Unmanned aerial vehicle adjustable planning method for small area coverage

Abstract

The invention provides an unmanned aerial vehicle adjustable planning method for small area coverage, and belongs to the technical field of unmanned aerial vehicle command and control. According to the method, firstly, an unmanned aerial vehicle and coverage area model is established according to coverage requirements and unmanned aerial vehicle performance parameters, and a foundation is laid for subsequent planning; then, constraint conditions of coverage area planning are represented, and it is ensured that planning meets actual application requirements; thirdly, on the basis of meeting the constraint condition, carrying out adjustable region clustering by utilizing a clustering algorithm, dividing the region into a plurality of subsets, and reasonably allocating tasks for the unmanned aerial vehicle; and finally, optimizing an area access sequence of task execution of the unmanned aerial vehicle by using an ant colony optimization algorithm so as to improve planning efficiency and reduce time consumption of task execution of the unmanned aerial vehicle. According to the method, the path planning efficiency and the task execution speed of the multi-unmanned aerial vehicle system in the multi-area coverage task are improved, the time consumption of task execution of the unmanned aerial vehicles is reduced, the area access sequence of the unmanned aerial vehicles is optimized, and an efficient and reliable solution is provided for practical application.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle command and control, and in particular to an adjustable planning method for unmanned aerial vehicles covering small areas. Background Art

[0002] In practical applications, multi-UAV collaboration can significantly improve flexibility and adaptability. However, in regional information systems, they make the system face greater challenges in performing tasks such as path planning, collaborative control, and intelligent decision-making. It is particularly difficult to consider multiple regions for path planning while meeting many constraints, because multivariable optimization problems have NP-Hard computational complexity and it is difficult to find the optimal solution.

[0003] Although many scholars have conducted relevant research on the problem of optimal adjustable planning of UAVs under conditions such as regional coverage and energy constraints, and proposed a variety of solutions, such as deriving energy consumption models from practical applications, dynamically adjusting the speed and power consumption of UAVs to reduce system energy consumption; decomposing path planning constraints, establishing objective functions and evaluation functions, and solving them using evaluation and evolutionary methods; and methods based on extended Kalman filtering to determine the adjustable planning process of UAVs by selecting waypoints that can improve the accuracy of target position estimation. However, current research focuses on the optimization and adjustment of UAVs in a single area. For example, according to the flight and energy capabilities of UAVs, the entire area is divided into multiple small areas and a round-trip planning path is generated to cover the sub-areas; based on collaborative search to enhance the perception capability of UAV systems, adjustable dynamic planning is performed to avoid the impact of unexpected failures.

[0004] However, these studies have limitations: On the one hand, there are relatively few studies on adjustable planning of drones for multi-area coverage, and existing results are only applicable to scenarios with a single area of ​​interest, making it difficult to directly apply them to the adjustment planning problem of multiple drone systems accessing multiple areas.

[0005] On the other hand, although the clustering-based area coverage adjustable planning method is highly efficient, the lack of optimization of the area visit order of the same UAV leads to unsatisfactory planning results in some cases.

[0006] In general, although there are currently a variety of technical means to consider unmanned system collaboration to achieve planning adjustments for coverage areas, research on the optimization of drone area access sequences in multi-region environments is still insufficient, and existing adjustable planning methods have some defects. Summary of the invention

[0007] The purpose of the present invention is to provide an adjustable planning method for UAVs for small area coverage, to improve the path planning efficiency and task execution speed of a multi-UAV system in a multi-area coverage task, to reduce the time consumed by UAVs in executing tasks, to optimize the area access sequence of UAVs, and to provide an efficient and reliable solution for practical applications.

[0008] To achieve the above object, the present invention provides an adjustable planning method for UAVs covering small areas, comprising the following steps: Step S1, establishing a UAV and coverage area model according to requirements such as mission time consumption in the coverage area, regional spatial distance, and performance parameters such as flight speed and altitude of the UAV; Step S2, characterizing the constraints of coverage area planning; Step S3: On the basis of satisfying the constraint conditions, a clustering algorithm is used to perform adjustable region clustering, the region is divided into multiple subsets, and tasks are assigned to the drones; Step S4: Utilize the ant colony optimization algorithm to optimize the area visit sequence of the UAV to perform the task, so as to improve the planning efficiency and reduce the time consumption of the UAV to perform the task.

[0009] Preferably, in step S1, a drone model is established, and the specific operations are as follows: Each drone is modeled as ,in, and Respectively represent The maximum flight speed and maximum flight altitude of each drone, Indicates the width of the drone's search range; When a drone is used to scan an area, its speed is limited to: ; in, Indicates A drone is used to scan Speed ​​limit in each area, express A collection of , Respectively represent the number of drones and the number of scanned areas; When a drone exist If the mission cannot be performed in the area, ; Calculate the time it takes for the drone to perform tasks in the area: ; in, Indicates drone exist The time it takes to perform tasks in the area; express The scan range of the area.

[0010] Preferably, in step S1, a coverage area model is established, and the specific operations are as follows: Calculate the Regions and Flight distance between regions , thus obtaining The spatial distance between any two regions in the region .

[0011] Preferably, in step S2, the constraint conditions characterizing the coverage area planning include: C1 constraint: When the drone is not selected to perform area coverage missions, it should stay at the base; C2 constraint: Each drone must take off from the base when performing a mission and return to the base after the mission is completed; C3 constraint: Each area can only be covered by one drone; C4 constraint: The drone should visit and cover all assigned areas one by one; C5 Constraint: The drone must return to base before running out of energy.

[0012] Preferably, in step S3, a clustering algorithm is used to cluster the adjustable regions, the regions are divided into multiple subsets, and tasks are assigned to the drones. The specific operations are as follows: Step S31, determining the cluster center according to the density value and distance value of each area; Step S32: After all cluster centers are selected, each remaining region is classified into the same cluster as its nearest neighbor; use To describe the various clusters of drone swarms, Indicates that it is classified into clusters For each cluster , whose center is the first region added to the cluster, namely: ; in, Representation clustering the center of; Step S33: From the currently available clusters, according to the time requirements of each cluster, select the cluster with the smallest time requirement as the cluster to be expanded first; ; in, represents the cluster with the smallest time requirement; Represents the currently available clustering set; Indicates coverage of a given cluster The time consumption requirements of each area; Indicates that it is classified into clusters The regional sequence of Indicates that it is classified into clusters No. Regions; Step S34: In the unallocated area, select a suitable area according to the relative distance to the allocated area and the C5 constraint condition. ; Step S35: Set the appropriate area Add to selected cluster In the sequence, according to its distance from the cluster head and tail, it is decided to insert it before the cluster head or after the tail, thereby updating the cluster sequence structure.

[0013] Preferably, in step S4, the ant colony optimization algorithm is used to optimize the area access sequence of the drone to perform the task, so as to improve the planning efficiency and reduce the time consumption of the drone to perform the task. The specific operations are as follows: Step S41, initialization phase, setting the initial pheromone value between any two areas and initializing the heuristic information according to the area distance; Step S42, local update rule, in the process of constructing the ant solution, local update of pheromone is performed after each movement; Step S43, global update rule: after all ants complete their journey, only the ants that have found the shortest path are allowed to update the pheromone, thus enhancing the pheromone on the global optimal path.

[0014] Therefore, the present invention adopts the above-mentioned adjustable planning method for drones covering small areas, and the beneficial technical effects are as follows: (1) Cluster-based small-area adjustment planning: Aiming at the coverage requirements of multiple small areas, the planning problem is transformed into a clustering solution and allocation process, which overcomes the high computational complexity of the NP-hard problem and can effectively improve the efficiency of UAV adjustment planning.

[0015] (2) Performance advantages: The present invention takes into account the optimization problem of the UAV area access sequence, overcomes the problem of unsatisfactory UAV planning effect when the area sequence is not optimized, and can effectively reduce the time consumed by the UAV to perform tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The present invention is a flow chart of an adjustable planning method for unmanned aerial vehicles covering small areas. DETAILED DESCRIPTION

[0017] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0018] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0019] Embodiment 1 like Figure 1 FIG. 1 is a flowchart of an adjustable planning method for a UAV for small area coverage according to the present invention, which specifically includes the following steps: 1. Establishment of drone and coverage area model.

[0020] 101. Drone modeling.

[0021] In this embodiment, a group of drones is required To execute Search missions in the area Each drone is modeled as ,in, and Respectively represent The maximum flight speed and maximum flight altitude of each drone, Represents the width of the drone's search range.

[0022] When drones are used to scan an area, their speed limits are: ; If the drone is not selected for scanning or runs out of energy, the drone exist If the mission cannot be performed in the area, In addition, the use of Indicates drone exist The time taken to execute a task in a region is calculated as follows: ; in, express The scan range of the area.

[0023] 102. Modeling of coverage area.

[0024] Assumptions Indicates Regions and The flight distance between the regions is calculated by the coordinates between them. ,express The spatial distance between any two areas in a region.

[0025] 2. Representation of constraints for coverage area planning.

[0026] When all areas of interest should be correctly covered by the drone, the following path constraints and energy constraints should be satisfied, and then a constrained motion model is obtained based on the drone modeling process in step 1.

[0027] C1 Constraint: Each drone should remain at base unless selected for an area coverage mission.

[0028] C2 Constraint: Once a drone is selected to scan an area, it should take off from the base and fly back to the base after completing the regional mission.

[0029] C3 constraint: To ensure efficient completion of tasks, it is assumed that each area can only be covered by one drone, and the drone cluster should not repeat tasks in a certain area or miss tasks in a certain area.

[0030] C4 constraint: Each drone involved in the coverage mission should visit and cover the area one by one.

[0031] C5 Constraint: Each drone should land at the base before running out of energy.

[0032] 3. Adjustable regional clustering stage.

[0033] 301. Select cluster centers.

[0034] The selection of cluster centers is a key step in the adjustable region clustering stage. In the present invention, if a region has a higher density than its neighboring regions, or is far away from other higher density regions, it is determined as the cluster center. In order to better calculate the cluster center of the drone, it is necessary to calculate each region. Two parameters: density value And the distance value Therefore, the cluster center can be defined as having both and In order to quickly find the cluster center of the drone, two quantitative parameters are introduced. and Consider the density value And the distance value The combined effect of Represents the sorted region index and follows this allocation strategy. It's a drone The cluster center of .

[0035] 302. Clustering starts at the cluster center.

[0036] When all cluster centers are selected, each remaining region is classified into the same cluster as its nearest neighbor. To describe the various clusters of drone swarms. Indicates that it is classified into clusters The regional sequence of Indicates that it is classified into clusters No. area.

[0037] For each cluster , whose center is the first region added to the cluster, namely: ; in, Representation clustering the center of; At this point, all clusters are available and all regions except the cluster centers are unassigned.

[0038] 303. Cluster selection.

[0039] In order to balance the time requirements of the clusters, the cluster with the smallest time requirement is selected from the available clusters, and the regions are added to this cluster first. The calculation process is: ; in, Represents the currently available clustering set, Indicates coverage of a given cluster The time consumption requirements of each area.

[0040] 304. Select suitable area.

[0041] Unallocated areas are clustered according to relative distance, and are basically assigned to the same cluster as the nearest neighbors of the allocated areas. In order to ensure flight safety and meet the C5 constraint, new areas can be added only if the coverage mission can be completed before the drone runs out of energy. Therefore, suitable areas are selected from the unallocated areas that are close to the allocated areas and meet the energy constraints. There are many strategies that can be used to find suitable regions, one of the most commonly used strategies is the closest to the end strategy. In this strategy, the unallocated region closest to the head or tail of the cluster being examined is selected.

[0042] 305. Cluster update.

[0043] The appropriate area obtained in the above stage Add to selected cluster in the sequence. like Compared to the cluster tail being closer to the cluster head, It is inserted before the head of the cluster and becomes the new head of the cluster queue; otherwise, it is placed at the end and becomes the new tail of the cluster queue. After all regions are allocated, the adjustable region clustering phase ends.

[0044] 4. Drone sequence optimization strategy.

[0045] The drones perform tasks in the order of the regions in the regional clustering. However, in most cases, this order cannot meet the drone planning requirements. Therefore, an ant colony method is used to plan the region sequence assigned to the drone and adjust the execution order of regional tasks to minimize the drone's time consumption.

[0046] 401. Initialization phase.

[0047] Based on the ant colony method, the initial value of the information between any two regions is set to be the same. Since the nearest neighbor optimization strategy is often used in the planning process to optimize the order of region visits, the heuristic information is initialized according to the distance between regions, where closer regions are more likely to be selected by the drone for the next visit.

[0048] 402. Local update rules.

[0049] The drone swarm adopts both local and global information update strategies to make full use of the information contained in the distance between regions. In the process of constructing each drone solution, each time the drone moves, a local update of information occurs. The local update rule can increase the diversity of solutions and avoid premature convergence of the optimization process.

[0050] 403. Global update rules.

[0051] The global update of information occurs after all drones have completed their routes and built solutions. Only the drone that has found the shortest path is allowed to update the information. During the global update process, only the information on the global optimal path will be enhanced, while other information will be weakened. The global update rule promotes the dissemination of empirical information and effectively improves the convergence speed of the algorithm.

[0052] It is worth noting that the contents not elaborated in detail in the present invention are all prior art and are well known to those skilled in the art.

[0053] Therefore, the present invention adopts the above-mentioned adjustable planning method for drones for small-area coverage, which improves the path planning efficiency and task execution speed of multi-drone systems in multi-area coverage tasks, reduces the time consumed by drones to execute tasks, optimizes the area access sequence of drones, and provides an efficient and reliable solution for practical applications.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. An adjustable planning method for UAVs covering small areas, characterized in that: The following steps are involved: Step S1: Establish a UAV and coverage area model according to coverage requirements and UAV performance parameters; Step S2, characterizing the constraints of coverage area planning; Step S3: On the basis of satisfying the constraint conditions, a clustering algorithm is used to perform adjustable region clustering, the region is divided into multiple subsets, and tasks are assigned to the drones; Step S4: Optimize the area visit sequence of the UAV to perform the task by using the ant colony optimization algorithm.

2. The adjustable planning method of UAV for small area coverage according to claim 1 is characterized in that: In step S1, a drone model is established. The specific operations are as follows: Each drone is modeled as ,in, and Respectively represent The maximum flight speed and maximum flight altitude of each drone, Indicates The width of the drone’s search area; When a drone is used to scan an area, its speed is limited to: ； in, Indicates A drone is used to scan Speed ​​limit in each area, express A collection of , Respectively represent the number of drones and the number of scanned areas; When a drone exist If the mission cannot be performed in the area, ; Calculate the time it takes for the drone to perform tasks in the area: ； in, Indicates drone exist The time it takes to perform tasks in the area; express The scan range of the area.

3. The adjustable planning method of UAV for small area coverage according to claim 2 is characterized in that: In step S1, a coverage area model is established, and the specific operations are as follows: Calculate the Regions and Flight distance between regions , thus obtaining The spatial distance between any two regions in the region .

4. The adjustable planning method for UAVs covering small areas according to claim 3 is characterized in that: In step S2, the constraints representing the coverage area planning include: C1 constraint: When the drone is not selected to perform area coverage missions, it should stay at the base; C2 constraint: Each drone must take off from the base when performing a mission and return to the base after the mission is completed; C3 constraint: Each area can only be covered by one drone; C4 constraint: The drone should visit and cover all assigned areas one by one; C5 Constraint: The drone must return to base before running out of energy.

5. The adjustable planning method of UAV for small area coverage according to claim 4 is characterized in that: In step S3, a clustering algorithm is used to cluster the adjustable regions, the regions are divided into multiple subsets, and tasks are assigned to the drones. The specific operations are as follows: Step S31, determining the cluster center according to the density value and distance value of each area; Step S32: After all cluster centers are selected, each remaining region is classified into the same cluster as its nearest neighbor; use To describe the various clusters of drone swarms, Indicates that it is classified into clusters For each cluster , whose center is the first region added to the cluster, namely: ； in, Representation clustering the center of Step S33: From the currently available clusters, according to the time requirements of each cluster, select the cluster with the smallest time requirement as the cluster to be expanded first; ； in, represents the cluster with the smallest time requirement; Represents the currently available clustering set; Indicates coverage of a given cluster The time consumption requirements of each area; Indicates that it is classified into clusters The regional sequence of Indicates that it is classified into clusters No. Regions; Step S34: In the unallocated area, select a suitable area according to the relative distance to the allocated area and the C5 constraint condition. ; Step S35: Set the appropriate area Add to selected cluster In the sequence, according to its distance from the cluster head and tail, it is decided to insert it before the cluster head or after the tail, thereby updating the cluster sequence structure.

6. The adjustable planning method of UAV for small area coverage according to claim 5, characterized in that: In step S4, the ant colony optimization algorithm is used to optimize the area visit sequence of the drone to perform the task, so as to improve the planning efficiency and reduce the time consumption of the drone to perform the task. The specific operations are as follows: Step S41, initialization phase, setting the initial pheromone value between any two areas and initializing the heuristic information according to the area distance; Step S42, local update rule, in the process of constructing the ant solution, local update of pheromone is performed after each movement; Step S43, global update rule: after all ants complete their journey, only the ants that have found the shortest path are allowed to update the pheromone, thus enhancing the pheromone on the global optimal path.

Images