Regional coverage task planning method for unmanned cluster

Abstract

The invention provides an unmanned cluster-oriented area coverage task planning method, and belongs to the technical field of swarm intelligence, and the method comprises the steps: carrying out the area distribution of a plurality of unmanned platforms according to a preset distribution rule based on a plurality of unmanned platforms and a plurality of areas, and generating an initial flight matrix; determining the unmanned platform with the longest task execution completion time in the initial flight matrix, performing local optimization on an area corresponding to the unmanned platform based on an access sequence optimization rule to reduce the task execution completion time of the unmanned platform, and generating a first flight matrix; determining the unmanned platform with the longest task execution completion time in the first flight matrix, performing global optimization on an area corresponding to the unmanned platform based on an adjacent area adjustment rule to reduce the task execution completion time of the unmanned platform, and generating a second flight matrix; and executing an area coverage task according to the second flight matrix. Through the processing scheme provided by the invention, the capability of the unmanned cluster to complete the complex task is improved.

Description

Technical Field

[0001] The present application relates to the field of swarm intelligence technology, and in particular to a method for planning area coverage tasks for unmanned swarms. Background Art

[0002] In recent years, with the continuous improvement of autonomy and intelligence, the application scenarios of unmanned platforms in civil and military fields have become more and more extensive. Unmanned platforms can be flexibly deployed in harsh environments that are difficult for humans to adapt to, and independently complete disaster detection, search and rescue, regional detection and intelligence collection. In addition, they can adaptively adjust the working mode and re-plan the flight path according to the dynamic changes of the environment and targets during the execution of the mission, which greatly improves the reliability and safety of completing the mission. However, due to the limited sensing range, computing power and energy supply, it is difficult for a single unmanned platform to meet the complex requirements of large-scale applications.

[0003] When performing complex tasks, unmanned swarms are more flexible and adaptable, and have excellent task execution efficiency and economic benefits. Unmanned swarms are usually composed of differentiated autonomous unmanned platforms, which have different flight capabilities and carry different mission payloads. They achieve group cooperation and coordination under the premise of individual autonomy, and have the ability to quickly adapt to dynamically changing mission requirements and target constraints, greatly improving the ability to perform tasks in changeable or even unknown environments. They have been widely used in military and civilian fields and have broad application prospects. For example, in the military field, autonomous unmanned swarms carrying differentiated sensors and detectors can collaboratively search complex geographical areas and achieve high-precision positioning of targets; in the agricultural field, multiple unmanned platforms can collaborate with each other in farmlands and forests to monitor the growth status of crops and vegetation in real time, providing data support for farm managers to make wise decisions.

[0004] Although unmanned swarms can greatly reduce system construction costs, improve system flexibility, and enhance system problem-solving capabilities, as the scale of swarms continues to expand, the task planning problem of unmanned swarms has become increasingly complex. Especially when performing comprehensive detection tasks such as target search and personnel search and rescue, the regional coverage task planning of unmanned swarms must not only arrange reasonable flight paths for each unmanned platform in the swarm to scan the entire mission area, but also ensure that the goals of the regional coverage task and the safety of each unmanned platform can be met to the greatest extent. This problem has been proven to have NP-hard computational complexity, and it is difficult to obtain the optimal solution within a limited time.

[0005] In a typical unmanned swarm area coverage mission planning problem, the unmanned swarm area coverage mission planning method needs to comprehensively consider various factors such as the environmental constraints faced by the swarm, the terrain of the region, the distance between regions, the capability constraints of the unmanned platform, and the target constraints of the mission. It can also allocate the areas to be covered and scanned for each unmanned platform and determine the access paths to these areas within a reasonable time to ensure the efficient completion of the mission. In recent years, the area coverage mission planning problem of unmanned swarms has received widespread attention. However, due to the complexity of the problem or the focus, previous studies often only focus on the geometric constraints of a single region, lack of consideration of the distribution of the region and the heterogeneity of the unmanned platform, and lack of efficient multi-region coverage mission planning methods for heterogeneous unmanned swarms. Summary of the invention

[0006] In view of this, an embodiment of the present application provides an area coverage mission planning method for unmanned clusters, which at least partially solves the problems existing in the prior art of only focusing on the geometric constraints of a single area, lacking consideration of the distribution of the area and the heterogeneity of the unmanned platform, and lacking an efficient multi-area coverage mission planning method for heterogeneous unmanned clusters.

[0007] The present application provides a method for planning an area coverage task for an unmanned swarm, the method comprising:

[0008] Based on multiple unmanned platforms and multiple areas, the multiple unmanned platforms are allocated areas according to a preset allocation rule to generate an initial flight matrix, wherein the initial flight matrix includes an area access list of each unmanned platform and an access order of each unmanned platform to the areas in the area access list;

[0009] Determine the unmanned platform with the longest mission completion time in the initial flight matrix, locally optimize the area in the area access list corresponding to the unmanned platform based on the rule of access order optimization to reduce the mission completion time of the unmanned platform, and generate a first flight matrix;

[0010] Determine the unmanned platform with the longest mission completion time in the first flight matrix, perform global optimization on the regions in the region access list corresponding to the unmanned platform based on the rules of neighboring region adjustment to reduce the mission completion time of the unmanned platform, and generate a second flight matrix;

[0011] The multiple unmanned platforms perform area coverage tasks according to the second flight matrix.

[0012] According to a specific implementation of an embodiment of the present application, the region allocation of the multiple unmanned platforms according to a preset allocation rule includes:

[0013] Select the unmanned platform with the shortest completion time for the current task, and add the unallocated area to the area access list of the unmanned platform;

[0014] For each unallocated area, respectively calculate the increment of the unmanned platform's task completion time when the unallocated area is allocated to the unmanned platform, add the unallocated area corresponding to the smallest increment to the unmanned platform's area access list, and set the access order;

[0015] An unmanned platform is allocated to each unallocated area in turn to complete the area allocation of multiple unmanned platforms.

[0016] According to a specific implementation method of an embodiment of the present application, the increment of the execution task completion time of the unmanned platform when the unallocated area is allocated to the unmanned platform is calculated respectively, the unallocated area corresponding to the smallest increment is added to the area access list of the unmanned platform, and the access order is set, including:

[0017] Let the unallocated area be R j The unmanned platform with the shortest task completion time is U s , the area access list of the unmanned platform is C s , Area Access List C s The access order queue of the area in is Q s ;

[0018] Calculate when R j Assigned to unmanned platform U s When U s Arrival Area R j Time Arr s，j , U s Complete R j Cover the time of the scan taskFin s，j and U s From R j Time to return to the initial area Back s，j , the initial area is the starting point of each unmanned platform;

[0019] For each unallocated area, based on the time Arr s，j 、TimeFin s，j and time Back s，j , respectively calculate the unallocated area R j Assigned to the unmanned platform U s The increment of the execution task completion time;

[0020] Add the unallocated area corresponding to the smallest increment to the unmanned platform U s The access order queue Q sand update the area access list C s And the unmanned platform U s The time to complete the task.

[0021] According to a specific implementation of an embodiment of the present application, the local optimization of the area in the area access list corresponding to the unmanned platform based on the rule of access order optimization includes:

[0022] Adjust the access order of the areas in the area access list corresponding to the unmanned platform in turn;

[0023] When performing the access order adjustment of the current area, the inter-area flight time after the current area is exchanged with the area adjacent to the current area is calculated, and when the inter-area flight time meets the first preset requirement, the current area is exchanged with the area adjacent to the current area, and the execution task completion time of the unmanned platform and the access order of the area in the area access list corresponding to the unmanned platform are updated;

[0024] If all the areas in the area access list corresponding to the unmanned platform cannot meet the preset requirements, the local optimization is completed.

[0025] According to a specific implementation of an embodiment of the present application, when performing the access order adjustment of the current area, calculating the inter-area flight time after exchanging the current area with an area adjacent to the current area, and when the inter-area flight time meets the first preset requirement, exchanging the current area with an area adjacent to the current area, including:

[0026] Let the current region be numbered The numbers of the preceding first region, the succeeding first region, and the succeeding second region adjacent to the current region are

[0027] When performing the access order adjustment of the current area, the inter-area flight time from the previous first area to the current area is calculated respectively. Inter-area flight time from the subsequent first area to the subsequent second area Inter-area flight time from the first preceding area to the first subsequent area And the inter-region flight time from the current region to the subsequent second region

[0028] When the inter-area flight time meets a first preset requirement, the current area is exchanged with a subsequent first area, and the first preset requirement is:

[0029]

[0030] According to a specific implementation of an embodiment of the present application, the globally optimizing the area in the area access list corresponding to the unmanned platform based on the rules of neighboring area adjustment includes:

[0031] The regions in the region access list corresponding to the unmanned platform are globally optimized based on rules of neighboring region exchange and region migration.

[0032] According to a specific implementation of an embodiment of the present application, the globally optimizing the area in the area access list corresponding to the unmanned platform based on the rules of neighboring area exchange and area migration includes:

[0033] Assume that the unmanned platform is U max , unmanned platform U max The corresponding regional access list is Q max , Area Access List Q max The area in

[0034] For each area Find not in Q max Medium distance area The nearest area R nb , obtain the visited area R by querying the first flight matrix nb Unmanned platform U ng , let region R nb On unmanned platform ng Regional access list Q ng The position of is tp, then

[0035] If the area From the zone access list Q max Migrate to zone access list Q ng Region R nb When the subsequent position of the area meets the second preset requirement, the area Migrate to Region R nb The subsequent location of the area From the zone access list Q max Delete from;

[0036] When each area If the second preset requirement is not met, the area is exchanged. Access list in zone Q max The location and area R nb Access list in zone Q ng When the position in the unmanned platform U is exchanged max and unmanned platform U ngIf the execution task completion time meets the third preset requirement, the area Access list in zone Q max The location and area R nb Access list in zone Q ng Exchange the positions in

[0037] Traverse the area access list Q max , when all regions cannot meet the second preset requirement and the third preset requirement, the global optimization is completed.

[0038] According to a specific implementation of the embodiment of the present application, the second preset requirement is:

[0039]

[0040] in, For region R nb To area The inter-regional flight time, For Region To area R nb The inter-regional flight time of the subsequent first region, For unmanned platform ng In the area The time required to complete the scanning task, T(U max ) origin For unmanned platform max Initial execution task completion time;

[0041] The third preset requirement is: the unmanned platform U after the area exchange max The mission completion time and the unmanned platform U after area exchange ng The task completion time of the unmanned platform U max The initial task completion time T(U max ) origin .

[0042] According to a specific implementation of the embodiment of the present application, the method further includes:

[0043] Determine the unmanned platform with the longest mission completion time in the second flight matrix, perform local optimization on the area in the area access list corresponding to the unmanned platform based on the rule of access order optimization to reduce the mission completion time of the unmanned platform, and generate a third flight matrix;

[0044] The multiple unmanned platforms perform area coverage missions according to the third flight matrix.

[0045] Beneficial effects:

[0046] The regional coverage task planning method for unmanned clusters in the embodiment of the present application includes three stages: initial solution generation based on multiple rules, local optimization based on the longest access time, and global optimization based on regional exchange. The planned initial flight matrix is ​​optimized from the unmanned platform level and the cluster level, which can effectively solve the multi-region coverage task planning problem under the unmanned cluster and effectively improve the ability of the unmanned cluster to complete complex tasks. In addition, the planning method proposed in this application can be used to solve similar vehicle routing, logistics scheduling and other problems by modifying the generation method of the initial solution and the optimization strategy of the solution, and has certain application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 The present invention is a flowchart of a method for planning an area coverage mission for an unmanned swarm according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0050] The following describes the implementation methods of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, in the absence of conflict, the following embodiments and the features in the embodiments can be combined with each other. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work belong to the scope of protection of the present application.

[0051] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present application, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this device and / or practice this method.

[0052] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. The illustrations only show the components related to the present application rather than being drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0053] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0054] The embodiment of the present application provides a regional coverage task planning method for unmanned clusters, which can solve the multi-region coverage task planning problem under heterogeneous unmanned clusters, generate optimal and approximately optimal planning schemes for this problem, and effectively improve the ability of unmanned clusters to complete complex tasks. The existing methods based on clustering or evolutionary computing types have a relatively long execution time, and these methods usually only focus on the regional access order of a single unmanned platform when optimizing the initial scheme, and do not consider the optimization of the scheme from the perspective of the cluster. Therefore, it is necessary to design a heuristic method with both global optimization and local refinement capabilities to better complete the task. The planning method designed by the present invention includes three stages: initial scheme generation based on multiple rules, local optimization based on the longest access time, and global optimization based on regional exchange. The planned scheme is optimized from the unmanned platform level and the cluster level. In addition, the scheduling method proposed here is not only applicable to regional coverage task planning problems, but can also be extended to scheduling problems such as vehicle routing and logistics scheduling, and has a wide range of application scenarios.

[0055] Refer to the following Figure 1 Described in detail, the method of this embodiment specifically includes the following steps:

[0056] Step S101: Based on multiple unmanned platforms and multiple areas, the multiple unmanned platforms are allocated areas according to a preset allocation rule to generate an initial flight matrix, wherein the initial flight matrix includes an area access list of each unmanned platform and an access order of each unmanned platform to areas in the area access list;

[0057] Step S102, determining the unmanned platform with the longest mission completion time in the initial flight matrix, performing local optimization on the regions in the region access list corresponding to the unmanned platform based on the rule of access order optimization to reduce the mission completion time of the unmanned platform, and generating a first flight matrix;

[0058] Step S103, determining the unmanned platform with the longest mission completion time in the first flight matrix, performing global optimization on the regions in the region access list corresponding to the unmanned platform based on the rules of neighboring region adjustment to reduce the mission completion time of the unmanned platform, and generating a second flight matrix;

[0059] Step S104: the multiple unmanned platforms perform area coverage tasks according to the second flight matrix.

[0060] In this embodiment, the method includes three stages: initial solution generation based on multiple rules, local optimization based on the longest access time, and global optimization based on regional exchange. The planned initial flight matrix is ​​optimized from the unmanned platform level and the cluster level, which can effectively solve the multi-region coverage task planning problem under the unmanned cluster and effectively improve the ability of the unmanned cluster to complete complex tasks. In addition, the planning method proposed in this application can be used to solve similar vehicle routing, logistics scheduling and other problems by modifying the generation method of the initial solution and the optimization strategy of the solution, and has certain application prospects.

[0061] During the specific implementation, it is necessary to build a regional coverage mission planning model, which includes the following contents:

[0062] Consider a cluster U = {U1, U2, ..., U n In m disconnected geographic areas R = {R1, R2, ..., R m} to perform coverage search tasks. Each unmanned platform can use a triplet U i = <V i , T i , W i > indicates that V i , T i , W iThey represent the flight speed of the unmanned platform, the maximum continuous flight time of the unmanned platform, and the scanning radius of the sensor configured on the unmanned platform. Generally speaking, the maximum continuous flight time of the unmanned platform is determined by factors such as flight speed, battery capacity, mission environment, and weight of the unmanned platform. For the sake of simplicity, the maximum continuous flight time of the unmanned platform constructed here is a known and fixed value. j The scan area is represented by A j , Region R j and R k The Euclidean distance between j，k Due to the heterogeneity of the flight capabilities and payload capacities of unmanned platforms, different unmanned platforms have different scanning times in the same geographical area. i，j Indicates unmanned platform U i In region R j The time required to complete the scanning task, TF i，j，k Indicates unmanned platform U i From area R j To area R k Flight time. TS i，j and TF i，j，k They can be calculated by the following formulas respectively

[0063]

[0064] Assume that all unmanned platforms are initially located in the same initial area R0, and the unmanned platform needs to return to the initial area after completing the coverage task. Since the unmanned platform does not need to scan the initial area, the scanning area of ​​R0 is 0, that is, R0 = 0. 0，j (j∈[1,m]) represents the initial region to region R j Based on the above formal representation of unmanned platforms and geographical areas, the regional coverage task of unmanned swarm can be defined as: given n independent and heterogeneous unmanned platforms UAV = {U1, U2, ..., U n} and m geographically dispersed regions R = {R1, R2, ..., R m}, assign area coverage scanning tasks to each unmanned platform, and determine the optimal flight path for each unmanned platform. Under the premise of meeting the energy consumption constraints of the unmanned platform, minimize the completion time of the unmanned cluster to complete all target area coverage scanning tasks.

[0065] Using the three-dimensional matrix X = {x i，j，k |i∈[1,n],j,k∈「0,m]} represents the flight path of the unmanned platform, where x i，j，k It is a Boolean variable, representing the unmanned platform U i Whether from region R j Fly to area Rk If the unmanned platform U i From R j Fly to Region R k , x i，j，k =1, otherwise x i，j，k =0.

[0066] Each unmanned platform that performs an area coverage mission must take off from the initial area. In other words, the unmanned platform either stays in the initial area or starts from the initial area to perform the mission. The number of times it takes off from the initial area is equal to the number of times it arrives at the initial area. i，0，k Indicates unmanned platform U i Whether to fly from the initial area to area R j , the following conditions must be met:

[0067]

[0068]

[0069] If an unmanned platform U i Fly to Region R j , the scanning task is performed, then U i Also need to be from R j Leave. In other words,

[0070]

[0071] Each unmanned platform selected to perform a mission should complete the mission and return to the initial area before its energy is exhausted. i，j Indicates unmanned platform U i In region R j The scanning time, TF i，j，k Indicates U i From area R j Fly to Region R k The flight time is , the following conditions are met:

[0072]

[0073] Using T(U i ) and T(U, R) represent the unmanned platform U i The time it takes to execute the task and the task completion time of the entire unmanned cluster, T(U i ) and T(U, R) are calculated as follows:

[0074]

[0075]

[0076] The regional coverage mission planning problem requires finding the optimal flight path matrix X for the unmanned swarm so that the mission completion time is minimized, that is,

[0077] min T(U,R) (8)

[0078] In one embodiment, the allocating regions of the plurality of unmanned platforms according to a preset allocation rule includes:

[0079] Select the unmanned platform with the shortest completion time for the current task, and add the unallocated area to the area access list of the unmanned platform;

[0080] For each unallocated area, respectively calculate the increment of the unmanned platform's task completion time when the unallocated area is allocated to the unmanned platform, add the unallocated area corresponding to the smallest increment to the unmanned platform's area access list, and set the access order;

[0081] An unmanned platform is allocated to each unallocated area in turn to complete the area allocation of multiple unmanned platforms.

[0082] Further, the step of respectively calculating the increment of the task completion time of the unmanned platform when the unallocated area is allocated to the unmanned platform, adding the unallocated area corresponding to the smallest increment to the area access list of the unmanned platform, and setting the access order includes:

[0083] Let the unallocated area be R j The unmanned platform with the shortest task completion time is U s , the area access list of the unmanned platform is C s , Area Access List C s The access order queue of the area in is Q s ;

[0084] Calculate when R j Assigned to unmanned platform U s When U s Arrival Area R j Time Arr s，j , U s Complete R j Cover the time of the scan taskFin s，j and U s From R j Time to return to the initial area Back s，j , the initial area is the starting point of each unmanned platform;

[0085] For each unallocated area, based on the time Arr s，j 、TimeFin s，i and time Backs，j , respectively calculate the unallocated area R j Assigned to the unmanned platform U s The increment of the execution task completion time;

[0086] Add the unallocated area corresponding to the smallest increment to the unmanned platform U s The access order queue Q s and update the area access list C s And the unmanned platform U s The time to complete the task.

[0087] In specific implementation, since the multi-area coverage task planning problem under a heterogeneous unmanned cluster is an NP-hard problem, there is no planning method that can solve the optimal solution for this problem in polynomial time. Therefore, this application proposes a method that can quickly find an approximate optimal planning solution for this problem. The method includes three execution stages: generation of an initial task planning solution based on multiple rules, local optimization of the solution based on the longest completion time, and global optimization of the solution based on neighboring area exchange. In the stage of generating the initial task planning solution based on multiple rules, the solution generates a planning solution for the task planning problem that satisfies the regional constraints and unmanned platform constraints, assigns an unmanned platform to perform scanning tasks to each area, and allocates the order of access to the area for the unmanned platform planning. The initial planning solution generated at this stage provides input for the optimization of subsequent stages.

[0088] Let C = {C1, C2, .., C n} represents the currently allocated region cluster, where C i Indicates the current unmanned platform U i Access collections of regions. Queues Represents C i Unmanned Platform U i The number of the area to be scanned in sequence, where and Unmanned platform U i The first and last areas currently visited. In order to minimize the completion time of the unmanned cluster as much as possible, it is necessary to balance the access time of each set. Therefore, each time the unmanned platform with the smallest current completion time is selected for operation, and the unallocated areas are considered to be added to the access set of the unmanned platform. s represents the selected cluster, then:

[0089]

[0090] Among them, D(C i ) is cluster C i The time requirement is the time cost of visiting and covering all areas in the cluster in sequence:

[0091]

[0092] In the unmanned platform U s The area access set C s When adding a region, for the region R that is not currently assigned j , respectively calculate when R j Assigned to unmanned platform U s When U s Arrival Area R j Time Arr s，j , U s Complete R j Cover the time of the scan taskFin s，j , U s From R j Return to the time to reach the initial area s，j , the calculation formulas are:

[0093]

[0094] Fin s，j =Arr s，j +TS s，j (12)

[0095] Back s，j =Fin s，j +TF s，j，0 (13)

[0096] Select the region R with the smallest unmanned platform completion time increment sec , that is, after adding this area, the unmanned platform can return to the initial area as quickly as possible, that is

[0097]

[0098] R sec Add to queue Q s The tail of U s The area access set and completion time information are obtained. Then the next iteration is carried out, a new area is selected according to the above process, and it is allocated to the best unmanned platform until all areas are allocated.

[0099] In one embodiment, the locally optimizing the area in the area access list corresponding to the unmanned platform based on the access order optimization rule includes:

[0100] Adjust the access order of the areas in the area access list corresponding to the unmanned platform in turn;

[0101] When performing the access order adjustment of the current area, the inter-area flight time after the current area is exchanged with the area adjacent to the current area is calculated, and when the inter-area flight time meets the first preset requirement, the current area is exchanged with the area adjacent to the current area, and the execution task completion time of the unmanned platform and the access order of the area in the area access list corresponding to the unmanned platform are updated;

[0102] If all the areas in the area access list corresponding to the unmanned platform cannot meet the preset requirements, the local optimization is completed.

[0103] Further, when performing the access order adjustment of the current area, calculating the inter-area flight time after exchanging the current area with an area adjacent to the current area, and when the inter-area flight time meets the first preset requirement, exchanging the current area with an area adjacent to the current area, includes:

[0104] Set the current region number to The numbers of the preceding first region, the succeeding first region, and the succeeding second region adjacent to the current region are

[0105] When performing the access order adjustment of the current area, the inter-area flight time from the previous first area to the current area is calculated respectively. Inter-area flight time from the subsequent first area to the subsequent second area Inter-area flight time from the first preceding area to the first subsequent area And the inter-region flight time from the current region to the subsequent second region

[0106] When the inter-area flight time meets a first preset requirement, the current area is exchanged with a subsequent first area, and the first preset requirement is:

[0107]

[0108] During specific implementation, the method (initial flight matrix) proposed in step S101 can find a planning scheme that satisfies all constraints for the multi-region coverage task planning problem under the unmanned cluster. However, this problem has a lot of solutions that meet the requirements, and the solutions found by the rule-based heuristic method are generally not the optimal solution, or even the suboptimal solution. In order to improve the quality of the solution, this embodiment proposes a local optimization method based on the completion time of the unmanned platform to optimize the initial planning scheme. In the original planning scheme, each unmanned platform has different regional coverage tasks and access paths, and the coverage areas and access paths between different unmanned platforms do not interfere with each other. The completion time of the cluster execution task is determined by the unmanned platform with the longest completion time. Local optimization means that when the area accessed by the unmanned platform is not changed, the unmanned platform with the longest completion time is selected, and the order of the unmanned platform accessing the area is adjusted to reduce the completion time of the unmanned platform to perform the task. After adjusting the area access order of the unmanned platform, the task completion time of this unmanned platform and the completion time of the cluster are recalculated, and the unmanned platform with the longest completion time is reselected for iterative optimization until the completion time of the cluster does not change. During the entire iterative optimization process, the area coverage mission of each unmanned platform does not change, but the order in which the unmanned platforms visit these areas changes.

[0109] Continue to use C = {C1, C2, ..., C n} represents the area access set of each unmanned platform in the initial planning scheme, Represents C i Unmanned Platform U i The order of visiting regions, and They are unmanned platform U i The numbers of the first and last areas visited. For the found planning scheme, that is, the generated flight path matrix X, first calculate the task completion time T(U i ), select the unmanned platform with the longest completion time for optimization operation, that is,

[0110]

[0111] For unmanned platforms max The region access sequence Q max , the method tries to find a region that meets the requirements and adjusts the access order of this region and subsequent regions to reduce U max Specifically, the method starts from Q max Start searching from the first region in , assuming that the kth region is currently being processed, i.e., region numbered , then the preceding area, the subsequent area, and the second subsequent area are Since adjusting the area access order does not affect the time it takes the unmanned platform to perform area coverage scanning, only the flight time between areas is considered here.

[0112]

[0113] If true, the kth and k+1th areas visited by the unmanned platform are swapped, and the list number of the partial areas visited by the unmanned platform is replaced by Adjust to If an area that meets the requirements is found, this round of search ends and the completion time of this unmanned platform is updated. Otherwise, continue in Q max Find the next area in the optimization. If all areas cannot meet the requirements, the method ends execution, otherwise it enters the next round of iterative optimization.

[0114] In one embodiment, the globally optimizing the area in the area access list corresponding to the unmanned platform based on the rules of neighboring area adjustment includes:

[0115] The regions in the region access list corresponding to the unmanned platform are globally optimized based on rules of neighboring region exchange and region migration.

[0116] Furthermore, the region in the region access list corresponding to the unmanned platform is globally optimized based on rules of neighboring region exchange and region migration, including:

[0117] Assume that the unmanned platform is U max , unmanned platform U max The corresponding regional access list is Q max , Area Access List Q max The area in

[0118] For each area Find not in Q max Medium distance area The nearest area R nb , obtain the visited area R by querying the first flight matrix nb Unmanned platform U ng , let region R nb On unmanned platform ng Regional access list Q ng The position of is tp, then

[0119] If the area From the zone access list Q max Migrate to zone access list Q ng Region R nb When the subsequent position of the area meets the second preset requirement, the area Migrate to Region R nb The subsequent location of the area From the zone access list Q max Delete from;

[0120] When each area If the second preset requirement is not met, the area is exchanged. Access list in zone Q max The location and area R nb Access list in zone Q ng When the position in the unmanned platform U is exchanged max and unmanned platform U ng If the execution task completion time meets the third preset requirement, the area Access list in zone Q max The location and area R nb Access list in zone Q ng Exchange the positions in

[0121] Traverse the area access list Q max , when all regions cannot meet the second preset requirement and the third preset requirement, the global optimization is completed.

[0122] The second preset requirement is:

[0123]

[0124] in, For region R nb To area The inter-regional flight time, For Region To area R nb The inter-regional flight time of the subsequent first region, For unmanned platform ng In the area The time required to complete the scanning task, T(U max ) origin For unmanned platform max Initial execution task completion time;

[0125] The third preset requirement is: the unmanned platform U after the area exchange max The mission completion time and the unmanned platform U after area exchange ng The task completion time of the unmanned platform U max The initial task completion time T(U max ) origin .

[0126] In specific implementation, the optimization method proposed in step S102 only performs local optimization from the perspective of a single unmanned platform, and can only optimize the regional access order of the unmanned platform, and the optimization strength is limited. Therefore, this embodiment starts from a global perspective, starting from the regional access list of multiple unmanned platforms, and adjusting the regional coverage tasks of the unmanned platform based on neighboring regional exchange and regional migration, which can further reduce the task completion time of the cluster, and after adjusting the regional access list of the unmanned platform, the local optimization method in the above embodiment can be used again to adjust the regional access order of the unmanned platform.

[0127] Specifically, for the flight path matrix X (first flight matrix) optimized in step S102, firstly, the unmanned platform U with the longest task completion time is calculated using formula (15): max , get the area access list Q of this unmanned platform max . Then, for Q max Each area in Find not in Q max Medium distance area The nearest area R nb , then query the flight path matrix X to get access to R nb Unmanned platform U ng , assuming R nb in U ng Regional access list Q ng The position is tp (i.e. ). The method proposed in this section first attempts to perform the insertion operation if the condition

[0128]

[0129] Established, that is, if R nb From Q max Migrate to Q ng , U ng The new task completion time is less than U max The initial completion time will be From Q max Delete it and place it in Q ng Medium R nb The subsequent position of , ends this round of search.

[0130] If condition (17) is not met, try to swap and R nb In Q max and Q ng In the position, if after the exchange, U ng and U max The new completion time is less than U max Initial completion time T(Umax ) origin , then exchange and R nb In Q max and Q ng The position in the search ends this round. max If no region that meets the requirements is found, the method ends execution, otherwise it enters the next round of iterative optimization.

[0131] In one embodiment, the method further comprises:

[0132] Determine the unmanned platform with the longest mission completion time in the second flight matrix, perform local optimization on the area in the area access list corresponding to the unmanned platform based on the rule of access order optimization to reduce the mission completion time of the unmanned platform, and generate a third flight matrix;

[0133] The multiple unmanned platforms perform area coverage missions according to the third flight matrix.

[0134] The method in the above embodiment includes three execution stages, namely, planning the initial plan, adjusting the unmanned platform area access order, and adjusting the area access list operation between unmanned platforms, wherein the latter two stages can be run iteratively. The overall execution process of the method is as follows:

[0135] 1. Input. Input unmanned cluster UAV = {U1, U2, ..., U n} and the region set R = (R1, R2, ..., R m}, establish a three-dimensional matrix X = {x i，j，k |i∈[1,n],j,k∈[0,m]} stores the solution results.

[0136] 2. Initial plan establishment stage: Select areas one by one according to the rules, and assign the best unmanned platform to this area until all areas are assigned.

[0137] 2.1 Use formula (9) to select the unmanned platform U with the shortest current completion time s ;

[0138] 2.2 Use formulas (13) and (14) to select the value that can make U s Returns the region R with the least time increase in the initial region sec , add this area to U s The regional access set Q s The tail of

[0139] 2.3 If all regions have been allocated, jump to stage 3 for execution; otherwise, jump to step 2.1 to continue iteration.

[0140] 3. Local optimization stage: Find the access order of the unmanned platform areas that need to be optimized according to the completion time of the unmanned platform, and perform local optimization on the initial planning scheme generated in stage 2.

[0141] 3.1 Use formula (15) to select the unmanned platform U with the largest current completion time max ;

[0142] 3.2 Find Q according to the judgment condition in formula (16) max The area that meets the demand, that is, adjusting the access order of this area and subsequent areas, can reduce U max completion time;

[0143] 3.3 If step 3.2 finds an area that meets the requirements, reduce U max If the time is completed, jump to 3.1 iteration execution; otherwise, jump to stage 4 execution.

[0144] 4. Global optimization stage: Find the unmanned platform area access list that needs to be optimized based on the completion time of the unmanned platform, and further optimize the solution after stage 3 optimization.

[0145] 4.1 Use formula (15) to select the unmanned platform U with the largest current completion time max ;

[0146] 4.2 Targeting Q max Each area in Find distance area The nearest area R nb , according to the judgment condition in formula (17) and R nb Whether insertion or swapping operations can be performed to reduce the overall task completion time of the cluster;

[0147] 4.3 If the insertion or exchange operation in step 4.2 is successful, jump to step 4.1 for iterative execution; otherwise, jump to stage 5 for execution.

[0148] 5. Output: Output the unmanned platform flight matrix X after optimization in the global optimization phase.

[0149] The embodiment provided by the present invention comprehensively considers indicators such as the scanning time of the area, the shortest distance to the allocated area, and the distance to the initial area to select an area that has not been allocated, allocate the best unmanned platform to this area, iteratively allocate each area, and find a planning scheme that meets all constraint requirements for the task planning problem; for the planning scheme found, select the unmanned platform with the longest completion time, adjust the order of the unmanned platform accessing the area, update the planning scheme and iterate; for the previously optimized planning scheme, select the unmanned platform with the longest completion time, insert or exchange the area accessed by the unmanned platform with other unmanned platforms according to the nearest neighbor principle, update the planning scheme and iterate.

[0150] In the initial planning scheme generation stage, the method needs to allocate the best unmanned platform for each area, and at the same time, it needs to determine the access path of the unmanned platform on the allocated area set. The method adopts a heuristic allocation strategy to allocate each area one by one. During an allocation process, the method first selects the unmanned platform with the shortest return time to the initial area, and then selects the area with the smallest increment of time for the unmanned platform to visit the initial area based on the area last visited by the unmanned platform, the distance between areas, and the distance from the area to the initial area, and adds this area to the end of the unmanned platform access area list. The method iterates the allocation of areas, and when all areas have been allocated the best unmanned platform, the initial planning scheme is generated.

[0151] In the local optimization stage of unmanned platform regional access, the method optimizes the initial planning scheme generated in the previous stage to further reduce the task completion time of the unmanned cluster on the regional set. This stage is an iterative process, and the completion time of all unmanned platforms in the scheme is evaluated each time. Since the completion time of the unmanned cluster is determined by the unmanned platform with the longest completion time, the unmanned platform with the longest completion time is selected, and the regional access path of the unmanned platform is adjusted to reduce the task completion time of the unmanned platform. The task completion time of the cluster is updated. At this time, the unmanned platform with the longest completion time may have changed. Then the unmanned platform with the longest completion time is selected to enter the next round of iteration. If the completion time of the cluster does not change after the update, the iteration ends and enters the next stage.

[0152] In the global optimization stage of unmanned platform regional access, the method optimizes the planning scheme after local optimization in the previous stage at the global level. This stage is also an iterative optimization process. In each round of iteration, the method selects the unmanned platform with the longest completion time from the scheme for optimization. The method first determines the regional access set of the unmanned platform with the longest task completion time, and then determines the geographically closest area for each area visited by this unmanned platform based on the nearest neighbor principle, and tries to migrate the area to the regional set of unmanned platforms that visit the adjacent area or exchange the positions of the two areas in their respective regional access sets. If performing the above migration or exchange operation can reduce the completion time of the unmanned platform, the method continues to iterate, otherwise the execution ends.

[0153] Compared with the prior art, the present invention has three main advantages. 1. At the problem level, by combining the characteristics of the problem to design an effective task planning method covering the three stages of initial planning scheme generation, single unmanned platform area access order adjustment, and multi-unmanned platform area exchange, the multi-area coverage task planning problem under unmanned clusters can be effectively solved. 2. At the method level, by carefully designing a heuristic method to solve the regional coverage task planning problem, the scope of problems that the method can solve is expanded. 3. At the scalability level, by modifying the generation method of the initial plan and the optimization strategy of the plan, the proposed heuristic method can be used to solve problems similar to those solved by the present invention, such as vehicle routing and logistics scheduling, and has certain application prospects.

[0154] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A regional coverage mission planning method for unmanned swarms, characterized in that: The method comprises: Based on multiple unmanned platforms and multiple areas, the multiple unmanned platforms are allocated areas according to a preset allocation rule to generate an initial flight matrix, wherein the initial flight matrix includes an area access list of each unmanned platform and an access order of each unmanned platform to the areas in the area access list; Determine the unmanned platform with the longest mission completion time in the initial flight matrix, locally optimize the area in the area access list corresponding to the unmanned platform based on the rule of access order optimization to reduce the mission completion time of the unmanned platform, and generate a first flight matrix; Determine the unmanned platform with the longest mission completion time in the first flight matrix, perform global optimization on the regions in the region access list corresponding to the unmanned platform based on the rules of neighboring region adjustment to reduce the mission completion time of the unmanned platform, and generate a second flight matrix; The multiple unmanned platforms perform area coverage tasks according to the second flight matrix.

2. The method for planning area coverage tasks for unmanned swarms according to claim 1, characterized in that: The performing area allocation for the plurality of unmanned platforms according to a preset allocation rule includes: Select the unmanned platform with the shortest completion time for the current task, and add the unallocated area to the area access list of the unmanned platform; For each unallocated area, respectively calculate the increment of the unmanned platform's task completion time when the unallocated area is allocated to the unmanned platform, add the unallocated area corresponding to the smallest increment to the unmanned platform's area access list, and set the access order; An unmanned platform is allocated to each unallocated area in turn to complete the area allocation of multiple unmanned platforms.

3. The method for planning area coverage tasks for unmanned swarms according to claim 2, characterized in that: The step of respectively calculating the increment of the task completion time of the unmanned platform when the unallocated area is allocated to the unmanned platform, adding the unallocated area corresponding to the smallest increment to the area access list of the unmanned platform, and setting the access order includes: Let the unallocated area be R j The unmanned platform with the shortest task completion time is U s , the area access list of the unmanned platform is C s , Area Access List C s The access order queue of the area in is Q s ; Calculate when R j Assigned to unmanned platform U s When U s Arrival Area R j Time Arr s，j , U s Complete R j Cover the time of the scan taskFin s，j and U s From R j Time to return to the initial area Back s，j , the initial area is the starting point of each unmanned platform; For each unallocated area, based on the time Arr s，j 、TimeFin s，j and time Back s，j , respectively calculate the unallocated area R j Assigned to the unmanned platform U s The increment of the execution task completion time; Add the unallocated area corresponding to the smallest increment to the unmanned platform U s The access order queue Q s and update the area access list C s And the unmanned platform U s The time to complete the task.

4. The method for planning area coverage tasks for unmanned swarms according to claim 1, characterized in that: The locally optimizing the area in the area access list corresponding to the unmanned platform based on the access order optimization rule includes: Adjust the access order of the areas in the area access list corresponding to the unmanned platform in turn; When performing the access order adjustment of the current area, the inter-area flight time after the current area is exchanged with the area adjacent to the current area is calculated, and when the inter-area flight time meets the first preset requirement, the current area is exchanged with the area adjacent to the current area, and the execution task completion time of the unmanned platform and the access order of the area in the area access list corresponding to the unmanned platform are updated; If all the areas in the area access list corresponding to the unmanned platform cannot meet the preset requirements, the local optimization is completed.

5. The method for planning area coverage tasks for unmanned swarms according to claim 4, characterized in that: The step of calculating the inter-region flight time after the current region is exchanged with the region adjacent to the current region when the access order adjustment of the current region is performed, and exchanging the current region with the region adjacent to the current region when the inter-region flight time meets a first preset requirement includes: Set the current region number to The numbers of the preceding first region, the succeeding first region, and the succeeding second region adjacent to the current region are When performing the access order adjustment of the current area, the inter-area flight time from the previous first area to the current area is calculated respectively. Inter-area flight time from the subsequent first area to the subsequent second area Inter-area flight time from the first preceding area to the first subsequent area And the inter-region flight time from the current region to the subsequent second region When the inter-area flight time meets a first preset requirement, the current area is exchanged with a subsequent first area, and the first preset requirement is:

6. The method for planning area coverage tasks for unmanned swarms according to claim 1, characterized in that: The globally optimizing the area in the area access list corresponding to the unmanned platform based on the rules of neighboring area adjustment includes: The regions in the region access list corresponding to the unmanned platform are globally optimized based on rules of neighboring region exchange and region migration.

7. The method for planning area coverage tasks for unmanned swarms according to claim 6, characterized in that: The globally optimizing the area in the area access list corresponding to the unmanned platform based on the rules of neighboring area exchange and area migration includes: Assume that the unmanned platform is U max , unmanned platform U max The corresponding regional access list is Q max ,Region Access List Q max The area in For each area Find not in Q max Medium distance area The nearest area R nb , obtain the visited area R by querying the first flight matrix nb Unmanned platform U ng , let region R nb On unmanned platform ng Regional access list Q ng The position of is tp, then If the area From the zone access list Q max Migrate to zone access list Q ng Region R nb When the subsequent position of the area meets the second preset requirement, the area Migrate to Region R nb The subsequent location of the area From the zone access list Q max Delete from; When each area If the second preset requirement is not met, the area is exchanged. Access list in zone Q max The location and area R nb In the area access list Q ng When the position in the unmanned platform U is exchanged max and unmanned platform U ng If the execution task completion time meets the third preset requirement, the area In the area access list Q max The location and area R nb Access list in zone Q ng Exchange the positions in ; Traverse the area access list Q max , when all regions cannot meet the second preset requirement and the third preset requirement, the global optimization is completed.

8. The method for planning area coverage tasks for unmanned swarms according to claim 7, characterized in that: The second preset requirement is: in, For region R nb To area The inter-regional flight time, For Region To area R nb The inter-regional flight time of the subsequent first region, For unmanned platform ng In the area The time required to complete the scanning task, T(U max ) origin For unmanned platform max Initial execution task completion time; The third preset requirement is: the unmanned platform U after the area exchange max The mission completion time and the unmanned platform U after area exchange ng The task completion time of the unmanned platform U max The initial task completion time T(U max ) origin .

9. The method for planning area coverage tasks for unmanned swarms according to claim 1, characterized in that: The method further comprises: Determine the unmanned platform with the longest mission completion time in the second flight matrix, perform local optimization on the area in the area access list corresponding to the unmanned platform based on the rule of access order optimization to reduce the mission completion time of the unmanned platform, and generate a third flight matrix; The multiple unmanned platforms perform area coverage missions according to the third flight matrix.

Images