it is 0; :For all points in the area to be scanned of Perform the summation; Indicates a point The number of times it is repeatedly covered by multiple covering circles; : The area of ​​the area to be cruised; : The discreteness of the region.

6. The method according to claim 1, wherein Improved k-means clustering algorithm including evaluation function for cluster center selection The calculation formula is: in, : the number of clusters; : the total number of sensor nodes; : Sensor node location; It is cluster centers, Representation node To cluster center distance.

7. The method according to claim 1, characterized in that Fitness function of particle swarm optimization The calculation formula is: in, Indicates a point Is it covering the circle? Within the coverage area, Indicates a point The number of times the circle is covered is reduced by one; : indicates the penalty value; : is the area of ​​the target area to be continued; : The discreteness of the region.

8. The method according to claim 1, characterized in that UAV detection radius in the simulation verification step The calculation formula is: in, is the flight altitude, is the sensor field of view angle.

9. The method according to claim 1, characterized in that The genetic algorithm further includes a mechanism for adaptively adjusting the crossover probability and mutation probability to optimize the algorithm's search performance; this mechanism includes: a. Define the initial value of the crossover probability as 0.9 and dynamically adjust the crossover probability based on the algorithm's iteration progress and the quality of the solution; b. Define the initial value of the mutation probability as 0.05 and dynamically adjust the mutation probability according to the changes in population diversity during the algorithm search process; c. Design an adaptive adjustment strategy to appropriately increase the mutation probability to maintain solution diversity when the population shows signs of convergence; d. The adaptive adjustment strategy also includes appropriately increasing the crossover probability in the local search phase to promote the generation of excellent solutions.

10. The method according to claim 1, characterized in that The simulation verification step further includes the following steps: a. Define the target area as a rectangular area of ​​a specific size and divide it into discrete grids. Each grid is used as a discrete point to simulate the UAV's cruising and monitoring. b. Set the detection radius of the drone so that the drone can cover a certain range of surrounding grids at each discrete point; c. Based on the number of drones and the layout of the target area, each drone is assigned a cruising grid area to achieve collaborative cruising; d. Use simulation software to simulate the drone cruising over the grid along the generated traversal path, collect coverage data, and verify the effectiveness of the path planning; e. Simulation verification also includes evaluation of key indicators such as the drone's cruise route coverage, cruise distance, and battery life to ensure the practicality and efficiency of the planned path.

Images