CN116579564A

CN116579564A - Regional reconnaissance task planning method, device, computer equipment and medium

Info

Publication number: CN116579564A
Application number: CN202310558347.7A
Authority: CN
Inventors: 叶青; 伍国华; 周玲
Original assignee: Central South University
Current assignee: Central South University
Priority date: 2023-05-17
Filing date: 2023-05-17
Publication date: 2023-08-11

Abstract

The application relates to a regional scout task planning method, a regional scout task planning device, computer equipment and a storage medium. The method comprises the following steps: taking the minimum time for completing all the scout tasks as a target, taking the scout task requirements, the truck and unmanned aerial vehicle configuration information as constraint conditions, and constructing a multi-vehicle multi-machine task planning model; based on the decomposition strategy, obtaining a task planning initial solution under each truck unmanned aerial vehicle distribution scheme; optimizing the initial solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle by a self-adaptive large neighborhood search algorithm integrating disturbance and tabu strategies to obtain the optimal solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle; and outputting task planning scheme information with shortest total task time according to the optimal solution of task planning under the allocation scheme of each truck unmanned aerial vehicle. The application realizes the multi-vehicle multi-machine task planning and better optimization effect.

Description

Regional reconnaissance task planning method, device, computer equipment and medium

Technical Field

The application relates to the field of task planning, in particular to a multi-vehicle multi-unmanned aerial vehicle collaborative region reconnaissance task planning method, a multi-vehicle multi-unmanned aerial vehicle collaborative region reconnaissance task planning device, computer equipment and a storage medium.

Background

At present, the rotor unmanned aerial vehicle is widely applied to regional reconnaissance, and has the advantages of low cost, good concealment, high low-altitude scanning precision and the like, but meanwhile, the problem of poor cruising ability is commonly existed. The range of the area covered by a single unmanned plane in one voyage is very limited, so that in the regional reconnaissance task in a larger range, the unmanned plane needs to repeatedly return to a base station to charge or replace a battery, so that great resource waste is caused, and the task completion efficiency is influenced. Under the condition, the working mode of the truck and the unmanned aerial vehicle can effectively save the mileage of the unmanned aerial vehicle, expand the range of the task and improve the efficiency of completing the task.

However, the application of the multi-vehicle and multi-unmanned cooperative system in area coverage is still to be further explored. At present, most of existing researches on the coverage problem of the cooperative area of the vehicle and the machine adopt a mode that one truck carries a plurality of unmanned aerial vehicles, the quantity of unmanned aerial vehicles carried by each truck is fixed, a single-vehicle multi-machine system has a certain limitation in the coverage of a larger area, the requirement of higher timeliness cannot be met, and after the quantity of unmanned aerial vehicles reaches a certain degree, the improvement of the efficiency by continuously increasing the unmanned aerial vehicles is not obvious. Therefore, when the multi-vehicle multi-machine collaborative mode is applied to regional reconnaissance, how to improve task completion efficiency, expand regional reconnaissance range, and how to more effectively utilize unmanned aerial vehicles, further research is needed in the aspects of exploring the number of unmanned aerial vehicles carried on each vehicle, and the like. The prior art also has the problem of poor adaptability.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a regional scout task planning method, apparatus, computer device, and storage medium that can improve the efficiency of the wide-range regional scout.

A regional scout mission planning method, the method comprising:

acquiring regional scout task information; the regional scout task information comprises regional information to be scout, truck information and unmanned aerial vehicle information; the regional information to be detected comprises base information, road network information and truck unmanned aerial vehicle meeting point information;

according to the regional scout task information, taking the minimum time for completing all scout tasks as a target, taking scout task requirements and truck and unmanned aerial vehicle configuration information as constraint conditions, and constructing a multi-vehicle multi-machine task planning model; the truck and unmanned aerial vehicle configuration information is determined by the truck information and the unmanned aerial vehicle information;

based on a decomposition strategy, decomposing the multi-vehicle multi-machine task planning model into a plurality of single-vehicle multi-machine models under a plurality of truck unmanned aerial vehicle distribution schemes, and respectively carrying out task planning on the single-vehicle multi-machine models to obtain task planning initial solutions under the truck unmanned aerial vehicle distribution schemes;

Optimizing the initial solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle through a self-adaptive large neighborhood search algorithm integrating disturbance and tabu strategies to obtain the optimal solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle;

and outputting task planning scheme information with shortest total task time according to the optimal solution of task planning under the allocation scheme of each truck unmanned aerial vehicle.

An area reconnaissance mission planning apparatus, the apparatus comprising:

the task information acquisition module is used for acquiring regional reconnaissance task information; the regional scout task information comprises regional information to be scout, truck information and unmanned aerial vehicle information; the regional information to be detected comprises base information, road network information and truck unmanned aerial vehicle meeting point information;

the multi-vehicle multi-machine task planning model construction module is used for constructing a multi-vehicle multi-machine task planning model by taking the minimum time for completing all the scout tasks as a target and taking the scout task requirements, the truck and the unmanned aerial vehicle configuration information as constraint conditions according to the regional scout task information; the truck and unmanned aerial vehicle configuration information is determined by the truck information and the unmanned aerial vehicle information;

the initial solution determining module is used for decomposing the multi-vehicle multi-machine task planning model into a plurality of single-vehicle multi-machine sub-models under a plurality of truck unmanned aerial vehicle allocation schemes based on a decomposition strategy, and respectively carrying out task planning on the single-vehicle multi-machine sub-models to obtain task planning initial solutions under the truck unmanned aerial vehicle allocation schemes;

The optimal solution determining module is used for optimizing the initial solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle through a self-adaptive large neighborhood searching algorithm integrating disturbance and tabu strategies to obtain the optimal solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle;

and the scheme information output module is used for outputting the task planning scheme information with the shortest total task time according to the optimal solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle.

A computer device comprising a memory storing a computer program and a processor which when executing the computer program performs the steps of:

A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:

According to the regional scout task planning method, the regional scout task planning device, the computer equipment and the storage medium, the minimum time for completing all scout tasks is taken as a target, scout task requirements, truck and unmanned aerial vehicle configuration information are taken as constraint conditions, and a multi-vehicle multi-machine task planning model is constructed; decomposing the multi-vehicle multi-machine task planning model into a plurality of single-vehicle multi-machine sub-models under a plurality of truck unmanned aerial vehicle allocation schemes based on a decomposition strategy, and respectively carrying out task planning for the single-vehicle multi-machine sub-models to obtain task planning initial solutions under the truck unmanned aerial vehicle allocation schemes; optimizing the initial solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle by a self-adaptive large neighborhood search algorithm integrating disturbance and tabu strategies to obtain the optimal solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle; and outputting task planning scheme information with shortest total task time according to the optimal solution of task planning under the allocation scheme of each truck unmanned aerial vehicle. The invention realizes multi-vehicle multi-machine task planning, the designed adaptive large neighborhood search algorithm integrating disturbance and tabu strategies combines the adaptive criterion of an ALNS algorithm, the neighborhood search strategy and the metapolis criterion of an SA algorithm, and simultaneously adds the tabu strategy to increase the memory of the historical search process, and the disturbance operation is added to enable the search process to be more diversified, and the invention also has a plurality of neighborhood structures with different characteristics, thereby realizing better optimization effect.

Drawings

FIG. 1 is a flow chart of a regional scout mission planning method according to an embodiment;

FIG. 2 is a schematic diagram of multi-vehicle multi-unmanned area reconnaissance in one embodiment;

FIG. 3 is a schematic diagram of initial solution generation for multi-vehicle multi-machine collaborative area scout mission planning in one embodiment;

FIG. 4 is a schematic diagram of an example of region division in one embodiment, where 4 (a) is the region to be scouted, 4 (b) is the mesh division result, and 4 (c) is the sub-region division result;

FIG. 5 is a schematic illustration of sub-zone division into drones in one embodiment;

FIG. 6 is a schematic representation of individual codes in one embodiment;

FIG. 7 is a schematic diagram of a randomly changing truck path in one embodiment;

FIG. 8 is a schematic diagram of a change in the longest truck path in one embodiment;

FIG. 9 is a schematic diagram of perturbation operation in one embodiment;

FIG. 10 is a graph showing the results of efficiency improvement for each 1 unmanned aircraft in one embodiment;

FIG. 11 is a block diagram of a regional scout mission planning apparatus in one embodiment;

fig. 12 is an internal structural diagram of a computer device in one embodiment.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

The regional scout task planning method provided by the application can be applied to the following application scenes. The area is provided with a known road network, the road network is provided with a plurality of known points of integration of the unmanned aerial vehicles of the trucks, which can carry out truck berthing and unmanned aerial vehicle taking off and landing, and the full coverage scanning of the area is completed by utilizing a plurality of unmanned aerial vehicles and a plurality of trucks so as to acquire the corresponding information of the area. And each truck carries an unmanned aerial vehicle to move on the regional road network, when reaching a junction of a certain truck unmanned aerial vehicle, all the unmanned aerial vehicles on the truck fly, the unmanned aerial vehicle returns to the truck to replace a battery after finishing a reconnaissance task in the subarea, and then the next task is executed until the subarea is completely covered, and the truck carries the unmanned aerial vehicle to go to the next junction. Thus, until all sub-areas are detected, all trucks return to the base with the drone.

In one embodiment, as shown in fig. 1, there is provided a regional scout mission planning method, including the steps of:

step 102, obtaining regional scout task information.

The regional scout task information comprises regional information to be scout, truck information and unmanned aerial vehicle information; the regional information to be scouted comprises base information, road network information and truck unmanned aerial vehicle meeting point information.

Fig. 2 is an example diagram of a multi-vehicle multi-drone collaborative area scout mission plan, with black triangles representing truck drone junction points and black squares representing base. In the figure, two unmanned aerial vehicles are carried together to finish tasks, three unmanned aerial vehicles are carried on a left truck to sequentially access the meeting point of the unmanned aerial vehicles of the No. 1, no. 2 and No. 3 trucks, the unmanned aerial vehicles fly at the meeting point and finish the regional scanning tasks of all white background grids, gray solid lines, light gray dotted lines and black dotted lines respectively represent scanning tracks of the 3 unmanned aerial vehicles in regions, and bold and black solid lines represent driving paths of the left truck. The right side truck carries two unmanned aerial vehicles to sequentially access the meeting point of the No. 4, no. 5 and No. 6 unmanned aerial vehicles, the unmanned aerial vehicles fly at the meeting point and complete the regional scanning task of all gray grids, black dotted lines and light gray dotted lines on the gray grids respectively represent the scanning tracks of 2 unmanned aerial vehicles in the region, and the thickened gray solid lines represent the driving paths of the right side truck. 2 trucks all start from the base, return to the base after completing the task, and each unmanned aerial vehicle can only scan one grid at a time.

And 104, constructing a multi-vehicle multi-machine task planning model by taking the minimum time for completing all the scout tasks as a target and taking the scout task requirements, the truck and unmanned aerial vehicle configuration information as constraint conditions according to the regional scout task information.

The problem in the present invention is assumed as follows:

(1) The truck can only run on the road network and cannot enter the area to execute the scout task;

(2) The unmanned aerial vehicle can only take off and land at the meeting point of the unmanned aerial vehicle of the truck accessed by the truck;

(3) The unmanned plane scans only one grid at a time, which is called a reconnaissance task;

(4) Trucks and drones both run at constant speed;

(5) Neglecting the time for taking off, landing and replacing the battery of the unmanned aerial vehicle;

(6) When the unmanned aerial vehicle executes a reconnaissance task, the truck stays in place to wait for the unmanned aerial vehicle to land;

(7) At each truck drone junction, the truck must wait until all drones have completed the reconnaissance mission back on the vehicle before proceeding to the next truck drone junction.

(8) When the unmanned aerial vehicle scans the grid, the grid vertex closest to the meeting point of the truck unmanned aerial vehicle is selected as a scanning starting point, and the grid vertex next closest to the meeting point of the truck unmanned aerial vehicle is selected as a scanning ending point.

The multi-vehicle multi-unmanned aerial vehicle collaborative area reconnaissance also comprises the following characteristics:

(9) The mileage of each truck is limited;

(10) The number of unmanned aerial vehicles carried by each truck may be different;

(11) After the truck flies the unmanned aerial vehicle, the unmanned aerial vehicle can be recovered in situ, and the unmanned aerial vehicle can be transported to the next junction to wait for the unmanned aerial vehicle under the condition that the range of the unmanned aerial vehicle is allowed.

(12) The flying spot and the landing spot of each unmanned aerial vehicle in one task must be on the same truck, and the unmanned aerial vehicle is not allowed to land on other trucks in the task execution process.

The mathematical symbols according to the present invention are shown in table 1.

Table 1 symbol illustrates

TABLE 1-1 set symbol illustrations

TABLE 1-2 parameter and decision variable sign illustrations

Specifically, the multi-vehicle multi-machine task planning model constructed by the invention is as follows:

(1) Objective function

The problem presented herein targets the least time to complete all reconnaissance tasks, i.e. the least time to get back to the truck or drone at the base at the latest.

Min{t|t≥max(t1 _h ,t2 _k ),h∈Tr,k∈D} (1)

(2) Constraint conditions

The constraint conditions are as follows:

t ₀ ＝(T _max -l _D /v _D ) (2)

constraint (2) represents the time required to calculate a drone to scout a mesh. Constraint (3) indicates that the vehicle on which the unmanned aerial vehicle is mounted must issue from the base, and return to the same base after all coverage tasks are completed. Constraint (4) indicates that the degree of egress from each truck drone junction is equal to the degree of ingress, thereby ensuring connectivity of the vehicle route. Constraint (5) indicates that each truck drone junction is accessible only once by one truck at most. Constraint (6) indicates that each mesh within each sub-area can only be reconnaissad by one drone. Constraint (7) indicates that the time required for the drone to finish a reconnaissance of one mesh at a time does not exceed its maximum endurance. Constraint condition (8) indicates that the maximum scanning total area of all unmanned aerial vehicles in each area is larger than or equal to the area of the area, so that the area can be completely covered, namely, each grid contained in the area is guaranteed to be detected by the unmanned aerial vehicle. Constraint (9) indicates that after the truck reaches the meeting point of the truck unmanned aerial vehicle, all unmanned aerial vehicles on the truck are put off. The constraint (10) indicates that each truck carries at least one drone. Constraint (11) indicates that the sum of the number of drones carried by all trucks is equal to the total number of drones. Equation (12) indicates that the path length of the truck cannot exceed the maximum range limit. The constraints (13), (14), (15) define the range of values of the 0-1 variable.

And 106, decomposing the multi-vehicle multi-machine task planning model into a plurality of single-vehicle multi-machine sub-models under a plurality of truck unmanned aerial vehicle distribution schemes based on a decomposition strategy, and respectively carrying out task planning for the single-vehicle multi-machine sub-models to obtain initial solutions of task planning under each truck unmanned aerial vehicle distribution scheme.

In the conventional multi-vehicle multi-machine problem, the number of unmanned aerial vehicles carried by each truck is generally defined in the problem setting, and the number of unmanned aerial vehicles carried by each truck is the same and fixed. The setting provides more ideas and possibilities for problem solving, and compared with the traditional multi-vehicle multi-machine problem, the problem solving method can find better solutions and is more in line with the situation of the actual problem. The initial solution generation at the algorithm of the present invention is thus divided into two steps: (1) Different truck unmanned aerial vehicle allocation schemes are generated, and the problem of multiple trucks and multiple machines is decomposed into the problem of multiple single trucks and multiple machines; (2) The initial solutions are constructed under different truck drone allocation schemes, respectively. As shown in fig. 3.

When different truck unmanned aerial vehicle allocation schemes are generated, because the area to be detected is larger, the unmanned aerial vehicle allocation method is convenient for task allocation of unmanned aerial vehicles, and firstly, area division is carried out. And dividing subareas according to the result of regional meshing, wherein the main purpose is to allocate grids to be scanned for each truck unmanned aerial vehicle junction so as to form a reconnaissance subarea set or subareas contained in each truck unmanned aerial vehicle junction. And then, according to the grid quantity of each subarea, distributing subareas to each unmanned aerial vehicle, so that grids distributed by each unmanned aerial vehicle are as average as possible. Finally, according to the number of trucks, the unmanned aerial vehicles are randomly distributed, all unmanned aerial vehicles are distributed to different trucks, 1 truck and a plurality of unmanned aerial vehicles are organized into a single-car-multi-machine system, and the subareas distributed by all unmanned aerial vehicles in the system serve as the areas to be covered of the system, so that each truck is distributed to 2 or more unmanned aerial vehicles.

When constructing an initial solution under different truck unmanned aerial vehicle allocation schemes, the generation of the initial solution follows the ideas of constructing a single-vehicle-multi-machine system firstly, then carrying out task planning on each single-vehicle-multi-machine system, and finally merging the single-vehicle-multi-machine task planning schemes, and the method comprises the following 2 parts: 1) Carrying out task planning on each single-vehicle-multi-machine system under different unmanned aerial vehicle allocation schemes of the trucks, carrying out truck path planning by using a dynamic planning algorithm, and carrying out unmanned aerial vehicle task planning by using a task allocation algorithm to form a task planning scheme; 2) And combining the single-multi-machine task planning schemes to form an initial task planning scheme of each multi-machine and taking the maximum value of the system time of each single-machine and each multi-machine as the task completion time.

And step 108, optimizing the initial solution of the task plan under the allocation scheme of each truck unmanned aerial vehicle by a self-adaptive large neighborhood search algorithm integrating disturbance and tabu strategies to obtain the optimal solution of the task plan under the allocation scheme of each truck unmanned aerial vehicle.

In the self-adaptive large neighborhood search algorithm integrating disturbance and tabu strategies, a simulated annealing mechanism and the tabu strategies are combined on the basis of self-adaptive large neighborhood search. Designing a plurality of neighborhood structures, accepting the new solution if the generated new solution is better than the current optimal solution in the neighborhood transformation process, and replacing the current optimal solution with the new solution; if the generated new solution is worse than the current optimal solution, determining whether to accept the solution by Metropolis criterion. Meanwhile, grids which need to be operated in the process of generating new solutions are added into the tabu list, so that repeated searching of the grids is avoided. If the tabu list is full, the grid is released according to the first-in first-out rule. In addition, when the historical optimal solution is h _max And when the iteration is not better, performing disturbance operation on the algorithm.

And 110, outputting task planning scheme information with shortest total task time according to the optimal solution of task planning under the allocation scheme of each truck unmanned aerial vehicle.

In the regional scout task planning method, the multi-vehicle multi-machine task planning model is constructed by taking the minimum time for completing all scout tasks as a target according to regional scout task information and taking scout task requirements, truck and unmanned aerial vehicle configuration information as constraint conditions; decomposing the multi-vehicle multi-machine task planning model into a plurality of single-vehicle multi-machine sub-models under a plurality of truck unmanned aerial vehicle allocation schemes based on a decomposition strategy, and respectively carrying out task planning for the single-vehicle multi-machine sub-models to obtain task planning initial solutions under the truck unmanned aerial vehicle allocation schemes; optimizing the initial solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle by a self-adaptive large neighborhood search algorithm integrating disturbance and tabu strategies to obtain the optimal solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle; and outputting task planning scheme information with shortest total task time according to the optimal solution of task planning under the allocation scheme of each truck unmanned aerial vehicle. The invention realizes multi-vehicle multi-machine task planning, the designed adaptive large neighborhood search algorithm integrating disturbance and tabu strategies combines the adaptive criterion of an ALNS algorithm, the neighborhood search strategy and the metapolis criterion of an SA algorithm, and simultaneously adds the tabu strategy to increase the memory of the historical search process, and the disturbance operation is added to enable the search process to be more diversified, and the invention also has a plurality of neighborhood structures with different characteristics, thereby realizing better optimization effect.

In one embodiment, the method further comprises: dividing the region to be detected into unitized grids, and determining grid center point information; the size of the grid is equal to the maximum scanning area of the unmanned aerial vehicle; constructing a distance matrix between grids and the meeting points of the truck unmanned aerial vehicle according to the meeting point information of the truck unmanned aerial vehicle and the grid center point information, distributing grids to be scanned for each meeting point of the truck unmanned aerial vehicle based on a nearby principle according to the distance matrix, and determining sub-region information corresponding to each meeting point of the truck unmanned aerial vehicle; distributing the subareas to each unmanned aerial vehicle according to the subarea information so as to minimize the sum of squares of the differences of the grid quantity distributed by each unmanned aerial vehicle; according to the acquired truck quantity information, carrying out random distribution on the unmanned aerial vehicles so that all the unmanned aerial vehicles are distributed to different trucks to form a plurality of truck unmanned aerial vehicle distribution schemes; wherein each truck unmanned aerial vehicle distribution scheme comprises a plurality of single-vehicle-multi-machine models; the single-vehicle-multiple-machine model is composed of a truck and at least two unmanned aerial vehicles.

Specifically, the generation of the allocation scheme of the truck unmanned aerial vehicle mainly allocates unmanned aerial vehicles to each truck according to the number of trucks, and because the situation that 1 truck carries 1 unmanned aerial vehicle is not considered, all possible allocation schemes are obtained on the basis of ensuring that each truck is allocated to more than or equal to 2 unmanned aerial vehicles, and the coverage area of each single-vehicle-multi-machine system is obtained under each allocation scheme. The method is mainly divided into the following two parts:

1) Region division

Because the area to be detected is larger, an area dividing algorithm is designed for facilitating task allocation to the unmanned aerial vehicle, the area is meshed, and a subarea dividing result is obtained. The region meshing method is as follows: firstly, analyzing the size of an area, then calculating the maximum scanning area of the unmanned aerial vehicle according to the range of the unmanned aerial vehicle (reserving a section of maneuvering range), and finally dividing the whole area to be detected into unitized grids, wherein the size of each grid is equal to the maximum scanning area of the unmanned aerial vehicle. The scanning range of the unmanned aerial vehicle can be regarded as a square area with a side length of DW.

Assuming that all unmanned aerial vehicles are of the same type, the unmanned aerial vehicle maximum scanning area SD _max The calculation method is as follows:

SD _max ＝(T _max ×v _D -l _D )×DW (16)

wherein T is _max Is the maximum flight time of the unmanned aerial vehicle, v _D For unmanned aerial vehicle flight speed, l _D And reserving a range for the unmanned aerial vehicle, wherein the DW is the scanning width of the unmanned aerial vehicle.

And dividing subareas according to the result of regional meshing, wherein the main purpose is to allocate grids to be scanned for each truck unmanned aerial vehicle junction so as to form a reconnaissance subarea set or subareas contained in each truck unmanned aerial vehicle junction. The strategy employed herein is a nearby principle, namely, assigning a grid to the closest truck drone junction to it. First, a distance matrix between each grid center point and each truck unmanned aerial vehicle junction is calculated, and then each grid is distributed to the closest truck unmanned aerial vehicle junction to form a preliminary subarea division result.

Fig. 4 shows an example of square area division with a side length of 8 km. The total area of the scout area shown in FIG. 4 (a) is 64km ² The square 0 in this area represents a base, with 5 points of truck drone convergence, represented by triangles, numbered sequentially from 1 to 5. The maximum endurance time of the unmanned aerial vehicle is 19.5mins, the flying speed is 80km/h, the reserved range is 10km, the scanning width of the unmanned aerial vehicle is 100m, the maximum scanning area of the unmanned aerial vehicle is 4km2 calculated by a formula (15), and the time for scanning each grid by the unmanned aerial vehicle is 12mins. The entire area can be divided into 16 grids and the grids are sequentially numbered as shown in fig. 4 (b). Fig. 4 (c) is a subregion division example, and according to the subregion division policy described above, the No. 1, 2, 3, and No. 4 truck unmanned aerial vehicle junction points are equally divided into grids, forming 4 subregions corresponding to each other, and the No. 5 truck unmanned aerial vehicle junction point is not divided into grids. Thus, when planning a vehicle path thereafter, the truck to be accessed is unmannedThe machine meeting points are 1, 2, 3 and 4, and the No. 5 truck unmanned aerial vehicle meeting point is not in the vehicle path. Different color patches represent different sub-areas. Table 2 shows the division results of the subareas.

2) Sub-area allocation to unmanned aerial vehicle

The following allocation policies are applied: and distributing the subareas to each unmanned aerial vehicle according to the number of the subarea grids, so that the grids distributed by each unmanned aerial vehicle are as average as possible. Therefore, the task amount of each single-multi-machine system and the number of unmanned aerial vehicles in the system form a positive correlation, even if the number of unmanned aerial vehicles distributed by each truck is different, under the strategy, the task completion time of each single-multi-machine system is more uniform, and because the total task time takes the time minimum value of each distribution scheme and the time of each distribution scheme takes the time maximum value of each single-multi-machine system in the scheme, the initial solution with shorter task completion time can be obtained. If the total number of unmanned aerial vehicles is larger than the number of meeting points, unmanned aerial vehicles which are not distributed to the grid are regarded as virtual unmanned aerial vehicles, and then the virtual unmanned aerial vehicles are randomly added into each single-vehicle-multi-vehicle system. As shown in fig. 5, a schematic diagram of dividing a subarea into 36 grids (numbered 1-36) and 6 subareas (numbered 1-6), wherein a total of 4 unmanned aerial vehicles are used for area reconnaissance, and the number of grids contained in each subarea is 6, 9, 3, 5 and 8, so that subareas 1 and 3 are allocated to unmanned aerial vehicles No. 1, subarea 2 is allocated to unmanned aerial vehicle No. 2, subarea 6 is allocated to unmanned aerial vehicle No. 3, subareas 4 and 5 are allocated to unmanned aerial vehicles No. 4, and the number of grids allocated to each unmanned aerial vehicle is 9, 8 and 10.

3) Unmanned aerial vehicle distributes to the truck: according to truck quantity, carry out random distribution to unmanned aerial vehicle, distribute all unmanned aerial vehicle to different trucks on, 1 truck and many unmanned aerial vehicle mechanism are a single car-multi-machine system, and the subregion that all unmanned aerial vehicles distributed in the system is as the waiting coverage area of this system, guarantees that every truck distributes 2 unmanned aerial vehicles and above. Table 2 shows an example of 8 allocation schemes for a multi-vehicle-multi-machine system consisting of 9 robots and 3 trucks.

Table 2 unmanned aerial vehicle allocation to trucks schematic

In one embodiment, the method further comprises: respectively planning a vehicle path and an unmanned aerial vehicle task set for the single-vehicle-multiple-machine sub-model through a dynamic planning algorithm and a task allocation algorithm to obtain task planning scheme data corresponding to the single-vehicle-multiple-machine sub-model; according to the task planning scheme data, if the longest vehicle path value of all the single-multi-machine sub-models under the same truck unmanned aerial vehicle allocation scheme does not exceed the preset truck maximum range value, merging task planning scheme data corresponding to a plurality of single-multi-machine sub-models under the same truck unmanned aerial vehicle allocation scheme to obtain a task planning initial solution under the truck unmanned aerial vehicle allocation scheme; otherwise, the truck unmanned aerial vehicle allocation scheme is deleted.

Specifically, under the allocation scheme of different truck unmanned aerial vehicles, the generation of an initial solution follows the ideas of constructing a single-vehicle-multi-machine system firstly, then carrying out task planning on each single-vehicle-multi-machine system, and finally merging the single-vehicle-multi-machine task planning schemes, wherein the ideas comprise the following 2 parts: 1) Carrying out task planning on each single-vehicle-multi-machine system under different unmanned aerial vehicle allocation schemes of the trucks, carrying out truck path planning by using a dynamic planning algorithm, and carrying out unmanned aerial vehicle task planning by using a task allocation algorithm to form a task planning scheme; 2) And combining the single-multi-machine task planning schemes to form an initial task planning scheme of each multi-machine and taking the maximum value of the system time of each single-machine and each multi-machine as the task completion time.

The construction flow of the initial solution is shown in the algorithm, and the algorithm needs to input the vertex (v) ₁ ，v ₂ ，...，v _q ) Integration point set of unmanned truck (n) ₁ ，n ₂ ，...，n _k ) Maximum range S of unmanned aerial vehicle _max Maximum range R of truck _max The number of trucks ACar, the number of unmanned aerial vehicles AUAV. The whole area can be meshed and sub-area division can be carried out according to the maximum navigation of the unmanned aerial vehicle through an area division algorithm, and a sub-area division result and a truck unmanned aerial vehicle meeting point number (line 3) corresponding to each area are obtained; thereby calculating The grid quantity (line 4) contained in each subarea is distributed to each unmanned aerial vehicle by a task distribution algorithm to obtain a gridnodetoUAV (line 5) to be covered area distributed by each unmanned aerial vehicle; distributing all unmanned aerial vehicles to all trucks by adopting a random distribution strategy to obtain different truck unmanned aerial vehicle distribution schemes assignUtoC (line 6); decomposing to obtain a plurality of bicycle-multi-machine systems subsys under the allocation scheme of each truck unmanned aerial vehicle; if the longest vehicle path in all the single-vehicle multi-machine systems under the scheme does not exceed the maximum range R of the truck _max (line 8): a dynamic planning algorithm and a task allocation algorithm are used for respectively planning a vehicle path and an unmanned aerial vehicle task set for each single-vehicle-multiple-machine system to form a task planning scheme (line 10), the single-vehicle-multiple-machine system task planning schemes are combined to form a multiple-vehicle-multiple-machine task planning scheme (line 11) under the unmanned aerial vehicle allocation scheme of the truck, and the task time is the maximum value of the task time of each single-vehicle-multiple-machine system; otherwise, the allocation scheme of the truck unmanned aerial vehicle is regarded as invalid, deletion is carried out (line 14), and finally, the initial solutions and the corresponding task time (lines 15 and 16) under all the effective allocation schemes of the truck unmanned aerial vehicle are obtained.

(3) Individual coding mode

A new individual coding scheme is designed. As shown in fig. 6, in real number encoding mode, an individual is composed of L parts (L is the number of trucks). Each section represents a single car-multiple machine system (shown in the purple box). Wherein each bicycle-multi-machine system consists of the following parts: part 1 represents the truck path within the system, representing the order in which trucks visit the truck drone junction (as shown by the orange boxes); the rest represents the task set of each unmanned aerial vehicle in each subarea in the system.

In one embodiment, the method further comprises: optimizing the initial solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle by a self-adaptive large neighborhood search algorithm integrating disturbance and tabu strategies to obtain the optimal solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle; the adaptive large neighborhood search algorithm integrating disturbance and tabu strategies comprises the following steps:

step 1: parameter initialization is performed, including: all the initial weights of the destruction operators and the repair operators in the neighborhood structure are set to be 1; the current iteration number I in the simulated annealing mechanism is set to be 1, and the initial temperature T is set to be the initial temperature T _s The method comprises the steps of carrying out a first treatment on the surface of the The grid quantity L stored in the current tabu list is set to 0, and the number h of unchanged historical optimal solutions is set to 1;

step 2: acquiring task planning scheme data S of each bicycle-multi-machine model _k ；

Step 3: task planning scheme data S of each bicycle-multi-machine model ₁ ，...，S _m Merging to form multi-vehicle multi-machine initial task planning scheme data S, and giving the current initial task planning scheme data S to an optimal solution S _best ；

Step 4: selecting a destructive operator operation d epsilon DO and a repair operator operation r epsilon RO according to the destructive operator weight DO and the repair operator weight RO in the self-adaptive neighborhood selection mechanism to obtain latest task planning scheme data S';

step 5: comparing the total task completion time of the new solution S 'and the current solution S to obtain a time difference, wherein the time difference is represented by delta f, if the new solution meets the constraint, the new solution is accepted, and the new solution S' is assigned to the current solution S; meanwhile, if Deltaf is smaller than 0, representing the new solution can shorten the task completion time, and assigning the new solution S' to the optimal solution S _best The method comprises the steps of carrying out a first treatment on the surface of the If Deltaf is more than 0, judging whether to accept the solution of the refund according to Metropolis criterion;

step 6: updating the damage operator weight do and the repair operator weight ro, and updating the tabu table;

step 7: if h < h is satisfied _max Repeating the steps 4 to 6; if the cycle number reaches h _max Then go to step 8;

step 8: executing disturbance operation omega to form a new task planning scheme S, and assigning h to be 1;

step 9: if the total iteration number I is greater than or equal to I _max Or T.ltoreq.T _e ThenThe algorithm is terminated, and the current optimal solution S is output _best The method comprises the steps of carrying out a first treatment on the surface of the Otherwise, updating the iteration times I and the current temperature T, and returning to the step 4.

In the self-adaptive large neighborhood search algorithm integrating disturbance and tabu strategies, a simulated annealing mechanism and the tabu strategies are combined on the basis of self-adaptive large neighborhood search. Designing a plurality of neighborhood structures, accepting the new solution if the generated new solution is better than the current optimal solution in the neighborhood transformation process, and replacing the current optimal solution with the new solution; if the generated new solution is worse than the current optimal solution, determining whether to accept the solution by Metropolis criterion. Meanwhile, grids which need to be operated in the process of generating new solutions are added into the tabu list, so that repeated searching of the grids is avoided. If the tabu list is full, the grid is released according to the first-in first-out rule. In addition, when the historical optimal solution is h _max And when the iteration is not better, performing disturbance operation on the algorithm. When the pseudo code of the algorithm is as follows.

Based on the above description, in order to improve the quality of the solution, the following four destructive operators and four repair operators are designed by combining different destructive and repair strategies in the algorithm. Any one of the destruction operators and one of the repair operators are combined into one neighborhood operator. And a perturbation operation is added in the algorithm to make the searching process more diversified.

(1) Destructive operator

A total of 4 destructive operators are designed herein, the set of destructive operators being defined as DO. These operators are mainly used to delete the grid from the sub-region to which it originally belongs. The following is a detailed case of 4 destruction operators:

1) Random destruction operator

Randomly selecting a grid in each sub-area for removal;

2) Optimal destruction operator

Selecting a grid farthest from a confluence point of the truck unmanned aerial vehicle in each sub-area for removal;

3) Randomly changing truck paths

A truck path is randomly selected, a truck drone junction in the path is randomly selected, and the mesh contained in the junction is pruned from the set of tasks of the truck-drone system, as shown in fig. 7.

4) Changing the longest truck path

The longest one of the truck paths is selected, one of the truck drone junction points in the path is randomly selected, and the mesh contained in the junction point is pruned from the task set of the truck-drone system, as shown in fig. 8.

(2) Repair operator

In total, 4 repair operators are designed, and a repair operator set is defined as RO. These operators are mainly used to reinsert the grid after the operations of the destruction operator into the sub-area. The following is a detailed case of 4 repair operators:

1) Random repair between the same bicycle-multimachine combination

Randomly adding the damaged grids into a subarea where a confluence point of the truck unmanned aerial vehicle in the remaining range of the unmanned aerial vehicle is located;

2) Greedy repair between the same bicycle-multimachine combination

Adding the destroyed grid into a subarea capable of optimizing the current solution;

3) Random repair between different bicycle-multiple machine combinations

4) Greedy repair between different bicycle-multimachine combinations

the combination of the 1 destruction operator and the 1 repair operator can form 1 neighborhood structure, and 16 neighborhood structures can be formed through the random combination of the 4 destruction operators and the 4 repair operators, so that the neighborhood structure is selected for the self-adaptive large neighborhood search. Meanwhile, the process of neighborhood structure transformation needs to meet the constraint of the maximum driving mileage of the truck and the maximum range of the unmanned aerial vehicle, and if the constraint is not met, the iteration process is skipped.

(3) Disturbance criterion

To make the search process more diverse, when the historical optimal solution cannot become more optimal after h_max iterations, the following perturbation operations are performed on the algorithm: in order to equalize the task completion time of each single-multiple-machine combination so as to minimize the total task time, the number of unmanned aerial vehicles carried on each truck is redistributed, 2 single-multiple-machine combinations with the largest difference in total task time are extracted, and 1 unmanned aerial vehicle in the combination with the shortest total task time is distributed to the combination with the longest total task time, as shown in fig. 9. Unmanned aerial vehicle mission planning within different single-vehicle-multiple-vehicle combinations is then resumed. The disturbance operation has great significance on the large neighborhood algorithm, and the algorithm result can be further improved after the disturbance operation is added.

It should be understood that, although the steps in the flowchart of fig. 1 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 1 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of other steps or sub-steps of other steps.

In one embodiment, simulation experiments were performed using the method of the present invention. Comprising the following steps:

calculation example and experimental parameter design:

first, 5 rule area cases (hereinafter, referred to as cases 1 to 5) are generated, and each case information is shown in table 3. The experimental parameters were designed as shown in table 4: the average speed of the truck is set to be 40km/h, and the average speed of the unmanned aerial vehicle is set to be 80km/h; the scanning interval of the unmanned aerial vehicle is set to be 100m; the unmanned aerial vehicle has a duration of 19.5mins, The reserved range is 10km, so that the maximum coverage area SD of each unmanned aerial vehicle is obtained _max ＝4km ² A drone would take 12 minutes to scan a grid. The initial temperature in the ALNSAWPT algorithm is 1000 ℃, the final temperature is 10 ℃, the cooling rate is 0.97, the maximum iteration number is 10000, the iteration number in the isothermal process is 10, and the maximum number of the tabu list storage grids is 500.

TABLE 3 basic information of each example

Table 4 algorithm parameter settings

Experimental results and analysis:

in the area of 144km ² -400 km ² In 5 examples of (2) trucks 4, 5, 6, 7, 8, 9, 10 unmanned aerial vehicles and 3 trucks 6, 7, 8, 9, 10 unmanned aerial vehicles are respectively carried out experiments, and under the condition that the number of trucks is the same, the influence of the number of unmanned aerial vehicles on the task time is increased, and under the condition that the number of unmanned aerial vehicles is the same, the influence of the number of trucks on the task time is increased, so that the improvement effect of the multi-vehicle multi-machine mode on the task completion efficiency is verified. Firstly, according to the number of trucks, unmanned aerial vehicles are randomly distributed, all unmanned aerial vehicles are distributed to different trucks, one truck and a plurality of unmanned aerial vehicles are organized into a single-car-multi-machine system, and the sub-area distributed by all unmanned aerial vehicles in the system is used as the area to be covered of the system, so that each truck is distributed to more than two (including two) unmanned aerial vehicles. Generating an initial solution under each truck unmanned aerial vehicle allocation scheme, optimizing the initial solution by using a self-adaptive large neighborhood search algorithm integrating disturbance and tabu strategies to obtain a multi-vehicle multi-unmanned aerial vehicle collaborative region scout task planning scheme, taking the shortest task completion time in each allocation scheme as the finally obtained region scout task completion time, and each calculation The experiment was repeated 10 times to average. Comparing the total task completion time of the task planning scheme obtained by the ALNSAWPT algorithm with the total task completion time of the initial task planning scheme, as shown in table 5, the experimental results show that in 5 examples, the ALNSAWPT algorithm can shorten the task time by 8.17%, 9.98%, 9.95%, 10.71% and 10.80% respectively on the basis of the initial solution, and the task completion time is shortened by 9.92% on average in various multi-vehicle multi-machine situations. The ALNSAWPT algorithm has a good optimization effect.

Table 5 degree of lifting of initial solution by ALNSAWPT algorithm

In order to embody the superiority of the ALNSAWPT algorithm, an adaptive large neighborhood search algorithm (ALNSAWT, adaptive Large Neighborhood Search Algorithm with Tabu Strategies) fused with a tabu strategy and an adaptive large neighborhood search algorithm (ALNSAWP, adaptive Large Neighborhood Search Algorithm with Perturbation and Tabu Strategies) fused with a disturbance are designed at the same time, experiments are carried out on the above-mentioned examples, and the experimental effects of the ALNSAWPT algorithm are compared. Experimental results show that the ALNSAWPT algorithm combines the tabu strategy and the disturbance operation at the same time, has better optimization effect, and as shown in table 6, compared with the ALNSAWP algorithm which only combines the tabu strategy and the ALNSAWP algorithm which only combines the disturbance operation, the total task time can be shortened by 3.68% -5.47%.

Table 6 comparison of ALNSAWPT with other algorithms

The effect of the number of drones and the number of trucks on the total mission time was then analyzed. Firstly, under the condition that the number of trucks is the same, the influence of the increase of the number of unmanned aerial vehicles on the task time is analyzed, and the results of all the calculation examples are averaged. As shown in fig. 10, in the case of 2 trucks, in the experiments using 4, 5, 6, 7, 8, 9 and 10 unmanned aerial vehicles respectively, the total task time can be shortened by 12.39%, 7.64%, 5.78%, 6.29%, 2.10% and 1.49% on average on the basis of the previous experiment; in the case of 3 trucks, the total task time can be shortened by 6.17%, 6.92%, 4.62% and 2.44% on average on the basis of the previous experiment in experiments using 6, 7, 8, 9 and 10 unmanned aerial vehicles respectively. The experimental result shows that the total task time of regional reconnaissance increases with the increase of the scale of the calculation example; when the number of trucks is the same, the total mission time for regional reconnaissance decreases as the number of drones increases, and the overall presents a marginal decrementing effect.

And secondly, analyzing the influence of the increase of the number of trucks on the task time under the condition that the number of unmanned aerial vehicles is the same. In each of examples 1 to 5, the task efficiency improvement degree caused by adding 1 truck is improved, the results of each example are averaged, as shown in table 7, when the number of unmanned aerial vehicles is the same, the total task time of regional reconnaissance is reduced along with the increase of the number of trucks, and in the experiments using 1, 2 and 3 trucks respectively, the total task time can be shortened by 50.62% and 31.60% respectively on average on the basis of the previous experiment.

TABLE 7 average efficiency improvement with increasing truck count for each example

Experimental results show that the adaptive large neighborhood search algorithm integrating disturbance and tabu strategies can effectively solve the problem of regional reconnaissance task planning of multiple vehicles and multiple unmanned aerial vehicles, and has a good optimization effect on initial solutions. Compared with a single-vehicle multi-machine system, the multi-vehicle multi-machine system can greatly reduce task completion time and improve task completion efficiency.

In one embodiment, as shown in fig. 11, there is provided an area reconnaissance mission planning apparatus, including: the system comprises a task information acquisition module 1102, a multi-vehicle multi-machine task planning model construction module 1104, an initial solution determination module 1106, an optimal solution determination module 1108 and a scheme information output module 1110, wherein:

the task information acquisition module 1102 is configured to acquire regional scout task information; the regional scout task information comprises regional information to be scout, truck information and unmanned aerial vehicle information; the regional information to be detected comprises base information, road network information and truck unmanned aerial vehicle meeting point information;

the multi-vehicle multi-machine task planning model construction module 1104 is configured to construct a multi-vehicle multi-machine task planning model according to the regional scout task information, with minimum time for completing all scout tasks as a target, and with scout task requirements, truck and unmanned aerial vehicle configuration information as constraint conditions; the truck and unmanned aerial vehicle configuration information is determined by truck information and unmanned aerial vehicle information;

The initial solution determining module 1106 is configured to decompose the multi-vehicle multi-machine task planning model into a plurality of single-vehicle multi-machine sub-models under a plurality of truck unmanned aerial vehicle allocation schemes based on a decomposition strategy, and perform task planning for the single-vehicle multi-machine sub-models respectively to obtain a task planning initial solution under each truck unmanned aerial vehicle allocation scheme;

the optimal solution determining module 1108 is configured to optimize the task planning initial solution under the allocation scheme of each truck unmanned aerial vehicle by using an adaptive large neighborhood search algorithm that merges disturbance and tabu strategies, so as to obtain a task planning optimal solution under the allocation scheme of each truck unmanned aerial vehicle;

and the solution information output module 1110 is configured to output task planning solution information with the shortest total task time according to the optimal solution of task planning under the allocation solution of each truck unmanned aerial vehicle.

The initial solution determining module 1106 is further configured to divide the area to be detected into a unitized grid, and determine grid center point information; the size of the grid is equal to the maximum scanning area of the unmanned aerial vehicle; constructing a distance matrix between grids and the meeting points of the truck unmanned aerial vehicle according to the meeting point information of the truck unmanned aerial vehicle and the grid center point information, distributing grids to be scanned for each meeting point of the truck unmanned aerial vehicle based on a nearby principle according to the distance matrix, and determining sub-region information corresponding to each meeting point of the truck unmanned aerial vehicle; distributing the subareas to each unmanned aerial vehicle according to the subarea information so as to minimize the sum of squares of the differences of the grid quantity distributed by each unmanned aerial vehicle; according to the acquired truck quantity information, carrying out random distribution on the unmanned aerial vehicles so that all the unmanned aerial vehicles are distributed to different trucks to form a plurality of truck unmanned aerial vehicle distribution schemes; wherein each truck unmanned aerial vehicle distribution scheme comprises a plurality of single-vehicle-multi-machine models; the single-vehicle-multiple-machine model is composed of a truck and at least two unmanned aerial vehicles.

The initial solution determining module 1106 is further configured to perform vehicle path and unmanned aerial vehicle task set planning for the single-multiple machine sub-model through a dynamic planning algorithm and a task allocation algorithm, so as to obtain task planning scheme data corresponding to the single-multiple machine sub-model; according to the task planning scheme data, if the longest vehicle path value of all the single-multi-machine sub-models under the same truck unmanned aerial vehicle allocation scheme does not exceed the preset truck maximum range value, merging task planning scheme data corresponding to a plurality of single-multi-machine sub-models under the same truck unmanned aerial vehicle allocation scheme to obtain a task planning initial solution under the truck unmanned aerial vehicle allocation scheme; otherwise, the truck unmanned aerial vehicle allocation scheme is deleted.

The optimal solution determining module 1108 is further configured to optimize the task planning initial solution under the allocation scheme of each truck unmanned aerial vehicle by using an adaptive large neighborhood search algorithm that merges disturbance and tabu strategies, so as to obtain a task planning optimal solution under the allocation scheme of each truck unmanned aerial vehicle; the adaptive large neighborhood search algorithm integrating disturbance and tabu strategies comprises the following steps:

step 9: if the total iteration number I is greater than or equal to I _max Or T.ltoreq.T _e The algorithm is terminated, and the current optimal solution S is output _best The method comprises the steps of carrying out a first treatment on the surface of the Otherwise, updating the iteration times I and the current temperature T, and returning to the step 4.

For specific limitations of the regional scout mission planning apparatus, reference may be made to the above limitations of the regional scout mission planning method, and no further description is given here. The modules in the regional scout mission planning device can be realized in whole or in part by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a terminal, and the internal structure thereof may be as shown in fig. 12. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of regional scout mission planning. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the structure shown in FIG. 12 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

In an embodiment a computer device is provided comprising a memory storing a computer program and a processor implementing the steps of the method embodiments described above when the computer program is executed.

In one embodiment, a computer readable storage medium is provided, on which a computer program is stored which, when executed by a processor, implements the steps of the method embodiments described above.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.

Claims

1. A regional scout mission planning method, the method comprising:

2. The method of claim 1, wherein constructing a multi-vehicle multi-machine mission planning model based on the regional scout mission information, targeting a minimum time to complete all scout missions, and constraining scout mission requirements, truck and drone configuration information, comprises:

according to the regional scout task information, taking the minimum time for completing all scout tasks as a target, taking scout task requirements, truck and unmanned aerial vehicle configuration information as constraint conditions, and constructing a multi-vehicle multi-machine task planning model as follows:

Min{t|t≥max(t1 _h ,t2 _k ),h∈Tr,k∈D} (1)

t ₀ ＝(T _max -l _D /v _D ) (2)

Wherein t1 _h Indicating the time for the h truck to return to the base, t2 _k The time for the kth unmanned aerial vehicle to return to the base is represented by Tr, the truck set tr= {1,2, …, l }, l represents the number of trucks, D represents the unmanned aerial vehicle set d= {1,2, …, m }, m represents the number of unmanned aerial vehicles, t ₀ Representing the time required for the unmanned aerial vehicle to scan a grid, T _max Represents the maximum endurance time of the unmanned aerial vehicle, l _D Representing a reserved voyage of the unmanned aerial vehicle, v _D Represents the average speed of the drone, N represents the set of truck drone junction points n= {1, …, b }, b represents the number of truck drone junction points,indicating that 1 is present if the h truck is driving from meeting point i to meeting point j, otherwise 0,/is present>Denoted in sub-region s, 1 if the grid number G is divided into the task set of the kth unmanned aerial vehicle, otherwise 0, where k= {1,2 …, m }, G e G, s e a, G denotes the grid set g= {1,2, …, n }, n denotes the number of grids, a denotes the area to be scouted>Representing sub-region A _s The contained grid set s epsilon A, G epsilon G A _S Represents the set of subregions of region a, s= {1,2 …, c }, c represents the number of subregions, R _ig Representing the distance from the meeting point i of the unmanned aerial vehicle of the truck to the center point of the g-number grid, i epsilon N ₀ ∪N,g∈G，SD _max Represents the maximum scanning area of the unmanned aerial vehicle, M _S Representing sub-region A _s Area, z _kh Indicating that if the kth unmanned aerial vehicle is carried by the h truck, it is 1, otherwise it is 0, ltr (s _h ) Represents the path length, STr, of the h truck _max Representing the maximum driving distance of the truck; equation (1) is an objective function, equations (2) - (15) are constraint conditions, constraint condition (2) represents time required for computing a single grid of unmanned aerial vehicle scout, constraint condition (3) represents time required by unmanned aerial vehicle to finish one grid of scout each time, constraint condition (8) represents that the maximum scanning total area of all unmanned aerial vehicle grids in each area is larger than or equal to the area of the area, constraint condition (4) represents that the degree of intersection of each truck unmanned aerial vehicle is equal to the ingress degree, thereby ensuring connectivity of a vehicle running route, constraint condition (5) represents that each truck unmanned aerial vehicle intersection can only be accessed by one truck at most, constraint condition (6) represents that each grid of each sub-area can only be scout by one unmanned aerial vehicle, constraint condition (7) represents that the time required by unmanned aerial vehicle to finish one grid of scout each time is not longer than the maximum endurance time, constraint condition (8) represents that the maximum scanning total area of all unmanned aerial vehicle grids in each area is larger than or equal to the area, constraint condition (9) guarantees that each of the area is completely covered by the area, constraint condition (9) represents that each unmanned aerial vehicle contained in each area is required by the scout, constraint condition (10) is not to be carried by one unmanned aerial vehicle, constraint condition (14) represents that the number of unmanned aerial vehicle (10) is not carried by at least one constraint condition (14) and the number of unmanned aerial vehicle (14) in each sub-area is represented by the constraint condition, the number of unmanned aerial vehicle (10) is not carried by the constraint condition, and the constraint condition (14) represents the number of unmanned aerial vehicle (3) is carried by the constraint condition, and the constraint condition (3) represents the number is, (15) defining the value range of the 0-1 variable.

3. The method of claim 1, wherein decomposing the multi-vehicle multi-machine mission planning model into a plurality of single-vehicle multi-machine sub-models under a plurality of truck drone allocation scheme based on a decomposition strategy comprises:

dividing the region to be detected into unitized grids, and determining grid center point information; the size of the grid is equal to the maximum scanning area of the unmanned aerial vehicle;

constructing a distance matrix between grids and the meeting points of the truck unmanned aerial vehicle according to the meeting point information of the truck unmanned aerial vehicle and the grid center point information, distributing grids to be scanned for each meeting point of the truck unmanned aerial vehicle based on a nearby principle according to the distance matrix, and determining subarea information corresponding to each meeting point of the truck unmanned aerial vehicle;

distributing the subareas to each unmanned aerial vehicle according to the subarea information so as to minimize the sum of squares of the differences of the grid quantity distributed by each unmanned aerial vehicle;

according to the acquired truck quantity information, carrying out random distribution on the unmanned aerial vehicles so that all the unmanned aerial vehicles are distributed to different trucks to form a plurality of truck unmanned aerial vehicle distribution schemes; wherein each truck unmanned aerial vehicle distribution scheme comprises a plurality of single-vehicle-multi-machine models; the single-vehicle-multiple-machine model is composed of a truck and at least two unmanned aerial vehicles.

4. A method according to claim 3, wherein performing mission planning for the single-vehicle-multiple-machine sub-models, respectively, to obtain initial solutions for mission planning under each truck unmanned aerial vehicle allocation scheme, comprises:

respectively planning a vehicle path and an unmanned aerial vehicle task set for the single-multi-machine sub-model through a dynamic planning algorithm and a task allocation algorithm to obtain task planning scheme data corresponding to the single-multi-machine sub-model;

according to the task planning scheme data, if the longest vehicle path value of all the single-multiple machine sub-models in the same truck unmanned aerial vehicle allocation scheme does not exceed a preset truck maximum range value, combining task planning scheme data corresponding to the multiple single-multiple machine sub-models in the same truck unmanned aerial vehicle allocation scheme to obtain a task planning initial solution in the truck unmanned aerial vehicle allocation scheme;

otherwise, deleting the truck unmanned aerial vehicle distribution scheme.

5. The method of claim 4, wherein optimizing the initial solution of the mission plan under each truck unmanned aerial vehicle allocation scheme by an adaptive large neighborhood search algorithm incorporating perturbation and tabu strategies, comprises:

Optimizing the initial solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle through a self-adaptive large neighborhood search algorithm integrating disturbance and tabu strategies to obtain the optimal solution of the task planning under the allocation scheme of each truck unmanned aerial vehicle; the adaptive large neighborhood search algorithm integrating disturbance and tabu strategies comprises the following steps:

Step 5: comparing the total task completion time of the new solution S 'and the current solution S to obtain a time difference, wherein the time difference is represented by delta f, if the new solution meets the constraint, the new solution is accepted, and the new solution S' is assigned to the current solution S; meanwhile, if Deltaf is smaller than 0, representing the new solution can shorten the task completion time, and assigning the new solution S' to the optimal solution S _best The method comprises the steps of carrying out a first treatment on the surface of the If Deltaf>0, judging whether to accept the refund solution according to the Metropolis criterion;

6. The method of claim 5, wherein in the operation of the destruction operator, the destruction operator comprises: random disruption operators, optimal disruption operators, randomly changing truck paths, and changing longest truck paths; in the repair operator operation, the repair operator includes: random repair between the same bicycle-multiple machine model combinations, greedy repair between the same bicycle-multiple machine model combinations, random repair between different bicycle-multiple machine model combinations, and greedy repair between different bicycle-multiple machine model combinations.

7. The method according to any one of claims 1 to 6, wherein the perturbation operation Ω is:

and (3) redistributing the number of the unmanned aerial vehicles carried on each truck, extracting 2 single-vehicle-multi-machine sub-model combinations with the largest total task time difference, distributing one unmanned aerial vehicle in the combination with the shortest total task time to the combination with the longest total task time, and then re-planning unmanned aerial vehicle tasks in different single-vehicle-multi-machine sub-model combinations.

8. An area reconnaissance mission planning apparatus, the apparatus comprising:

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.

10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.