CN107037827B - Unmanned aerial vehicle aerial work task allocation and flight path planning combined optimization method and device - Google Patents

Abstract

The invention relates to a method and a device for joint optimization of task allocation and track planning of aerial work of an unmanned aerial vehicle. The method provided by the invention can automatically obtain the task and the flight path planning of the unmanned aerial vehicle, so that the unmanned aerial vehicle can automatically execute the operation task according to the task and the flight path planning, and the influence of manual operation is avoided. In addition, the unmanned aerial vehicle based on this result execution job task also can be in the operation of limited duration of flight to the polylith farmland and obtain the biggest gross profit and can improve the efficiency of operation effectively.

Description

Unmanned aerial vehicle aerial work task allocation and flight path planning combined optimization method and device

Technical Field

The embodiment of the invention relates to the technical field of unmanned aerial vehicles, in particular to a method and a device for joint optimization of unmanned aerial vehicle aerial work task allocation and flight path planning.

Background

With the continuous deepening of the agricultural mechanization degree, the unmanned aerial vehicle rapidly becomes an important mode in the agricultural operation process with the advantages of high operation efficiency, low labor intensity, low comprehensive cost and the like, and has wide application in agricultural aviation operations such as precision seeding, vegetation detection, pesticide spraying and the like. For example, the germination condition and the weed degree of herbaceous plants can be detected by an unmanned aerial vehicle, or the planthoppers can be controlled by spraying pesticides on rice fields by using the unmanned aerial vehicle, and the like.

Current drones can be broadly classified into two broad categories, multi-rotor (e.g., quad-rotor, six-rotor, or eight-rotor drones, etc.) and fixed-wing. Wherein fixed wing unmanned aerial vehicle is comparatively wide application in agricultural operation with advantages such as long, the area of cruising is big, flying speed is fast, height.

However, in the process of implementing the present invention, the inventor finds that, because the current operation of the fixed-wing unmanned aerial vehicle is mainly performed by artificial remote control, the effect of the actual operation is greatly influenced by the operation level of the operator, and the flight path planned by the artificial and instant looking mode deviates from the theoretical flight path seriously, which leads to the high operation leakage rate and repetition rate of the unmanned aerial vehicle. And when multiple operators operate multiple fixed wing drones, there is also a lack of coordination before each other.

In addition, in the process of carrying out pesticide spraying aviation operation by using the fixed-wing unmanned aerial vehicle, a plurality of factors influencing operation execution are provided, such as the time for executing work, the flight time of the unmanned aerial vehicle and the like. For example, when the working time of a day is divided into two sections, how to spray pesticides on a plurality of farmlands by a plurality of fixed-wing unmanned aerial vehicles under the condition of limited flight time and how to carry out the most reasonable flight path planning makes the total income of the farmlands after spraying maximum (that is, the total sum of pesticide effects of the farmlands completing the spraying task maximum) become a problem to be solved urgently.

Disclosure of Invention

An embodiment of the invention provides a method and a device for joint optimization of mission allocation and flight path planning of unmanned aerial vehicle aerial work, which are used for solving the problems that in the prior art, the flight of the unmanned aerial vehicle is greatly influenced by manual operation, and when the working time of one day is divided into two sections, how to obtain the maximum total profit for operation of a plurality of farmlands by utilizing a plurality of fixed-wing unmanned aerial vehicles within the limited flight time length is high.

In a first aspect, an embodiment of the present invention provides a joint optimization method for mission allocation and flight path planning of unmanned aerial vehicle, where when multiple fixed-wing unmanned aerial vehicles perform a mission on multiple rectangular farmlands, the method includes:

acquiring two time windows for executing the task, information of the fixed-wing unmanned aerial vehicle and farmland information of the candidate farmlands;

coding the time window information, the unmanned aerial vehicle information and the farmland information, and randomly generating a plurality of chromosomes; screening out chromosomes meeting preset constraint conditions defined by a preset MUAV-MTW-DTOP model from the plurality of chromosomes generated randomly, and constructing an initial solution set of the MUAV-MTW-DTOP model; the MUAV-MTW-DTOP model is an objective function which enables the multi-frame fixed-wing unmanned aerial vehicle flying in a preset flying mode to obtain the maximum total benefit in the task; the preset constraints comprise flight time constraints and two time window constraints of each fixed-wing unmanned aerial vehicle;

and continuously crossing and mutating the initial solution set based on a preset genetic algorithm until the initial solution set meets the constraint of the iteration times, ending the crossing and the mutation, selecting the optimal solution which enables the model to obtain the maximum total income from the mutated solution set, and taking the optimal solution as the task allocation and track planning result of the operation.

In a second aspect, a further embodiment of the present invention provides a joint optimization apparatus for mission allocation and flight path planning for unmanned aerial vehicles, which includes, when multiple fixed-wing unmanned aerial vehicles perform a mission on multiple rectangular fields:

the information acquisition unit is used for acquiring two time windows for executing the task, information of the fixed-wing unmanned aerial vehicle and farmland information of the candidate farmlands;

the initial solution acquisition unit is used for encoding the time window information, the information of the unmanned aerial vehicle and the farmland information and randomly generating a plurality of chromosomes; screening out chromosomes meeting preset constraint conditions defined by a preset MUAV-MTW-DTOP model from the plurality of chromosomes generated randomly, and constructing an initial solution set of the MUAV-MTW-DTOP model; the MUAV-MTW-DTOP model is an objective function which enables the multi-frame fixed-wing unmanned aerial vehicle flying in a preset flying mode to obtain the maximum total benefit in the task; the preset constraints comprise flight time constraints and two time window constraints of each fixed-wing unmanned aerial vehicle;

and the optimal solution calculation unit is used for continuously crossing and mutating the initial solution set based on a preset genetic algorithm until the crossing and the mutation are finished after the constraint of the iteration times is met, selecting an optimal solution which enables the model to obtain the maximum total income from the mutated solution set, and taking the optimal solution as a task allocation and flight path planning result of the operation.

The invention provides a joint optimization method for task allocation and flight path planning of unmanned aerial vehicle aerial work, aiming at the condition that a plurality of fixed-wing unmanned aerial vehicles execute a task on a plurality of rectangular farmlands, the method comprises the steps of firstly obtaining two time windows for executing the task, information of the fixed-wing unmanned aerial vehicles and farmland information of a plurality of candidate farmlands, then obtaining an optimal solution which can enable the model to obtain the maximum total income according to the information based on a preset MUAV-MTW-DTOP model and a genetic algorithm, and taking the optimal solution as a task allocation and flight path planning result of the operation. Compared with the existing manual remote control mode, the method provided by the invention can automatically obtain the task and flight path planning of the unmanned aerial vehicle in the operation according to the preset model and algorithm, so that the unmanned aerial vehicle can automatically execute the operation task according to the task and flight path planning, and the influence of manual operation is avoided. In addition, the method provided by the invention takes the optimal solution of the preset farmland maximization profit model as a flight path planning result, so that the unmanned aerial vehicle executing the operation task based on the result can operate a plurality of farmlands within the limited flight time and obtain the maximum total profit, thereby effectively improving the operation efficiency and enabling the operation form of the unmanned aerial vehicle to be applied to wider farmland operation tasks.

Drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:

FIG. 1 is a flowchart of an embodiment of a joint optimization method for task allocation and flight path planning in unmanned aerial vehicle aerial work provided by the invention;

FIGS. 2(a) -2(c) are schematic diagrams of a farmland spraying mode provided by an embodiment of the invention;

fig. 3 is a schematic view of a flight path between farmlands of an unmanned aerial vehicle under a dynamic constraint condition, provided by an embodiment of the invention;

FIG. 4 is a flow chart of a genetic algorithm provided by an embodiment of the present invention;

FIG. 5 is an example of chromosome coding provided by an embodiment of the invention;

FIG. 6 is an example of chromosome crossing provided by an embodiment of the invention;

FIG. 7 is an example of chromosomal variation provided by an embodiment of the invention;

FIG. 8 is a schematic diagram of the distribution of 20 farmland provided by an embodiment of the invention;

FIG. 9 is a schematic diagram of the convergence of the optimal solution provided by the embodiment of the present invention;

fig. 10 is a schematic structural diagram of an embodiment of an apparatus for joint optimization of unmanned aerial vehicle aerial work task allocation and flight path planning, provided by the embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In a first aspect, an embodiment of the present invention provides a method for joint optimization of mission allocation and flight path planning for unmanned aerial vehicle aerial work, where when multiple fixed-wing unmanned aerial vehicles perform a mission on multiple rectangular farmlands, as shown in fig. 1, the method includes:

s101, obtaining two time windows for executing the task, information of the fixed-wing unmanned aerial vehicle and farmland information of the candidate farmlands;

s102, coding the time window information, the unmanned aerial vehicle information and the farmland information, and randomly generating a plurality of chromosomes; screening out chromosomes meeting preset constraint conditions defined by a preset MUAV-MTW-DTOP model from the plurality of chromosomes generated randomly, and constructing an initial solution set of the MUAV-MTW-DTOP model; the MUAV-MTW-DTOP model is an objective function which enables the multi-frame fixed-wing unmanned aerial vehicle flying in a preset flying mode to obtain the maximum total benefit in the task; the preset constraints comprise flight time constraints and two time window constraints of each fixed-wing unmanned aerial vehicle;

s103, continuously crossing and mutating the initial solution set based on a preset genetic algorithm until the initial solution set meets the constraint of iteration times, ending crossing and mutating, selecting an optimal solution which enables the model to obtain the maximum total income from the mutated solution set, and taking the optimal solution as a task allocation and flight path planning result of the operation.

The invention provides a joint optimization method for task allocation and flight path planning of unmanned aerial vehicle aerial work, aiming at the condition that a plurality of fixed-wing unmanned aerial vehicles execute a task on a plurality of rectangular farmlands, the method comprises the steps of firstly obtaining two time windows for executing the task, information of the fixed-wing unmanned aerial vehicles and farmland information of a plurality of candidate farmlands, then obtaining an optimal solution which can enable the model to obtain the maximum total income according to the information based on a preset MUAV-MTW-DTOP model and a genetic algorithm, and taking the optimal solution as a task allocation and flight path planning result of the operation. Compared with the existing manual remote control mode, the method provided by the invention can automatically obtain the task and flight path planning of the unmanned aerial vehicle in the operation according to the preset model and algorithm, so that the unmanned aerial vehicle can automatically execute the operation task according to the task and flight path planning, and the influence of manual operation is avoided. In addition, the method provided by the invention takes the optimal solution of the preset farmland maximization profit model as a flight path planning result, so that the unmanned aerial vehicle executing the operation task based on the result can operate a plurality of farmlands within the limited flight time and obtain the maximum total profit, thereby effectively improving the operation efficiency and enabling the operation form of the unmanned aerial vehicle to be applied to wider farmland operation tasks.

In practical implementation, it can be understood that the objective function and the constraint condition included in the MUAV-MTW-DTOP model in the above method are important bases for obtaining the optimal planning result of the present invention, and may be set in various ways, and one of the alternative setting ways is described in detail below.

(1) Unmanned plane

Denoted by U is N for executing a task to be sprayed_USet of isomorphic drones

Each unmanned aerial vehicle can only carry one pesticide; during flight, all drones have the same minimum turning radius R_UAnd a flying speed V, and all carry a spraying radius R_DThe spray head of (1).

Considering the characteristic that the unmanned aerial vehicle executes the pesticide spraying task, the embodiment of the invention makes the following assumptions:

(a) the unmanned aerial vehicles have the capability of automatically avoiding obstacles, and can adopt an autonomous avoidance control strategy under the condition of collision, so that the generated path deviation is very small and negligible relative to the total flight path length;

(b) the unmanned aerial vehicles all fly at the same cruising speed and cruising height, so that the influence of the factors on the spraying effect is not considered;

(c) influence of external environment on the flight track of the unmanned aerial vehicle is not considered in the flight process of the unmanned aerial vehicle;

(d) the unmanned aerial vehicle can carry pesticides required by executing tasks in the flight process, but the fuel is limited;

(2) farmland

Is provided with

The starting point and the end point of the unmanned aerial vehicle are respectively, and the starting point and the end point are the same in the embodiment of the invention;

for N to be sprayed with pesticides_ARectangular farmland, and farmland A_iIs side length of L_ix,L_iyThe vertex coordinate is (A)_i1,A_i2,A_i3,A_i4) Area is S_iA rectangle of (2); the set of the starting point, the terminal point and the farmland area of the unmanned aerial vehicle is

When unmanned plane U_uFor farmland A_iWhen the pesticide is sprayed In the coverage mode, the entry point of the unmanned aerial vehicle flying into the farmland is In_iuThe departure point of the flying away from the farmland is Out_iuAnd assume that the drone must completely spray the entire field before leaving. Meanwhile, the pesticide can be sprayed on each farmland only once at most.

(3) Flight path of preset flight mode

In the process of executing agricultural aviation operation tasks by the unmanned aerial vehicle, the agricultural operation tasks are required to be completed by spraying the agricultural material in a covering mode in farmlands, and the unmanned aerial vehicle also needs to fly among different farmlands to realize switching among the tasks, so that two types of flight paths are generated, namely flight paths in farmlands and in farmlands.

Flight path inside the field: in farmland A_iInside, unmanned aerial vehicle flies under the dynamics constraint and will have turning radius to use the parallel scanning strategy to carry out the overlay pesticide and spray. In the process, the unmanned plane U_uFrom farmland A_iIn (2) of_iuPoint entering, the path after entering the farmland is parallel to a certain edge of the farmland, and then from Out_iuPoint away, at this moment, unmanned plane U_uThe spent time for spraying the farmland is t_iwu。

The parallel scanning strategy is an optimal strategy for the unmanned aerial vehicle to carry out coverage type pesticide spraying on a rectangular farmland area, is also the most conveniently realized strategy in control, and is widely applied to the execution process of an area coverage task. This strategy has two common implementations, namely flight parallel to the short sides of the rectangle and flight parallel to the long sides of the matrix. For example, there are two ways to overlay scan the rectangular area in fig. 2(a) using a parallel scanning strategy, with the flight trajectory of the drone parallel to the short side of the rectangle (as shown in fig. 2 (b)) and the flight trajectory of the drone parallel to the long side of the rectangle (as shown in fig. 2 (c)). Meanwhile, the position of the unmanned aerial vehicle entering the rectangular area and the actual turning radius of the unmanned aerial vehicle on the rectangular boundary can influence the length of the flight path of the unmanned aerial vehicle in the farmland at the same time. Although the unmanned aerial vehicle can enter from any point of the boundary of the farmland, the entry point when the unmanned aerial vehicle covers the shortest flight path of the rectangular area is that the distance from the vertex of the farmland is R_DPoint (2) of (c). Therefore, the embodiment of the invention discretizes the entry point of the unmanned aerial vehicle into the rectangular area, and selects the distance R between the 4 vertexes of the rectangle on the long side and the short side of the rectangle _D8 points of (1) { R ]_D1,R_D2...R_D8As an entry point for the drone into the rectangular area. For a rectangular area, the shortest path for a drone to fly inside the area must enter the rectangular area from one of the 8 entry points, and the exit point Out of the rectangular area for the drone to exit on this shortest path_iuI.e. uniquely determined.

Flight path between farmlands:

the unmanned aerial vehicle finishes the execution of the farmland A_iAfter the pesticide spraying task, if the farmland A needs to be sprayed_jSpraying pesticide, then the unmanned aerial vehicle must also be in two farmlands A_i,A_jFly along the Dubins path. The starting point of the path is unmanned plane U_uIn farmland A_iOut of_iuThe termination point is the unmanned plane U_uIn farmland A_jIn (b) at the entry point In_juThe time taken by the Dubins path is denoted as t_ijwu。

According to the generation principle of the Dubins path, the shortest Dubins path between two points can be generated by combining an arc-segment path and a straight-segment path, and there are six cases where D ═ { LSL, RSR, RSL, LSR, RLR, LRL }. Wherein, L represents a section of arc that unmanned aerial vehicle turned to the left, and R represents a section of arc that unmanned aerial vehicle turned to the right, and S represents that unmanned aerial vehicle flies with the straight line mode.

For example, fig. 3 depicts a flight path between fields for a drone under dynamic constraints. According to the generation principle of the Dubins path, the unmanned plane follows A of the rectangular area_iThe point starts to turn right and fly along an arc path (R), then fly along a straight path (S), and finally turn left and fly along an arc path (L) to reach A of a rectangular area_jThe path of the point. Because a parallel scanning strategy is adopted in the farmland, the angle of the unmanned aerial vehicle entering or leaving a rectangular farmland area is necessarily the boundary vertical to the farmland, so that in the optimization process of the shortest Dubins path, the starting angle of the starting point and the entering angle of the ending point of the unmanned aerial vehicle are determined, and the changing factor is that the U of the unmanned aerial vehicle is a U_uIn farmland A_iOut of_iuAnd field A_jIn (b) at the entry point In_ju

In order to complete the task allocation problem, the embodiment of the invention selects all the profits of all the unmanned aerial vehicles when completing the task as the maximized benefit function.

The objective function of the MUAV-MTW-DTOP model is as follows:

the constraint conditions of the MUAV-MTW-DTOP model are as follows:

wherein N is_UFor unmanned plane U_uThe number of the cells; n is a radical of_AFor farmland A_iThe number of the cells; w is the w-th time window, and is 1 or 2; n th₀,N_A+1The block farmland represents the starting point and the end point of the unmanned aerial vehicle; SQ_iIndicating farmland A_iArea (unit hectare); p_iwIndicating that field A is completed in the w-th time window_iThe benefits of the task, i.e., the profits; s_iwuRefers to unmanned plane U_uFor farmland A in the w time window_iThe time to start spraying the pesticide; t is t_ijwuRefers to unmanned plane U_uFor farmland A in the w time window_i,A_jThe time of flying according to a preset flying mode; s_jwuRefers to unmanned plane U_uFor farmland A in the w time window_jThe time to start spraying the pesticide; m is a preset value; o is_iwu,C_iwuAre respectively unmanned aerial vehicle U_uFor farmland A in the w time window_iA start time and an end time at which a task can be executed; t is t_iwuRefers to unmanned plane U_uFor farmland A in the w time window_iThe time for executing the task according to a preset flight mode; e_wuRepresenting the maximum flight duration of the drone. If x_iwuIndicate unmanned plane U as 1_uTo farmland A in the w time window_iComplete the drug spraying task, otherwise the unmanned aerial vehicle does notIs directed to farmland A_iExecuting the task; if y_ijwuIndicate unmanned plane U as 1_uThrough farmland A_i,A_jOtherwise, the unmanned aerial vehicle does not pass through the farmland A_i,A_j；

And the target function formula (1) is used for maximizing the total income of the farmland after the pesticide spraying task is completed. Constraint (2) is to ensure that all routes, i.e. all drones, start at a₀End point is

In this disclosure, the coordinates of the starting point and the ending point are the same, and the number of routes is the sum of the routes of the unmanned aerial vehicle in two time windows. Constraint (3) allows each edge to have connectivity. The constraint equation (4) is to explain the service time required for accessing the farmland. Constraint (5) ensures that each field is visited at most once. Constraint (6) entails that the time to perform the task must be within two time windows and that the drone may not be within a time window from the starting point to the field and from the field back to the starting point. Constraint (7) translates the limit on fuel required by the drone into a limit on the flight time of each drone. Constraint (8) is the definition of variables such as target, path, etc.

It is understood that, after the MUAV-MTW-DTOP model is obtained, the method provided by the embodiment of the present invention can solve the optimal solution of the MUAV-MTW-DTOP model according to a preset genetic algorithm. The predetermined genetic algorithm for finding the optimal solution can be implemented by various methods, and one of the alternative ways is described in detail below.

The preset genetic algorithm is used for solving the problem of joint optimization of the unmanned aerial vehicle aviation operation task allocation and flight path planning, and the feasible solution is a chromosome. The population is composed of a plurality of chromosomes, the scale of the population is defined according to problems, the population is updated through the crossing and variation of the chromosomes to form a new population, wherein the crossing means that two parent chromosomes form two new child chromosomes according to the crossing probability, the variation means that one chromosome forms one new chromosome according to the variation probability, and the loop iteration is continuously carried out, so that the optimal child, namely the optimal solution, is selected. The specific flow is shown in fig. 4. The coding, interleaving, mutation, fitness function rules involved in fig. 4 are as follows:

(1) encoding

The MDD coding method comprises coding of time information of a target area farmland, an unmanned aerial vehicle, a target area entry point and an unmanned aerial vehicle accessing a first farmland in each route. And the first line of the coded chromosome is a target area, the second line is an unmanned aerial vehicle for executing tasks, the third line is an entry point when the unmanned aerial vehicle executes the tasks on the target area, and the fourth line is the time for the unmanned aerial vehicle to visit the first farmland in each route. Wherein the target region belongs to the set {1, 2.. N_AN, the drone belongs to the set {0,1_UAnd f, enabling the entry point of the farmland in the target area to belong to a set {1, 2.. 8}, and enabling the time for accessing the first farmland in each route to belong to any time window.

For example, in FIG. 5, there are four paths for the chromosome in two time windows, namely, the UAV U₁And unmanned plane U₂The represented route. At [9:00:00,12:00:00]In the time window, unmanned plane U₁First at time 9.926 from R_D6Point-entering farmland A₂From R to R_D7Point-entering farmland A₆Finally returning to the starting point A₀(ii) a Unmanned plane U₂First at time 9.071 from R_D2Point-entering farmland A₇From R to R_D1Point-entering farmland A₁Finally returning to the starting point A₀(ii) a At [14:00:00,17:00:00]In the time window, unmanned plane U₁From R at time 15.57_D3Point-entering farmland A₅And then returns to the starting point a₀(ii) a Unmanned plane U₂From R at time 15.53_D4Point-entering farmland A₄And then returns to the starting point a₀(ii) a And object A₃Has not been accessed.

(2) Fitness function and selection

The fitness of a chromosome is the total profit for all visited target regions, which is related to the profit variation function of the target region, i.e. to the visited region, the area of the region. Fitness can be obtained by equation (1) and Fitness ═ Profit. The selection of chromosomes is made by roulette, wherein the likelihood of each chromosome being selected is the same.

(3) Crossing

Firstly, randomly determining the crossing position, crossing the genes in the two chromosomes of the parent, and because the fourth row of the genes represents the starting time of the unmanned aerial vehicle accessing the first farmland in the same route, in the crossed offspring chromosomes, replacing the fourth rows of other genes related to the same route with new time.

For example, in fig. 6, parents a and B are two chromosomes to be crossed selected randomly, the crossing position 7 is randomly determined in parent a, and then the seventh column of parent a is replaced with the 4 th column of parent B, for parent a, the unmanned plane U is used in the second time window₁The time of the fifth column and the fourth row is replaced by 16.75, so that the time of the fifth column and the fourth row is also replaced by 16.75 to keep the time consistency of the first farmland accessed in the same route in the same time, thereby obtaining the offspring A; similarly, in the parent B, 9.926 is required to replace the time in the fourth row of the third column to obtain the child B.

(4) Variation of

The mutation in the genetic algorithm is to prevent the genetic algorithm from falling into local optima, so that the genetic algorithm has the possibility of gene mutation, the mutation can be one gene or a plurality of genes, and the chromosome mutation in the embodiment of the invention mainly has the following conditions: the sequence variation of the farmland, the variation of the unmanned aerial vehicle, the entry point variation of the target area, and the time variation of the unmanned aerial vehicle accessing the first farmland in each route.

For example, as shown in fig. 7, when the chromosome a is mutated by the drone, the seventh row of the chromosome a changes from 1 to 2, and as a result, the field 6 is originally transformed by the drone U₁Access change to drone U₂And (6) accessing. Simultaneously in order to guarantee unmanned aerial vehicle U₁The start time of the first farmland visit in the first time window is consistent, and the seventh column and the fourth row of the chromosome A are updated to 9.071 on the basis of the variation

To show the superiority of the method provided by the embodiment of the present invention, how to solve the MUAV-MTW-DTOP model by using the genetic algorithm according to the above function setting, and thus obtain the final task allocation result, will be described in detail below with reference to several specific examples.

Specifically, the genetic algorithm is used for solving the MUAV-MTW-DTOP model in the environment of MATLAB 2013, and experiments are carried out.

Suppose that two unmanned aerial vehicles spray pesticides on 20 farmland areas in two time windows [9,12], [14,17], and use the genetic algorithm to obtain a distribution scheme, wherein the crossover probability of the genetic algorithm is taken to be 0.9, the variation probability is 0.5, the population scale is 600, and the iteration number is 100. The specific parameters involved in the experimental procedure are described below:

(1) unmanned plane

In the experiment of the embodiment of the invention, the specific configuration of the unmanned aerial vehicle is shown in table 1, the speed of the unmanned aerial vehicle is 4m/s, the maximum spraying radius is 5m, and the maximum duration is 2600 s. And the course angles of the unmanned aerial vehicle at the starting point and the ending point are both 0.

Table 1 basic parameter configuration table for unmanned aerial vehicle

Unmanned aerial vehicle parameters	A0\AN+1	V	RD	Ewu
					Unmanned aerial vehicle information	(0,0)	4m/s	5m	2600s

(2) Farmland area and income thereof

This field area had 20 fields to be sprayed, as shown in detail in fig. 8. Specific farmland coordinates and earnings are shown in table 2.

TABLE 2 Farmland coordinate information and yield

The optimal solution obtained by using the genetic algorithm for the above scenario has the benefit of 15.601, and has converged at the 38 th generation, and the convergence rate is higher, as shown in fig. 9. One of the optimal allocation schemes is shown in table 3, where in all fields, fields 3,9,12,20 are not being visited and other fields have been completed with spraying within the time window.

TABLE 3 optimal solution Table

In a second aspect, an embodiment of the present invention further provides an apparatus for joint optimization of mission allocation and flight path planning for unmanned aerial vehicle aviation operation, wherein when a plurality of fixed-wing unmanned aerial vehicles perform a mission on a plurality of rectangular fields, the apparatus comprises:

an information obtaining unit 201, configured to obtain two time windows for executing the task, information of the fixed-wing drone, and farmland information of the multiple candidate farmlands;

an initial solution obtaining unit 202, configured to encode the time window information, the information of the unmanned aerial vehicle, and the farmland information, and randomly generate a plurality of chromosomes; screening out chromosomes meeting preset constraint conditions defined by a preset MUAV-MTW-DTOP model from the plurality of chromosomes generated randomly, and constructing an initial solution set of the MUAV-MTW-DTOP model; the MUAV-MTW-DTOP model is an objective function which enables the multi-frame fixed-wing unmanned aerial vehicle flying in a preset flying mode to obtain the maximum total benefit in the task; the preset constraints comprise flight time constraints and two time window constraints of each fixed-wing unmanned aerial vehicle;

and the optimal solution calculation unit 203 is configured to perform continuous crossing and variation on the initial solution set based on a preset genetic algorithm until the crossing and variation are finished after the constraint of the iteration times is satisfied, select an optimal solution which enables the model to obtain the maximum total yield from the varied solution set, and use the optimal solution as a task allocation and track planning result of the operation.

In specific implementation, the objective function of the MUAV-MTW-DTOP model is as follows:

the constraint conditions of the MUAV-MTW-DTOP model are as follows:

wherein N is_UFor unmanned plane U_uThe number of the cells; n is a radical of_AFor farmland A_iThe number of the cells; w is the w-th time window, and is 1 or 2; n th₀,N_A+1The block farmland represents the starting point and the end point of the unmanned aerial vehicle; SQ_iIndicating farmland A_iThe area of (d); p_iwIndicating that field A is completed in the w-th time window_iThe benefit of the task; s_iwuRefers to unmanned plane U_uFor farmland A in the w time window_iThe time to start spraying the pesticide; t is t_ijwuRefers to unmanned plane U_uFor farmland A in the w time window_i,A_jThe time of flying according to a preset flying mode; s_jwuRefers to unmanned plane U_uFor farmland A in the w time window_jThe time to start spraying the pesticide; m is a preset value; o is_iwu,C_iwuAre respectively unmanned aerial vehicle U_uFor farmland A in the w time window_iA start time and an end time at which a task can be executed; t is t_iwuRefers to unmanned plane U_uFor farmland A in the w time window_iThe time for executing the task according to a preset flight mode; e_wuRepresenting the maximum flight time of the unmanned aerial vehicle; if x_iwu1, then indicate unmanned plane U_uTo farmland A in the w time window_iThe medicine spraying task is completed, otherwise the unmanned aerial vehicle does not spray the farmland A_iExecuting the task; if y_ijwu1, then indicate unmanned plane U_uThrough farmland A_i,A_jOtherwise, the unmanned aerial vehicle does not pass through the farmland A_i,A_j。

In specific implementation, the preset flight mode is as follows:

the method comprises the following steps of flying in a farmland according to a parallel scanning flying mode: entering the farmland from a first entry point on a first side in a direction perpendicular to the first side of the farmland, wherein the distance between the first entry point and the nearest vertex of the farmland is the scanning radius of the unmanned aerial vehicle, and the first side is any side of the region to be detected; when the unmanned aerial vehicle needs to turn, the unmanned aerial vehicle turns with the minimum turning radius larger than or equal to that of the unmanned aerial vehicle;

flying between farmlands according to a Dubins path flying mode, wherein the Dubins path flying mode is that the Dubins path flying mode flies according to a path formed by combining an arc path and a straight path, and the path formed by combining the arc path and the straight path comprises { LSL, RSR, RSL, LSR, RLR and LRL }; wherein, L represents the arc section route that unmanned aerial vehicle turned to the left, and R represents the arc section route that unmanned aerial vehicle turned to the right, and S represents that unmanned aerial vehicle flies with the straight line mode.

In a specific implementation, the initial solution obtaining unit is further configured to:

encoding the solution of the MUAV-MTW-DTOP model into a chromosome composed according to a preset structure by adopting an MMD encoding method to form an initial solution; the chromosome comprises information of a target area farmland, an unmanned aerial vehicle, a target area entry point and time for the unmanned aerial vehicle to access the first farmland in each route; wherein the target area farmland belongs to a set {1,2_AN, the drone belonging to the set {0,1_UAn entry point of the target region farmland belongs to a set {1, 2.. 8}, and the time for accessing the first farmland in each route belongs to a time window;

the first action of the chromosome is the random full arrangement of the target area farmlands, the second action is the random combination of unmanned aerial vehicles for executing tasks, the third action is the random combination of entry points when the unmanned aerial vehicles execute the tasks on the target area, the fourth action is the random combination of the time for the unmanned aerial vehicles to visit the first farmlands in each route, and the starting time for the same unmanned aerial vehicle to visit the first farmlands in each route is the same.

In specific implementation, the optimal solution calculating unit is further configured to perform the following steps:

step 1, generating an initial solution by using the MMD coding method, judging whether the initial solution meets the constraint condition of the MUAV-MTW-DTOP model, if not, continuously generating the initial solution meeting the constraint condition of the MUAV-MTW-DTOP model, generating a parent population with a preset scale and calculating the fitness of the parent population;

step 2, selecting two individuals (A, B) in a parent population to cross by using a roulette method, wherein the crossing rule is that the crossing position in the individual A is randomly selected, then the gene in the individual B, which is the same as the first line of the crossing position of the individual A, is searched, the gene at the crossing position in the chromosomes A and B is replaced, all genes, which are newly exchanged in the chromosomes A and B and are the same as the second line of the genes and the fourth line of the genes is in the same time window, are searched, and the fourth line of the searched gene is replaced by the fourth line of the newly exchanged genes in the chromosomes A and B to obtain chromosomes C and D; judging whether chromosomes C and D meet the constraint conditions of the MUAV-MTW-DTOP model, if so, replacing the chromosomes A and B in the population by using the chromosomes C and D, otherwise, ending, and continuously iterating and updating the population in the step 1 to obtain a new offspring population;

and 3, selecting one chromosome in the population in the step 2 for mutation by using a roulette method, wherein the method for mutating the chromosome is at least one of the following mutation methods, and the method comprises the following steps: carrying out random full arrangement on the first row sequence of the chromosomes; and carrying out variation on the unmanned aerial vehicle information of the second line of the chromosome, wherein the variation range is {0, 1.. N_U}; and mutating an entry point of a target region farmland of a third row of chromosomes, wherein the entry point belongs to a set {1, 2.. 8 }; and a fourth temporal variation of the chromosome belonging to two time windows;

the whole chromosome variation steps include: firstly, if the first row sequence of the chromosome is mutated, a farmland sequence full array is randomly generated; selecting whether the second row is mutated or not and the position of the mutation, and if so, randomly generating a mutated unmanned aerial vehicle to replace the original position; randomly selecting whether the third row has variation and a variation position, and if the variation exists, randomly generating a variation entry point to replace the original position; randomly selecting whether the fourth line is mutated or not and a mutation position, randomly generating mutation time if the fourth line is mutated, replacing the original position, finding a gene which is the same as the second line of the mutation position and in the same time window as the fourth line, mutating the fourth line into new mutation time, and finally obtaining a new mutation chromosome;

judging whether the varied chromosomes meet the constraint conditions of the MUAV-MTW-DTOP model or not, if so, replacing the chromosomes in the population, otherwise, finishing the replacement and continuously iterating and updating the population in the step 2 to obtain a new offspring population;

step 4, calculating the population fitness of the offspring in the step 3 and selecting the optimal solution of all solutions in the iteration;

step 5, judging whether the current iteration times reach a preset value or not, if not, combining the child population and the parent population in the step 3 according to a certain proportion to form a new parent population, and returning to the step 2; if so, ending the iteration, and taking the finally obtained optimal solution as a task allocation and track planning result of the unmanned aerial vehicle.

Since the device for the joint optimization of the assignment of the unmanned aerial vehicle aerial work task and the flight path planning, which is introduced in this embodiment, is a device capable of executing the method for the joint optimization of the assignment of the unmanned aerial vehicle aerial work task and the flight path planning, which is introduced in this embodiment of the present invention, based on the method for the joint optimization of the assignment of the unmanned aerial vehicle aerial work task and the flight path planning, those skilled in the art can understand the specific implementation manner and various variations of the device for the joint optimization of the assignment of the unmanned aerial vehicle aerial work task and the flight path planning, and therefore, how to implement the method for the joint optimization of the assignment of the unmanned aerial vehicle aerial work task and the flight path planning, which is introduced in this embodiment, is not described in detail herein. As long as the person skilled in the art implements the device adopted by the method for joint optimization of unmanned aerial vehicle aerial work task allocation and flight path planning in the embodiment of the present invention, the device belongs to the protection scope of the present application.

In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.

Some component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some or all of the components of a gateway, proxy server, system according to embodiments of the present invention. The present invention may also be embodied as apparatus or device programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing the present invention may be stored on computer-readable media or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.

Claims (8)

1. A joint optimization method for task allocation and flight path planning of unmanned aerial vehicle aerial work is characterized in that when a plurality of fixed-wing unmanned aerial vehicles execute a task on a plurality of rectangular farmlands, the method comprises the following steps:

acquiring two time windows for executing the task, information of the fixed-wing unmanned aerial vehicle and farmland information of a plurality of candidate farmlands;

coding the time window information, the unmanned aerial vehicle information and the farmland information, and randomly generating a plurality of chromosomes; screening out chromosomes meeting preset constraint conditions defined by a preset MUAV-MTW-DTOP model from the plurality of chromosomes generated randomly, and constructing an initial solution set of the MUAV-MTW-DTOP model; the MUAV-MTW-DTOP model is an objective function which enables the multi-frame fixed-wing unmanned aerial vehicle flying in a preset flying mode to obtain the maximum total benefit in the task; the preset constraints comprise flight time constraints and two time window constraints of each fixed-wing unmanned aerial vehicle;

continuously crossing and mutating the initial solution set based on a preset genetic algorithm until the initial solution set meets the constraint of iteration times, ending the crossing and mutation, selecting an optimal solution which enables the model to obtain the maximum total income from the mutated solution set, and taking the optimal solution as a task allocation and track planning result of the operation;

wherein the objective function of the MUAV-MTW-DTOP model is as follows:

the constraint conditions of the MUAV-MTW-DTOP model are as follows:

wherein N is_UFor unmanned plane U_uThe number of the cells; n is a radical of_AFor farmland A_iThe number of the cells; w is the w-th time window, and is 1 or 2; n th₀,N_A+1The block farmland represents the starting point and the end point of the unmanned aerial vehicle; SQ_iIndicating farmland A_iThe area of (d); p_iwIndicating that field A is completed in the w-th time window_iThe benefit of the task; s_iwuRefers to unmanned plane U_uFor farmland A in the w time window_iThe time to start spraying the pesticide; t is t_ijwuRefers to unmanned plane U_uFor farmland A in the w time window_i,A_jThe time of flying according to a preset flying mode; s_jwuRefers to unmanned plane U_uFor farmland A in the w time window_jThe time to start spraying the pesticide; m is a preset value; o is_iwu,C_iwuAre respectively unmanned aerial vehicle U_uFor farmland A in the w time window_iA start time and an end time for executing the task; t is t_iwuRefers to unmanned plane U_uFor farmland A in the w time window_iThe time for executing the task according to a preset flight mode; e_wuRepresenting the maximum flight time of the unmanned aerial vehicle; if x_iwu1, then indicate unmanned plane U_uTo farmland A in the w time window_iThe medicine spraying task is completed, otherwise the unmanned aerial vehicle does not spray the farmland A_iExecuting the task; if y_ijwu1, then indicate unmanned plane U_uThrough farmland A_i,A_jOtherwise the nobodyThe machine has not passed through the farmland A_i,A_j。

2. The method according to claim 1, characterized in that the predetermined flight pattern is:

the method comprises the following steps of flying in a farmland according to a parallel scanning flying mode: entering the farmland from a first entry point on a first side in a direction perpendicular to the first side of the farmland, wherein the distance between the first entry point and the nearest vertex of the farmland is the scanning radius of the unmanned aerial vehicle, and the first side is any side of the region to be detected; when the unmanned aerial vehicle needs to turn, the unmanned aerial vehicle turns with the minimum turning radius larger than or equal to that of the unmanned aerial vehicle;

flying between farmlands according to a Dubins path flying mode, wherein the Dubins path flying mode is that the Dubins path flying mode flies according to a path formed by combining an arc path and a straight path, and the path formed by combining the arc path and the straight path comprises { LSL, RSR, RSL, LSR, RLR and LRL }; wherein, L represents the arc section route that unmanned aerial vehicle turned to the left, and R represents the arc section route that unmanned aerial vehicle turned to the right, and S represents that unmanned aerial vehicle flies with the straight line mode.

3. The method of claim 1, wherein the step of selecting chromosomes satisfying the predetermined constraints defined by the predetermined MUAV-MTW-DTOP model from among the randomly generated chromosomes, and the step of constructing the initial solution set of the MUAV-MTW-DTOP model comprises:

encoding the solution of the MUAV-MTW-DTOP model into a chromosome composed according to a preset structure by adopting an MMD encoding method to form an initial solution; the chromosome comprises information of a target area farmland, an unmanned aerial vehicle, a target area entry point and time for the unmanned aerial vehicle to access the first farmland in each route; wherein the target area farmland belongs to a set {1,2_AN, the drone belonging to the set {0,1_UAn entry point of the target region farmland belongs to a set {1, 2.. 8}, and the time for accessing the first farmland in each route belongs to a time window;

the first action of the chromosome is the random full arrangement of the target area farmlands, the second action is the random combination of unmanned aerial vehicles for executing tasks, the third action is the random combination of entry points when the unmanned aerial vehicles execute the tasks on the target area, the fourth action is the random combination of the time for the unmanned aerial vehicles to visit the first farmlands in each route, and the starting time for the same unmanned aerial vehicle to visit the first farmlands in each route is the same.

4. The method according to claim 3, wherein the initial solution set is continuously crossed and varied based on a preset genetic algorithm until the crossing and variation are finished after the constraint of the iteration times is satisfied, an optimal solution which enables the model to obtain the maximum total income is selected from the varied solution set, and the optimal solution is used as a task allocation and flight path planning result of the operation, and the method comprises the following steps:

step 1, generating an initial solution by using the MMD coding method, judging whether the initial solution meets the constraint condition of the MUAV-MTW-DTOP model, if not, continuously generating the initial solution meeting the constraint condition of the MUAV-MTW-DTOP model, generating a parent population with a preset scale and calculating the fitness of the parent population;

step 2, selecting an individual A and an individual B in a parent population to be crossed by using a roulette method, wherein the crossing rule is that the crossing position in the individual A is randomly selected, then the gene in the individual B, which is the same as the first line of the crossing position of the individual A, is searched, the gene at the crossing position in the chromosomes A and B is replaced, all genes, which are newly exchanged in the chromosomes A and B and are the same as the second line of the genes and the fourth line of the genes is in the same time window, are searched, and the fourth line of the searched genes is replaced by the fourth line of the newly exchanged genes in the chromosomes A and B to obtain chromosomes C and D; judging whether chromosomes C and D meet the constraint conditions of the MUAV-MTW-DTOP model, if so, replacing the chromosomes A and B in the population by using the chromosomes C and D, otherwise, ending, and continuously iterating and updating the population in the step 1 to obtain a new offspring population;

and 3, selecting one chromosome in the population in the step 2 for mutation by using a roulette method, wherein the mutation mode of the chromosome is at least one of the following mutation modesThe method comprises the following steps: carrying out random full arrangement on the first row sequence of the chromosomes; and carrying out variation on the unmanned aerial vehicle information of the second line of the chromosome, wherein the variation range is {0, 1.. N_U}; and mutating an entry point of a target region farmland of a third row of chromosomes, wherein the entry point belongs to a set {1, 2.. 8 }; and a fourth temporal variation of the chromosome belonging to two time windows;

the whole chromosome variation steps include: firstly, if the first row sequence of the chromosome is mutated, a farmland sequence full array is randomly generated; selecting whether the second row is mutated or not and the position of the mutation, and if so, randomly generating a mutated unmanned aerial vehicle to replace the original position; randomly selecting whether the third row has variation and a variation position, and if the variation exists, randomly generating a variation entry point to replace the original position; randomly selecting whether the fourth line is mutated or not and a mutation position, randomly generating mutation time if the fourth line is mutated, replacing the original position, finding a gene which is the same as the second line of the mutation position and in the same time window as the fourth line, mutating the fourth line into new mutation time, and finally obtaining a new mutation chromosome;

judging whether the varied chromosomes meet the constraint conditions of the MUAV-MTW-DTOP model or not, if so, replacing the chromosomes in the population, otherwise, finishing the replacement and continuously iterating and updating the population in the step 2 to obtain a new offspring population;

step 4, calculating the population fitness of the offspring in the step 3 and selecting the optimal solution of all solutions in the iteration;

step 5, judging whether the current iteration times reach a preset value or not, if not, combining the child population and the parent population in the step 3 according to a certain proportion to form a new parent population, and returning to the step 2; if so, ending the iteration, and taking the finally obtained optimal solution as a task allocation and track planning result of the unmanned aerial vehicle.

5. The utility model provides an unmanned aerial vehicle air work task allocation and flight path planning unite optimization device which characterized in that, when many fixed wing unmanned aerial vehicles carry out a task to polylith rectangle farmland, the device includes:

the information acquisition unit is used for acquiring two time windows for executing the task, information of the fixed-wing unmanned aerial vehicle and farmland information of a plurality of candidate farmlands;

the initial solution acquisition unit is used for encoding the time window information, the information of the unmanned aerial vehicle and the farmland information and randomly generating a plurality of chromosomes; screening out chromosomes meeting preset constraint conditions defined by a preset MUAV-MTW-DTOP model from the plurality of chromosomes generated randomly, and constructing an initial solution set of the MUAV-MTW-DTOP model; the MUAV-MTW-DTOP model is an objective function which enables the multi-frame fixed-wing unmanned aerial vehicle flying in a preset flying mode to obtain the maximum total benefit in the task; the preset constraints comprise flight time constraints and two time window constraints of each fixed-wing unmanned aerial vehicle;

the optimal solution calculation unit is used for continuously crossing and mutating the initial solution set based on a preset genetic algorithm until the initial solution set meets the constraint of iteration times, ending the crossing and the mutation, selecting an optimal solution which enables the model to obtain the maximum total income from the mutated solution set, and taking the optimal solution as a task allocation and flight path planning result of the operation;

wherein the objective function of the MUAV-MTW-DTOP model is as follows:

the constraint conditions of the MUAV-MTW-DTOP model are as follows:

wherein N is_UFor unmanned plane U_uThe number of the cells; n is a radical of_AFor farmland A_iThe number of the cells; w is the w-th time window, and is 1 or 2; n th₀,N_A+1The block farmland represents the starting point and the end point of the unmanned aerial vehicle; SQ_iIndicating farmland A_iThe area of (d); p_iwIndicating that field A is completed in the w-th time window_iThe benefit of the task; s_iwuRefers to unmanned plane U_uFor farmland A in the w time window_iThe time to start spraying the pesticide; t is t_ijwuRefers to unmanned plane U_uFor farmland A in the w time window_i,A_jThe time of flying according to a preset flying mode; s_jwuRefers to unmanned plane U_uFor farmland A in the w time window_jThe time to start spraying the pesticide; m is a preset value; o is_iwu,C_iwuAre respectively unmanned aerial vehicle U_uFor farmland A in the w time window_iA start time and an end time for executing the task; t is t_iwuRefers to unmanned plane U_uFor farmland A in the w time window_iThe time for executing the task according to a preset flight mode; e_wuRepresenting the maximum flight time of the unmanned aerial vehicle; if x_iwu1, then indicate unmanned plane U_uTo farmland A in the w time window_iThe medicine spraying task is completed, otherwise the unmanned aerial vehicle does not spray the farmland A_iExecuting the task; if y_ijwu1, then indicate unmanned plane U_uThrough farmland A_i,A_jOtherwise, theUnmanned aerial vehicle does not pass through farmland A_i,A_j。

6. The device according to claim 5, characterized in that the predetermined flight pattern is:

the method comprises the following steps of flying in a farmland according to a parallel scanning flying mode: entering the farmland from a first entry point on a first side in a direction perpendicular to the first side of the farmland, wherein the distance between the first entry point and the nearest vertex of the farmland is the scanning radius of the unmanned aerial vehicle, and the first side is any side of the region to be detected; when the unmanned aerial vehicle needs to turn, the unmanned aerial vehicle turns with the minimum turning radius larger than or equal to that of the unmanned aerial vehicle;

flying between farmlands according to a Dubins path flying mode, wherein the Dubins path flying mode is that the Dubins path flying mode flies according to a path formed by combining an arc path and a straight path, and the path formed by combining the arc path and the straight path comprises { LSL, RSR, RSL, LSR, RLR and LRL }; wherein, L represents the arc section route that unmanned aerial vehicle turned to the left, and R represents the arc section route that unmanned aerial vehicle turned to the right, and S represents that unmanned aerial vehicle flies with the straight line mode.

7. The apparatus of claim 5, wherein the initial solution obtaining unit is further configured to:

encoding the solution of the MUAV-MTW-DTOP model into a chromosome composed according to a preset structure by adopting an MMD encoding method to form an initial solution; the chromosome comprises information of a target area farmland, an unmanned aerial vehicle, a target area entry point and time for the unmanned aerial vehicle to access the first farmland in each route; wherein the target area farmland belongs to a set {1,2_AN, the drone belonging to the set {0,1_UAn entry point of the target region farmland belongs to a set {1, 2.. 8}, and the time for accessing the first farmland in each route belongs to a time window;

the first action of the chromosome is the random full arrangement of the target area farmlands, the second action is the random combination of unmanned aerial vehicles for executing tasks, the third action is the random combination of entry points when the unmanned aerial vehicles execute the tasks on the target area, the fourth action is the random combination of the time for the unmanned aerial vehicles to visit the first farmlands in each route, and the starting time for the same unmanned aerial vehicle to visit the first farmlands in each route is the same.

8. The apparatus of claim 7, wherein the optimal solution computing unit is further configured to perform the following steps:

step 1, generating an initial solution by using the MMD coding method, judging whether the initial solution meets the constraint condition of the MUAV-MTW-DTOP model, if not, continuously generating the initial solution meeting the constraint condition of the MUAV-MTW-DTOP model, generating a parent population with a preset scale and calculating the fitness of the parent population;

step 2, selecting an individual A and an individual B in a parent population to be crossed by using a roulette method, wherein the crossing rule is that the crossing position in the individual A is randomly selected, then the gene in the individual B, which is the same as the first line of the crossing position of the individual A, is searched, the gene at the crossing position in the chromosomes A and B is replaced, all genes, which are newly exchanged in the chromosomes A and B and are the same as the second line of the genes and the fourth line of the genes is in the same time window, are searched, and the fourth line of the searched genes is replaced by the fourth line of the newly exchanged genes in the chromosomes A and B to obtain chromosomes C and D; judging whether chromosomes C and D meet the constraint conditions of the MUAV-MTW-DTOP model, if so, replacing the chromosomes A and B in the population by using the chromosomes C and D, otherwise, ending, and continuously iterating and updating the population in the step 1 to obtain a new offspring population;

and 3, selecting one chromosome in the population in the step 2 for mutation by using a roulette method, wherein the method for mutating the chromosome is at least one of the following mutation methods, and the method comprises the following steps: carrying out random full arrangement on the first row sequence of the chromosomes; and carrying out variation on the unmanned aerial vehicle information of the second line of the chromosome, wherein the variation range is {0, 1.. N_U}; and mutating an entry point of a target region farmland of a third row of chromosomes, wherein the entry point belongs to a set {1, 2.. 8 }; and a fourth temporal variation of the chromosome belonging to two time windows;

the whole chromosome variation steps include: firstly, if the first row sequence of the chromosome is mutated, a farmland sequence full array is randomly generated; selecting whether the second row is mutated or not and the position of the mutation, and if so, randomly generating a mutated unmanned aerial vehicle to replace the original position; randomly selecting whether the third row has variation and a variation position, and if the variation exists, randomly generating a variation entry point to replace the original position; randomly selecting whether the fourth line is mutated or not and a mutation position, randomly generating mutation time if the fourth line is mutated, replacing the original position, finding a gene which is the same as the second line of the mutation position and in the same time window as the fourth line, mutating the fourth line into new mutation time, and finally obtaining a new mutation chromosome;

judging whether the varied chromosomes meet the constraint conditions of the MUAV-MTW-DTOP model or not, if so, replacing the chromosomes in the population, otherwise, finishing the replacement and continuously iterating and updating the population in the step 2 to obtain a new offspring population;

step 4, calculating the population fitness of the offspring in the step 3 and selecting the optimal solution of all solutions in the iteration;

step 5, judging whether the current iteration times reach a preset value or not, if not, combining the child population and the parent population in the step 3 according to a certain proportion to form a new parent population, and returning to the step 2; if so, ending the iteration, and taking the finally obtained optimal solution as a task allocation and track planning result of the unmanned aerial vehicle.

Images