CN106908066B

CN106908066B - Unmanned aerial vehicle monitoring covering single-step optimization flight path planning method based on genetic algorithm

Info

Publication number: CN106908066B
Application number: CN201710278469.5A
Authority: CN
Inventors: 王彤; 刘嘉昕; 马欣
Original assignee: Xian University of Electronic Science and Technology
Current assignee: Xian University of Electronic Science and Technology
Priority date: 2017-04-25
Filing date: 2017-04-25
Publication date: 2019-12-20
Anticipated expiration: 2037-04-25
Also published as: CN106908066A

Abstract

The invention discloses a flight path planning method for unmanned aerial vehicle monitoring coverage single-step optimization based on a genetic algorithm, which has the following thought: setting a flying area of the unmanned aerial vehicle cluster as A, setting a designated task monitoring area in the flying area A as S, defining the independent variables of the yaw angles of the N unmanned aerial vehicles, and respectively setting the yaw angles of the N unmanned aerial vehicles at the initial moment and the position coordinate matrix of the N unmanned aerial vehicles at the initial moment in the A; k belongs to {0,1,2, …, K }, wherein K represents the kth route planning, K is the set total steps of the route planning, and the route planning from the kth route to the (K +1) th route planning is recorded as 1 single step; respectively obtaining flight path positions of N unmanned aerial vehicles at (k +1) t time in A, wherein i is 1,2, … and N; further obtaining the real-time maximum monitoring coverage of the S by the N unmanned aerial vehicles finished by the S single step, and adding 1 to S; the initial value of s is 1; and adding 1 to K until K is larger than K to obtain the real-time maximum monitoring coverage of the S by the N unmanned aerial vehicles finished by the S 'single step, so that the N unmanned aerial vehicles can continuously monitor the S maximum range, and the S' single step is used for planning the K-1-th route to the K-th route.

Description

Unmanned aerial vehicle monitoring covering single-step optimization flight path planning method based on genetic algorithm

Technical Field

The invention belongs to the technical field of unmanned aerial vehicles, and particularly relates to a single-step optimization flight path planning method for monitoring coverage of an unmanned aerial vehicle based on a genetic algorithm, which is suitable for realizing continuous monitoring of the maximum coverage area of an unmanned aerial vehicle group on a designated area.

Background

An Unmanned Aerial Vehicle (UAV) is a short name of an Unmanned Aerial Vehicle (Unmanned Aerial Vehicle), and occupies an important application position in military and civil use by virtue of the characteristics of no personnel risk, low cost and good concealment; in practical application of unmanned aerial vehicle investigation, sometimes, due to certain specific tasks, monitoring of the maximum coverage area of a specified area is required; in order to pursue high-efficiency application, a reference track of the unmanned aerial vehicle needs to be planned in advance by a ground command center, so that the unmanned aerial vehicle can fly according to the reference track according to reconnaissance requirements. Therefore, the unmanned aerial vehicle coverage optimizing track planning technology is an important content of the flight mission of the unmanned aerial vehicle.

At present, the research on the unmanned aerial vehicle area coverage problem is generally less at home and abroad, wherein the research on the unmanned aerial vehicle area coverage problem is more representative; in 2004, Ivan Maza and Anibal Ollero proposed a small team unmanned aerial vehicle area coverage track planning, by dividing the whole area into a plurality of sub-areas, the basic path shape of each sub-area is Z-shaped, when a certain unmanned aerial vehicle fails or crashes, re-planning is carried out immediately; in 2006, the research of Agarwal also adopts the idea of area division, a flight area is divided into a plurality of rectangular sub-areas, the areas are allocated according to the capability of each unmanned aerial vehicle for executing the covering task, the unmanned aerial vehicle is simplified to only allow 90-degree and 180-degree turning, but the turning radius is not considered in the defect of the covering scheme; in 2010, Chenhai et al proposed a track planning algorithm for a convex polygon area, which converts the problem of track planning coverage of the convex polygon area into a problem of solving the width of the convex polygon, and the unmanned aerial vehicle only needs to fly along a Z-shaped route along the direction of a support parallel line when the width appears, but does not consider the influence of the minimum turning radius on the Z-shaped route in the flying process.

Most of the methods for planning the coverage tracks of the areas aim at the condition that the starting point and the end point of the required tracks are fixed, and the basic principle is that the optimal tracks are formed by cutting the areas, avoiding obstacles, restricting oil consumption and turning times, so that the specific unmanned aerial vehicle can cover each area after cutting through a 'cattle-ploughing' flight route; most of the researches plan the flight path based on the deterministic algorithm, the algorithm has certain defects, when the flight path planning is carried out on a large-range complex environment, the path searching has overlarge calculated amount, low efficiency and poor optimizing capability, and the requirements on the calculating efficiency and the reliability of the flight path planning cannot be ensured. In addition, in practical situations, some tasks require the unmanned aerial vehicle to achieve the maximum coverage area for continuous monitoring of a specified area, and the flight path planning required by the flight tasks often has no fixed starting point and fixed ending point.

Disclosure of Invention

The invention aims to provide a single-step flight path planning method for unmanned aerial vehicle monitoring coverage and optimizing based on a genetic algorithm, which is a single-step flight path planning method for unmanned aerial vehicle monitoring coverage and optimizing based on a genetic algorithm, wherein the single-step flight path planning method for unmanned aerial vehicle monitoring coverage and optimizing is based on a genetic algorithm and used for unmanned aerial vehicle cluster monitoring area coverage and optimizing, the problem of area coverage flight path planning is organically combined with the genetic algorithm, the flight task of an unmanned aerial vehicle cluster can be effectively solved, and the flight path planning problem that the monitoring coverage area of a designated area is maximum and the required flight path has no fixed starting point and end point can be realized.

In order to achieve the technical purpose, the invention is realized by adopting the following technical scheme.

An unmanned aerial vehicle monitoring coverage single-step optimization flight path planning method based on a genetic algorithm comprises the following steps:

step 1, setting a flying area of an unmanned aerial vehicle cluster as A, and setting a designated task monitoring area in the flying area A as S, wherein the unmanned aerial vehicle cluster comprises N unmanned aerial vehicles, each unmanned aerial vehicle is provided with an airborne radar, and each unmanned aerial vehicle flies at a constant speed;

step 2, defining independent variables of yaw angles of the N unmanned aerial vehicles, and respectively setting the yaw angles of the N unmanned aerial vehicles at the initial moment and a position coordinate matrix of the N unmanned aerial vehicles at the initial moment in the flyable area A;

initialization: k belongs to {0,1,2, …, K }, wherein K represents the kth route planning, K is the set total steps of the route planning, the initial value of K is 0, and the route planning from the kth route planning to the (K +1) th step is recorded as 1 single step;

and 3, assuming that the flight path position of the ith unmanned aerial vehicle in the flyable area A at kt moment isThe unmanned aerial vehicle can fly in the flyable area A, and further the flyable area of the flight path positions of the N unmanned aerial vehicles in the flyable area A at the (k +1) t moment is obtained respectively, wherein t represents a single-step flightTrace planning time intervals; 1,2, …, N;

step 4, obtaining the real-time maximum monitoring coverage of the N unmanned aerial vehicles which finish the S single step to the designated task monitoring area S according to the flight path positions of the N unmanned aerial vehicles in the flyable area A at the (k +1) t moment, and adding 1 to S; the initial value of S is 1, S belongs to {1,2, …, S '}, and S' is the total number of single steps;

and 5, adding 1 to K, and sequentially and repeatedly executing the steps 3 and 4 until K is larger than K to obtain the real-time maximum monitoring coverage of the S 'single-step completed N unmanned aerial vehicles on the specified task monitoring area S, so that the N unmanned aerial vehicles can continuously monitor and cover the maximum range of the specified task monitoring area S, and the S' single step is from the K-1 flight path planning to the K flight path planning.

The invention has the beneficial effects that: the method of the invention changes the flight yaw angle of the unmanned aerial vehicle groupThe method has the advantages that the sum of the reconnaissance coverage areas of the unmanned aerial vehicle cluster at the appointed time is used as an algorithm fitness function, and a path planning problem and a genetic algorithm are organically combined, so that a brand-new path planning problem different from the traditional area coverage optimizing path planning situation can be solved, namely, the path planning problem that the starting point and the end point of a path are not specified, and the maximum continuous monitoring coverage range of the appointed area is realized when the unmanned aerial vehicle cluster flies along the path is required.

Drawings

The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.

FIG. 1 is a flow chart of a flight path planning method for unmanned aerial vehicle monitoring coverage single step optimization based on genetic algorithm;

fig. 2 is a coverage of 6 unmanned aerial vehicles in the designated area S to be monitored in step 50, where a white point is a position of the unmanned aerial vehicle, and a black area is a sum of coverage areas of the 6 unmanned aerial vehicles;

fig. 3 is a diagram of a flight path planning result obtained by using the method of the present invention, in which a solid line enclosed area is a flyable area a of 6 unmanned aerial vehicles, a dotted line enclosed area is an area S to be monitored, and a dot-dash line is a respective flight path of 6 unmanned aerial vehicles;

fig. 4 is a graph showing the percentage change of the monitoring coverage area of the unmanned aerial vehicle group, wherein the abscissa is the percentage of the monitoring coverage area of the unmanned aerial vehicle group in the designated area S to be monitored, and the ordinate is the number of steps of the flight path planning by using the method, and the unit is the step.

Detailed Description

Referring to fig. 1, it is a flow chart of a flight path planning method for unmanned aerial vehicle monitoring coverage single step optimization based on genetic algorithm of the present invention; the unmanned aerial vehicle monitoring coverage single-step optimization flight path planning method based on the genetic algorithm comprises the following steps:

step 1, setting environmental parameters of a flight path planning problem. The current problem assumes that the drone is flying within a prescribed flyable area and ensures that the maximum coverage area requirement for continuous surveillance is achieved for a designated reconnaissance area within the flyable area. Setting a flyable area A of the unmanned aerial vehicle, and setting a designated task monitoring area in the flyable area A as S; secondly, setting a series of unmanned aerial vehicle motion parameters according to the system maneuvering performance of the unmanned aerial vehicle; finally, the reconnaissance radius R of the unmanned aerial vehicle is set according to the airborne radar range equation_s。

The specific substeps of step 1 are:

1.1 when unmanned aerial vehicle carries out the flight mission, the safe region that allows unmanned aerial vehicle flight is the unmanned aerial vehicle can fly the region, and it can fly the region for A to establish unmanned aerial vehicle, if fly away from this unmanned aerial vehicle can fly region A, then probably hit by threats such as the air gun fire of prevention of hostile force, ground to the guided missile momentum, directional radiation device, lead to the flight mission to fail.

A designated task monitoring area in an unmanned aerial vehicle flyable area A is set as S, and a flight task of flight path planning requires real-time maximum monitoring coverage on the designated task monitoring area S, so that a radar can continuously acquire a ground potential threat target of the designated task monitoring area S.

1.2 the unmanned aerial vehicle movement parameter is a state parameter representing the unmanned aerial vehicle moving on the ground or flying in the air, and the state parameter is used for passingDetermining the motion of the unmanned aerial vehicle, wherein the motion parameters related to the flight path planning problem are as follows: setting a yaw angle of an unmanned aerial vehicleThe device is used for representing an included angle between the flight speed direction of the unmanned aerial vehicle and the positive direction of the x axis of the horizontal coordinate system; setting a roll angle gamma of the unmanned aerial vehicle, wherein the roll angle gamma is used for representing an included angle between a symmetrical plane of the unmanned aerial vehicle and a vertical plane containing an x axis of a horizontal coordinate system; setting a turning angle theta of the unmanned aerial vehicle and a turning radius R of the unmanned aerial vehicle; and an airborne radar is arranged on the unmanned aerial vehicle, and the airborne radar is a transmitter and a receiver.

The connecting line between the turning starting point and the turning ending point of the unmanned aerial vehicle in the turning process and the initial course track of the unmanned aerial vehicle form an included angle which is the turning angle theta of the unmanned aerial vehicle; the arc radius formed by the original course of the unmanned aerial vehicle and the new outward tangent of the unmanned aerial vehicle is the turning radius R of the unmanned aerial vehicle; the unmanned aerial vehicle is controlled by the unmanned aerial vehicle to determine the roll angle gamma of the unmanned aerial vehicle due to the limitation of self mobility performance, and the minimum turning radius R of the unmanned aerial vehicle exists under the limitation of the roll angle gamma of the unmanned aerial vehicle_minAnd with the minimum turning radius R of the drone_minThe corresponding maximum turning angle of the unmanned aerial vehicle is recorded as the maximum turning angle theta of the unmanned aerial vehicle_max(ii) a The turning angle theta of the unmanned aerial vehicle is not larger than the maximum turning angle theta of the unmanned aerial vehicle_maxI.e. theta ≦ theta_max(ii) a The turning radius R of the unmanned aerial vehicle is not less than the minimum turning radius R of the unmanned aerial vehicle_minI.e. R.gtoreq.R_min(ii) a In the embodiment of the invention, the roll angle gamma of the unmanned aerial vehicle is 30 degrees.

Setting the average flying speed of the unmanned aerial vehicle as v_pFor representing the average value of the flight speed of the unmanned aerial vehicle within the single-step flight path planning time interval t; assuming the average value v of the flying speed of the unmanned aerial vehicle in the single-step flight path planning time interval t during the flight process_pAlways kept unchanged.

1.3 flight mission requires unmanned aerial vehicle to realize continuous monitoring of assigned mission monitoring area S in maximum range, the invention simplifies the monitoring range of single unmanned aerial vehicle into the mode of using the unmanned aerial vehicle as circle center and airborne radarMaximum distance of action R_sIs a circle with a radius, and the maximum action distance R of the airborne radar_sAccording to the radar distance equation, the following can be obtained:

wherein, P_tThe peak power of the airborne radar is represented, G represents the antenna gain of the airborne radar, lambda represents the wavelength of electromagnetic waves emitted by the airborne radar, sigma represents the scattering cross section area of a ground potential threat target in the detection range of the airborne radar, k' represents a Boltzmann constant, and T₀Expressing standard room temperature, B expressing the bandwidth of the airborne radar, F expressing the ratio of the signal-to-noise ratio of the input end to the signal-to-noise ratio of the output end of the airborne radar, L expressing the self loss of the airborne radar, S expressing the signal power of the output end of the airborne radar, N expressing the noise power output by the airborne radar, (S/N)_ominThe minimum output signal-to-noise ratio required by the airborne radar is shown, and the subscript omin shows the minimum output operation.

And 2, abstracting a track planning problem for realizing continuous monitoring in the maximum range on the designated area into a mathematical optimization problem. Firstly, using N unmanned aerial vehicles to yawAmount of change ofAs an argument to the flight path planning problem; secondly, setting initial conditions of the flight path planning problem and using vectorsAnd matrix P⁰The coordinates represent the yaw angle of the unmanned aerial vehicle cluster at the zero moment and the position of the unmanned aerial vehicle cluster in the flyable area A; and finally, setting a fitness function and a termination criterion of the algorithm, taking the sum of the reconnaissance coverage areas of the unmanned aerial vehicle cluster at the designated moment as the fitness function, and setting a maximum iteration algebra G to terminate the algorithm.

Initialization: k belongs to {0,1,2, …, K }, wherein K represents the kth route planning, K is the set total steps of the route planning, the initial value of K is 0, and the route planning from the kth route planning to the (K +1) th step is marked as 1 single step.

The specific substeps of step 2 are:

2.1 defining yaw Angle independent variables of N unmanned aerial vehicles Indicating the yaw angle variation of the ith unmanned aerial vehicle in the single-step flight path planning time interval t,θ_maxthe maximum turning angle of the unmanned aerial vehicle is represented, N is the total number of the unmanned aerial vehicles, and N is a positive integer greater than 0.

2.2 setting initial conditions of the flight path planning problem: respectively setting the initial moment yaw angles of the N unmanned aerial vehicles and the position coordinate matrixes of the N unmanned aerial vehicles in the flyable area A at the initial moment, namely respectively using vectorsRepresenting course vectors of N unmanned aerial vehicles at zero time by using matrix P⁰The position coordinate matrix of N unmanned aerial vehicles in the flyable area A at zero time is represented by the following expressions:

wherein the content of the first and second substances,indicating the yaw angle of the ith drone at time zero,represents the flight path position of the ith unmanned plane in the flyable area A at the zero moment, the x-axis coordinate of the track position of the ith unmanned plane in the flyable area A at the zero moment,and the y-axis coordinate of the track position of the ith unmanned aerial vehicle in the flyable area A at zero time is represented, and the superscript T represents the transposition operation.

2.3 setting the fitness function termination criterion of the single-step flight path planning algorithm: the flight mission of the flight path planning requires N unmanned aerial vehicles to realize continuous monitoring in the maximum range in a designated mission monitoring area S, so the total scout coverage area of the N unmanned aerial vehicles at kt time is selected as a fitness function of a single-step flight path planning algorithm, wherein K represents the K-th step of flight path planning, K belongs to {0,1,2, …, K }, K is the set total step number of flight path planning, the initial value of K is 0, and the K to the K +1 steps are marked as 1 single step; recording the time interval for planning the single step track as t, and planning the single step track from the kth step track to the (k +1) th step track; the fitness function termination criteria of the single-step flight path planning algorithm are as follows: and setting the maximum iteration algebra G of the genetic algorithm, and terminating the single-step track planning task when the iteration of the genetic algorithm is performed for G times.

And step 3, determining the feasible position.

3.1 suppose that the flight path position of the ith unmanned aerial vehicle in the flyable area A at kt moment isTurning angle theta of the ith unmanned aerial vehicle in flight_iDoes not exceed the maximum turning angle theta of the unmanned aerial vehicle_maxAnd then the ith unmanned aerial vehicle can fly at the track position in the flyable area A at kt moment. According toYaw angle of ith unmanned aerial vehicle in single-step flight path planning time interval tAmount of change ofDetermines an unsmooth arc line, and each point on the unsmooth arc line can be used as the track position of the ith unmanned aerial vehicle in the flyable area A at the (k +1) t momentBecause this non-smooth arc is a collection of flyable positions for track positions, only one of which is the true track position; therefore, for simplification of processing, the unsmooth arc line is approximated to a smooth arc and is recorded as the arc track of the ith unmanned aerial vehicle, and the approximation is reasonable.

3.2 according to the geometrical relation, the variation of the yaw angle of the ith unmanned aerial vehicle in the single-step track planning time interval tTurning angle theta with ith unmanned aerial vehicle_iThe relationship of (1) is:when the maximum turning angle theta of the unmanned aerial vehicle_maxWhen the angle of the flight path is 22.5 °, the absolute value of the yaw angle change of the ith unmanned aerial vehicle in the single-step flight path planning time interval t is less than or equal to 45 °, that is, the absolute value of the yaw angle change isThe arc track chord length d of the ith unmanned aerial vehicle and the arc length l of the ith unmanned aerial vehicle can be obtained through geometric derivation, d is larger than or equal to l and is smaller than or equal to 1.02d, and the arc track chord length of the ith unmanned aerial vehicle and the arc length of the ith unmanned aerial vehicle are approximately considered to be equal in value; therefore, the flight path position of the ith unmanned aerial vehicle in the flyable area A from the time (k +1) tThe smooth circular arc formed by the unmanned aerial vehicle is the track position of the ith unmanned aerial vehicle in the flyable area A at kt momentAnd taking the arc track chord length d of the ith unmanned aerial vehicle as a smooth arc with the radius as the circle center, and taking the smooth arc as a flight path position flyable region of the ith unmanned aerial vehicle in the flyable region A at the moment (k +1) t.

3.3, i is respectively 1 to N, and the substeps 3.1 and 3.2 are sequentially repeated, so that a flight path position flyable region of the 1 st unmanned aerial vehicle in the flyable region A at the moment (k +1) t to a flight path position flyable region of the Nth unmanned aerial vehicle in the flyable region A at the moment (k +1) t are respectively obtained and recorded as a flight path position flyable region of the N unmanned aerial vehicles in the flyable region A at the moment (k +1) t.

And 4, single-step optimizing flight path planning based on the genetic algorithm. Firstly, taking the yaw angle independent variable x of the N unmanned aerial vehicles as an individual gene of a genetic algorithm, and further determining an individual gene linear coding scheme of the genetic algorithm. Then setting the population Z after the g-th iteration in the genetic algorithm^gTo obtain the population Z after the g-th iteration^gAnd calculating the population Z^gThe fitness function value of each of the L individuals, and the evaluation population Z^gThe fitness of the method is based on the roulette method to simulate natural selection, excellent individuals are selected from the excellent individuals to carry out crossing and mutation operations, the fitness of a new population formed by all the individuals subjected to the crossing and mutation operations is evaluated again, and the excellent individuals are continuously selected to enter the next round of crossing and mutation. The above processes are circulated until the maximum iteration algebra K is met, and the optimal individual in the population at the moment is selected as the solution of the single-step flight path planning; wherein G belongs to {1,2, …, G }, G represents the G-th iteration, and the initial value of G is 1; k belongs to {0,1,2, …, K }, wherein K represents the kth route planning, K sets the total steps of the route planning, the initial value of K is 0, and the route planning from the kth route planning to the (K +1) th route planning is marked as a single step.

The specific substeps of step 4 are:

4.1 according to (k +1) t moment N unmanned aerial vehicles can fly to area AThe flight path position in the unmanned aerial vehicle can be in a flight domain, and the independent yaw angle variables of the N unmanned aerial vehicles are recorded asAnd to the yaw angle independent variable of N unmanned aerial vehiclesCarrying out linear coding to obtain the independent variable x of the yaw angle of the N unmanned aerial vehicles which are subjected to the linear coding,wherein the content of the first and second substances,representing the variation of the yaw angle of the ith unmanned aerial vehicle in the single-step track planning time interval t; due to the change of yaw angle of the ith unmanned aerial vehicle in the single-step flight path planning time interval tTurning angle theta with ith unmanned aerial vehicle_iThe relationship of (1) is:therefore, the variation of yaw angle of the ith unmanned aerial vehicle in the single-step track planning time interval tE represents belonging; selecting a decimal linear interpolation coding method in the interval to carry out linear coding on the independent variable x of the yaw angle of the N unmanned aerial vehicles, namely orderingθ_maxRepresents the maximum turning angle of the drone, rand represents [0,1 ]]A random number within the interval.

4.2 setting population in genetic algorithm: calculating to obtain the group Z of N unmanned aerial vehicles when the number of individuals is L after the g iteration according to a decimal linear interpolation coding method^gThe expression is as follows:

wherein x is_jRepresents the genes carried by the jth individual after the g-th iteration, and the deviation angle variation of the ith unmanned aerial vehicle and the jth individual subjected to linear coding after the g-th iteration in the single-step track planning time interval t is represented, the gene represents the deviation angle variation of the N unmanned aerial vehicles in the single-step track planning time interval t, and L is an even number greater than 0.

4.3 setting a fitness function in the genetic algorithm, and evaluating the fitness of each individual in the population Z of the N unmanned aerial vehicles when the number of the individuals is L: the group Z of N unmanned aerial vehicles when the number of individuals after the g-th iteration is L^gAnd as the variation of the yaw angle of the N unmanned aerial vehicles for single-step track planning, the single step refers to the k-th track planning to the k + 1-th track planning, and the feasible positions of the j-th individual N unmanned aerial vehicles at the (k +1) th time after the g-th iteration are calculated according to the following relational expression Representing the feasible position of the ith unmanned aerial vehicle of the jth individual after the g-th iteration at the (k +1) th time, wherein the expression is as follows:

wherein the content of the first and second substances,x-axis coordinates representing feasible positions of the ith drone at the (k +1) th time instant of the jth individual after the g-th iteration,y-axis coordinates representing feasible positions of the ith unmanned aerial vehicle at the (k +1) th time point of the jth individual after the g-th iteration,x-axis coordinate, v, of track position of ith unmanned aerial vehicle in flyable area A at kt time after g-th iteration_pRepresents the average flying speed of the unmanned aerial vehicle,the yaw angle of the ith unmanned aerial vehicle at kt after the g-th iteration is shown,showing the yaw angle variation of the ith unmanned plane and the jth individual which are subjected to linear coding after the g-th iteration within the single-step track planning time interval t,and (3) representing the y-axis coordinate of the track position of the ith unmanned aerial vehicle in the flyable area A at kt after the g-th iteration, wherein cos represents the cosine operation, sin represents the sine operation, and superscript T represents the transposition operation.

4.4, j is respectively 1 to L, the substep 4.3 is repeated, and then the feasible positions of the 1 st individual N unmanned aerial vehicles at the (k +1) th time t after the g-th iteration are respectively obtainedIteration to gFeasible positions of the later L individual N unmanned aerial vehicles at the (k +1) th time tRecording feasible positions of N unmanned aerial vehicles corresponding to L individuals after the g-th iteration at the (k +1) th time tThe expression is as follows:

4.5, the genetic algorithm fitness function of the feasible positions of the jth individual N drones at the (k +1) th time after the g-th iteration is recorded as Y, which is specifically expressed as follows:

function (·) represents solving the area function of the monitoring area, and the obtained area function value of the monitoring area is a fitness value; recording the total area monitored by the jth individual N unmanned aerial vehicles as S_j，S_j＝S_j1∪…∪S_ji∪…∪S_jNJ ═ 1,2, …, L, i ═ 1,2, …, N, u, denotes a union operation, S_jiThe area S which represents the area monitored by the jth individual and the ith unmanned aerial vehicle and is monitored by the jth individual and the ith unmanned aerial vehicle_jiSatisfies the following conditions:

wherein the content of the first and second substances,coordinates of the j individual and the i unmanned plane monitored by the (k +1) step in the area on the x axis after the g iteration,after the g-th iterationCoordinates of the areas monitored by the j individuals and the ith unmanned aerial vehicle in the (k +1) th step on the y axis, x 'represents the independent variable of the genetic algorithm fitness function of the N unmanned aerial vehicles on the x axis, y' represents the independent variable of the genetic algorithm fitness function of the N unmanned aerial vehicles on the y axis, R_sThe maximum action distance of the airborne radar.

Feasible positions of the jth individual N unmanned aerial vehicles after the g-th iteration at the (k +1) th time tSubstituting the feasible positions of the N unmanned aerial vehicles of the jth individual after the g-th iteration into the genetic algorithm fitness function of the feasible positions of the jth individual at the (k +1) th moment, and calculating to obtain the fitness value f of the jth individual after the g-th iteration_j。

4.6 let j take 1 to L respectively, repeat substep 4.5, and then obtain fitness value of 1 st individual after the g-th iteration to fitness value of L-th individual after the g-th iteration respectively, and mark as Z^gFitness value, Z, corresponding to the middle L individuals^gAnd the G-th iteration is the population of N unmanned aerial vehicles when the number of individuals is L.

4.7 Individual selection operations.

Using roulette in Z^gSelecting excellent individuals from fitness values corresponding to the medium L individuals, and performing L' round selection according to the number of the populations, wherein the specific operations are as follows: each round producing one [0,1 ]]Obtaining L random numbers by random numbers in the interval, respectively using the L random numbers as selection pointers, and correspondingly dividing the wheel disc into L partsBlocks, n-th block corresponding to n-th blockProbability of individual being selected, and size of nth block and nth blockThe probability of an individual being selected is in a 1:1 ratio,q∈{1,2,…,L'}，l' is the set total number of wheels,equal to the value of L, and equal to the value of L'.

4.71 determination of superior individuals Using wheel selection, the second in each roundThe probability that an individual is selected is Is any one of the L individuals and can be selected repeatedly.

4.72 after the q-th roundThe probability of an individual being selected is recorded as Is any one of the L individuals and can be selected repeatedly.

4.73 q is 1 to L' respectively, repeat 4.72, and then the 1 st and the second round are obtained respectivelyProbability of individual being selectedTo the L' th wheel and thenProbability of individual being selectedIs marked as Z^gL' excellent individuals of (1) Is shown asGenes carried by excellent individuals;l' represents Z^gTotal number of excellent individuals contained in (1); l 'is equal to L'.

4.8 crossover and mutation operations of individual genes: to Z^gL' excellent individuals of (1)The crossover and mutation operations of the genes are sequentially carried out, Z^gThe genes carried by each excellent individual in the group of the excellent individuals respectively comprise N 'positions, and the values of N' are equal to those of N and correspond to one another.

4.81 pairs of Z^gL' excellent individuals of (1)To carry outWith alternate operation of the wheels, each cross operation producing a [0,1 ]]Random numbers in the interval, if the random numbers generated by the j' th round of crossing operation are larger than the set genetic algorithm crossing probability P_crossThen is at the firstRandomly selecting a position pos on a gene carried by a good individual_cross，And will be firstGenes carried by excellent individualsAnd a firstGenes carried by excellent individualsAre divided into two parts, respectivelyGenes carried by excellent individualsThe first partAnd a firstGenes carried by excellent individualsSecond part of (2)In combination with the firstGenes carried by excellent individualsSecond part of (2)And a firstGenes carried by excellent individualsFirst part ofCombined, thereby forming a crossGenes carried by a new individualAnd after crossingGenes carried by a new individualThe specific intersection process is shown as follows:

4.82 let j' take 1 toRepeat 4.81 while traversing Z^gThe L' excellent individuals in the group are further respectively obtained, and the gene x carried by the 1 st new individual after the cross_{c_1}Gene x carried by the new individual up to the L' after crossover_{c_L}", L" new individuals after the crossover,the gene carried by each new individual after the crossing comprises N' positions,l' is an even number greater than 0.

4.83 on the crossed L' new individualsThe wheel mutation operation, L' andequal in value, producing one [0,1 ] per round]Random number in the interval, ifThe random number generated by the round is not more than the set genetic algorithm variation probability P_mutationThen the gene x carried by the j' th new individual after the crossover_{c_j}"Gene x carried as m-th individual after crossover variation_mThe initial value of m is 1, and m is added with 1; if it is firstThe random number generated by the round is larger than the set genetic algorithm variation probability P_mutationThen, after the crossover, the gene x carried by the j' th new individual_{c_j}"randomly choose a position pos to perform gene variation, i.e., the gene x carried by the j' th new individual after crossing_{c_j}"replacement of an element at position pos toIs [ -2 θ ]_max,2θ_max]A random value in the range is obtained, and then the gene x carried by the m' th individual after cross variation is obtained_m'The initial value of m 'is 1, and m' is added with 1; the specific operation is as follows:

gene x carried by j' new individual after crossing_{c_j}"Inclusion of N' positions is noted And then the gene x carried by the m' th individual after the cross mutation_m'Comprises the following steps:

where pos ∈ {1,2, …, N' },denotes the gene x carried by the j' th new individual after crossover_{c_j}"the element at the position of the second pos,represents the gene x carried by the m' th individual after the cross mutation_m'The element at the position of the second pos.

4.84 orderRespectively take 1 toRepeat 4.83 to obtain the genes carried by the 1 st individual after cross mutationGenes carried by the L' th individual after cross mutationL' individuals Z marked as after cross variation_mutation(ii) a Wherein L' is substituted withThe values are equal and correspond to each other one by one, and the gene carried by each individual after cross variation comprises N' positions.

Wherein x is_m"means a gene carried by the m-th individual after the cross mutation, shows the yaw angle variation of the ith unmanned plane after linear coding and the mth individual after cross mutation within the single-step track planning time interval t, and m 'belongs to {1,2, …, L' }.

4.85 at the same time, in order to prevent the non-convergence of the solution caused by the roulette selection method, the Elite retention strategy is adopted for processing, and the L' individual Z after the cross variation is calculated_mutationFeasible positions of the m' th individual N unmanned aerial vehicles at the (k +1) th t moment Representing the feasible position of the ith unmanned aerial vehicle of the m' th individual at the (k +1) th time t, the expression is as follows:

wherein the content of the first and second substances,x-axis coordinates representing the feasible location of the ith drone at time (k +1) t for the m "th individual,y-axis coordinates representing the feasible location of the ith drone at time (k +1) t for the m "th individual,the x-axis coordinate, v, of the track position of the ith unmanned aerial vehicle in the flyable area A at the kt moment_pRepresents the average flying speed of the unmanned aerial vehicle,the yaw angle of the ith unmanned aerial vehicle at kt moment is shown,shows the yaw angle variation of the ith unmanned plane and the mth individual after linear coding in the k +1 step single-step flight path planning time interval t,and (3) representing the y-axis coordinate of the track position of the ith unmanned aerial vehicle in the flyable area A at the kt moment, wherein cos represents the cosine operation, sin represents the sine operation, and superscript T represents the transposition operation.

4.86 taking 1 to L ' of m ' respectively, repeating 4.85 times to obtain L ' individual Z after cross mutation_mutationFeasible positions of the 1 st individual N unmanned aerial vehicles at the (k +1) th timeTo the cross-mutated L' individuals Z_mutationMiddle and L' individual N unmanned aerial vehicleFeasible position at the (k +1) th time tMarking as feasible positions of N unmanned aerial vehicles corresponding to the L' individuals after cross mutation at the (k +1) th time pointThe expression is as follows:

l' individuals Z after cross mutation_mutationFeasible positions of the m' th individual N unmanned aerial vehicles at the (k +1) th t momentFitness function of genetic algorithmIs specifically represented as follows:

function (·) represents solving the area function of the monitoring area, and the obtained area function value of the monitoring area is a fitness value; u denotes a union operation, S_jRepresents the total area, S, monitored by the jth individual N drones_iIndicates the area monitored by the ith unmanned aerial vehicle of the jth individual and the area S monitored by the ith unmanned aerial vehicle_iSatisfies the following conditions:

coordinates in the x-axis of the area monitored by the jth individual ith drone,the coordinates of the monitored area of the ith unmanned aerial vehicle of the jth individual on the y axis are represented by x 'which represents the independent variable of the genetic algorithm fitness function of the N unmanned aerial vehicles on the x axis, y' which represents the independent variable of the genetic algorithm fitness function of the N unmanned aerial vehicles on the y axis, and R_sThe maximum action distance of the airborne radar.

4.87 Cross-mutated L "individuals Z_mutationFeasible positions of the m' th individual N unmanned aerial vehicles at the (k +1) th t momentSubstituting into the L' individuals Z after cross mutation_mutationFeasible positions of the m' th individual N unmanned aerial vehicles at the (k +1) th t momentFitness function of genetic algorithmIn the method, L' individual Z after cross mutation is calculated_mutationFitness value f 'of m-th individual'_m″。

4.88 taking 1 to L ' of m ' respectively, repeating 4.87 to obtain L ' individual Z after cross mutation_mutationFitness value f 'of the 1 st individual'₁To the cross-mutated L' individuals Z_mutationL 'individual fitness values f'_L", denotes a fitness value f, f ═ f 'for L" individuals after cross mutation'₁,…,f'_m″,…,f'_L″]。

Will Z^gAnd performing descending order arrangement on the fitness values corresponding to the middle L individuals and the fitness values corresponding to the L ' individuals after cross variation according to the fitness values to obtain L + L ' individuals with the fitness values in descending order arrangement, then selecting the front L individuals from the L + L ' individuals with the fitness values in descending order arrangement, adding 1 to g, taking the selected front L individuals as the population of the N unmanned aerial vehicles when the number of the individuals after the g iteration is L, and returning to the substep 4.3.

4.9 repeating substeps 4.3 to 4.8 until a population Z of N unmanned aerial vehicles is obtained when the number of individuals after G iteration is L^GTaking the group Z of N unmanned aerial vehicles when the number of individuals after the G-th iteration is L^GThe gene carried by the individual with the maximum medium fitness value serves as the optimal value x of the yaw angle variation of the N unmanned aerial vehicles at the (k +1) th time t within the single-step flight path planning time interval t_opt， The optimal yaw angle variation of the ith unmanned aerial vehicle at the (k +1) th moment in the single-step track planning time interval t is shown, and the optimal course vector of the N unmanned aerial vehicles at the (k +1) th moment is obtained through calculation of the following course updating relational expression The yaw angle of the ith unmanned aerial vehicle at kt moment is shown,and (3) representing the yaw angle of the ith unmanned aerial vehicle at the (k +1) t moment.

Calculating the position coordinate matrix P of the N unmanned aerial vehicles in the flyable area A at the (k +1) th time t through the following position updating relational expression^k+1And the position coordinate matrix P of the N unmanned aerial vehicles in the flyable area A at the (k +1) th time point^k+1And (3) obtaining the optimal track position of the N unmanned aerial vehicles in the flyable area A at the (k +1) th time.

The position coordinate matrix P of the N unmanned aerial vehicles in the flyable area A at the (k +1) th time point^k+1The expression of (a) is:

wherein the content of the first and second substances,represents the flight path position of the ith unmanned plane in the flyable area A at the (k +1) th time point t,x-axis coordinates representing the track position of the ith drone within the flyable area a at time (k +1) t,y-axis coordinates representing the track position of the ith drone within the flyable area a at time (k +1) t,an x-axis coordinate representing a track position of the ith drone within the flyable area a at a time kt,y-axis coordinate, v, representing track position of ith unmanned aerial vehicle in flyable area A at kt moment_pRepresenting the average flying speed of the unmanned aerial vehicle, cos represents the operation of finding the cosine, and superscript T represents the transposition operation.

The optimal course vector of N unmanned aerial vehicles at the (k +1) th time tAnd the position of N unmanned planes in the flyable area A at the (k +1) th time pointCoordinate matrix P^k+1The N unmanned aerial vehicles which are taken as the S-th single step completion cover the designated task monitoring area S in real time in the largest mode, and the sum of S and 1 is added; s has an initial value of 1, S ∈ {1,2, …, S '}, S' being the total number of single steps.

Specifically, the optimal course vector of N unmanned aerial vehicles at the (k +1) th t moment is usedAnd a position coordinate matrix P of N unmanned aerial vehicles in the flyable area A at the (k +1) th time point^k+1And (3) as an initial condition of the next step of single-step optimization flight path planning based on the genetic algorithm, using time serial processing, planning the next step of flight path position in the current step of flight time by using the method in the step 4, and continuously obtaining a plurality of optimal flight path positions after single-step planning, thereby realizing the continuous monitoring of the designated task monitoring area S in the maximum range by the N unmanned aerial vehicles.

The effect of the present invention is further verified and explained by the following simulation experiment.

Simulation conditions:

the simulation assumes that 6 unmanned aerial vehicles with the reconnaissance radius of 70km are used for monitoring a designated task monitoring area S of 200km multiplied by 200km, a flyable area A where the unmanned aerial vehicle cluster is located is a rectangular area of 250km multiplied by 250km, the designated task monitoring area is located at the center of the flyable area, and the course vectors of the 6 unmanned aerial vehicles at zero timeAnd a position coordinate matrix P of the unmanned aerial vehicle in the flyable area A at the zero moment⁰Respectively as follows:

the flight path of each step of the unmanned aerial vehicle group is a single-step flight path planning method for monitoring, covering and optimizing the unmanned aerial vehicle group based on the genetic algorithm, the flight path obtained by the experiment is the result of performing 100-step single-step planning, and the detailed simulation parameters are shown in the following table 1:

TABLE 1

(II) simulation content and result analysis

The results of performing 100-step route planning by using the single-step route planning method for unmanned aerial vehicle fleet monitoring coverage optimization based on the genetic algorithm are shown in fig. 2 to 4.

Fig. 2 is a coverage of 6 unmanned aerial vehicles in the designated area S to be monitored in step 50, where a white point is a position of the unmanned aerial vehicle, and a black area is a sum of coverage areas of the 6 unmanned aerial vehicles; as can be seen from fig. 2, the monitoring coverage of 6 drones can approximately completely cover the designated area S to be monitored.

Fig. 3 is a diagram of a flight path planning result obtained by using the method of the present invention, in which a solid line enclosed area is a flyable area a of 6 unmanned aerial vehicles, a dotted line enclosed area is an area S to be monitored, and a dot-dash line is a respective flight path of 6 unmanned aerial vehicles; as can be seen from fig. 3, the route points obtained by planning are all distributed in the flyable area a, which indicates that the route points obtained by the method are all effective and feasible.

Fig. 4 is a graph showing the percentage change of the monitoring coverage area of the unmanned aerial vehicle group, wherein the abscissa is the percentage of the monitoring coverage area of the unmanned aerial vehicle group in the designated area S to be monitored, and the ordinate is the number of steps for planning the flight path by using the method, and the unit is the step; as can be seen from the unmanned aerial vehicle fleet monitoring coverage area curve in fig. 4, the flight path obtained based on the method of the present invention enables the coverage area percentage of the unmanned aerial vehicle fleet to be continuously maintained above 99.5% after convergence, which proves that the single-step flight path planning method for unmanned aerial vehicle fleet monitoring coverage optimization based on the genetic algorithm can realize the continuous monitoring of the unmanned aerial vehicle fleet to the maximum coverage area of the designated area.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention; thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims

1. An unmanned aerial vehicle monitoring coverage single-step optimization flight path planning method based on a genetic algorithm is characterized by comprising the following steps:

and 3, assuming that the flight path position of the ith unmanned aerial vehicle in the flyable area A at kt moment isThe unmanned aerial vehicle can fly in the flyable area A, and further the flyable area of the flight path positions of the N unmanned aerial vehicles in the flyable area A at the (k +1) t moment is obtained respectively, wherein t represents the single-step flight path planning timeSpacing; 1,2, …, N;

2. The method for planning a flight path by monitoring unmanned aerial vehicles through a genetic algorithm to cover single-step optimization according to claim 1, wherein in step 1, each unmanned aerial vehicle is provided with an airborne radar, and further comprising:

the maximum action distance of the airborne radar is R_sThe expression is as follows:

wherein, P_tThe peak power of the airborne radar is represented, G represents the antenna gain of the airborne radar, lambda represents the wavelength of electromagnetic waves emitted by the airborne radar, sigma represents the scattering cross section area of a ground potential threat target in the detection range of the airborne radar, k' represents a Boltzmann constant, and T₀Expressing standard room temperature, B expressing the bandwidth of the airborne radar, F expressing the ratio of the signal-to-noise ratio of the input end to the signal-to-noise ratio of the output end of the airborne radar, L expressing the self loss of the airborne radar, S expressing the signal power of the output end of the airborne radar, D expressing the noise power output by the airborne radar, (S/N)_ominThe minimum output signal-to-noise ratio required by the airborne radar is shown, and the subscript omin shows the minimum output operation.

3. The genetic algorithm-based unmanned aerial vehicle monitoring coverage single-step optimization flight path planning method according to claim 1, wherein the substep of the step 2 is:

2.1 defining yaw Angle independent variables of N unmanned aerial vehicles Indicating the yaw angle variation of the ith unmanned aerial vehicle in the single-step flight path planning time interval t,θ_maxrepresenting the maximum turning angle of the unmanned aerial vehicle, wherein N is the total number of the unmanned aerial vehicles, and is a positive integer greater than 0;

2.2 setting the initial time yaw angle of N unmanned aerial vehicles and the position coordinate matrix of the N unmanned aerial vehicles in the flyable area A at the initial time, namely respectively using vectorsRepresenting course vectors of N unmanned aerial vehicles at zero time by using matrix P⁰The position coordinate matrix of N unmanned aerial vehicles in the flyable area A at zero time is represented by the following expressions:

wherein the content of the first and second substances,indicating the yaw angle of the ith drone at time zero,represents the flight path position of the ith unmanned plane in the flyable area A at the zero moment, the x-axis coordinate of the track position of the ith unmanned plane in the flyable area A at the zero moment,the y-axis coordinate of the track position of the ith unmanned aerial vehicle in the flyable area A at zero moment is represented, and the superscript T represents transposition operation;

2.3 setting the fitness function termination criterion of the single-step flight path planning algorithm: the flight mission of the flight path planning requires N unmanned aerial vehicles to realize continuous monitoring in the maximum range in a designated mission monitoring area S, so the total scout coverage area of the N unmanned aerial vehicles at kt time is selected as a fitness function of a single-step flight path planning algorithm, wherein K represents the K-th step of flight path planning, K belongs to {0,1,2, …, K }, K is the set total step number of flight path planning, the initial value of K is 0, and the K to the K +1 steps are marked as 1 single step; the fitness function termination criteria of the single-step flight path planning algorithm are as follows: and setting the maximum iteration algebra G of the genetic algorithm, and terminating the single-step track planning task when the iteration of the genetic algorithm is performed for G times.

4. The method for planning the flight path of unmanned aerial vehicle monitoring coverage single-step optimization based on the genetic algorithm as claimed in claim 1, wherein in step 3, the flight path position flyable domain of the N unmanned aerial vehicles at the (k +1) t moment in the flyable region a is obtained by the following process:

3.1 suppose that the flight path position of the ith unmanned aerial vehicle in the flyable area A at kt moment isTurning angle theta of the ith unmanned aerial vehicle in flight_iDoes not exceed the maximum turning angle theta of the unmanned aerial vehicle_maxThen, the ith unmanned aerial vehicle can fly at the track position in the flyable area A at kt moment;

according to the yaw angle of the ith unmanned aerial vehicle in the single-step track planning time interval tAmount of change ofDetermining a smooth circular arc, wherein each point on the smooth circular arc can be used as the track position of the ith unmanned aerial vehicle in the flyable area A at the (k +1) t momentRecording the smooth arc line as the arc track of the ith unmanned aerial vehicle;

3.2 respectively calculating the arc track chord length d of the ith unmanned aerial vehicle and the arc track arc length l of the ith unmanned aerial vehicle, satisfying that d is not less than l and not more than 1.02d, and determining the track position of the ith unmanned aerial vehicle in the flyable area A at the moment of (k +1) tThe smooth circular arc formed by the unmanned aerial vehicle is the track position of the ith unmanned aerial vehicle in the flyable area A at kt momentTaking the arc track chord length d of the ith unmanned aerial vehicle as a smooth arc with the radius as the circle center, and further taking the smooth arc as a flight path position flyable region of the ith unmanned aerial vehicle in the flyable region A at the moment (k +1) t;

5. The genetic algorithm-based unmanned aerial vehicle monitoring coverage single-step optimization flight path planning method according to claim 1, wherein in step 4, the S-th single-step completed N unmanned aerial vehicles achieve real-time maximum monitoring coverage of a designated mission monitoring area S by the following steps:

4.1 recording the yaw angle independent variable of the N unmanned aerial vehicles as a flyable region according to the flight path position of the N unmanned aerial vehicles in the flyable region A at the (k +1) t momentAnd to the yaw angle independent variable of N unmanned aerial vehiclesCarrying out linear coding to obtain the independent variable x of the yaw angle of the N unmanned aerial vehicles which are subjected to the linear coding, θ_maxrepresents the maximum turning angle of the drone, rand represents [0,1 ]]A random number in the interval, superscript T represents transposition operation;

wherein x is_jRepresents the genes carried by the jth individual after the g-th iteration, representing the yaw angle variation of the ith unmanned aerial vehicle and the jth individual which are subjected to linear coding after the g-th iteration within a single-step track planning time interval t, wherein the gene represents the yaw angle variation of the N unmanned aerial vehicles within the single-step track planning time interval t, and L is an even number greater than 0;

4.3 group Z of N unmanned aerial vehicles when the number of individuals after the g-th iteration is L^gAs the variation of the yaw angle of the N unmanned aerial vehicles for single-step track planning, a single step refers to the k-th step to the k + 1-th step, and the feasible positions of the j-th individual N unmanned aerial vehicles at the (k +1) th time after the g-th iteration are calculated according to the following relational expression Representing the feasible position of the ith unmanned aerial vehicle of the jth individual after the g-th iteration at the (k +1) th time, wherein the expression is as follows:

wherein the content of the first and second substances,x-axis coordinates representing feasible positions of the ith unmanned aerial vehicle of the jth individual after the g-th iteration at the (k +1) th time point,y-axis coordinates representing feasible positions of the ith unmanned aerial vehicle of the jth individual after the g-th iteration at the (k +1) th time point,x-axis coordinate, v, of track position of ith unmanned aerial vehicle in flyable area A at kt time after g-th iteration_pRepresents the average flying speed of the unmanned aerial vehicle,the yaw angle of the ith unmanned aerial vehicle at kt after the g-th iteration is shown,showing the yaw angle variation of the ith unmanned plane and the jth individual which are subjected to linear coding after the g-th iteration within the single-step track planning time interval t,representing the y-axis coordinate of the track position of the ith unmanned aerial vehicle in the flyable area A at kt moment after the g-th iteration, wherein cos represents the cosine operation, sin represents the sine operation, and superscript T represents the transposition operation;

4.4, j is respectively 1 to L, the substep 4.3 is repeated, and then the feasible positions of the 1 st individual N unmanned aerial vehicles at the (k +1) th time t after the g-th iteration are respectively obtainedFeasible positions of the L-th individual N unmanned aerial vehicles at the (k +1) th time after the g-th iterationRecording feasible positions of N unmanned aerial vehicles corresponding to L individuals after the g-th iteration at the (k +1) th time tThe expression is as follows:

function (·) represents solving the area function of the monitoring area, and the obtained area function value of the monitoring area is a fitness value;

feasible positions of the jth individual N unmanned aerial vehicles after the g-th iteration at the (k +1) th time tSubstituting the feasible positions of the N unmanned aerial vehicles of the jth individual after the g-th iteration into the genetic algorithm fitness function of the feasible positions of the jth individual at the (k +1) th moment, and calculating to obtain the fitness value f of the jth individual after the g-th iteration_j；

4.6 let j take 1 to L respectively, repeat substep 4.5, and then obtain fitness value of 1 st individual after the g-th iteration to fitness value of L-th individual after the g-th iteration respectively, and mark as Z^gFitness value, Z, corresponding to the middle L individuals^gThe number of individuals after the g iteration is L, and the number of the unmanned aerial vehicles is N;

4.7 roulette method in Z^gSelecting excellent individuals from fitness values corresponding to the medium L individuals to obtain Z^gL' excellent individuals of (1)

4.8 pairs of Z^gL' excellent individuals of (1)Sequentially carrying out gene crossing and mutation operations to obtain the fitness value f corresponding to the L' individuals after crossing and mutation;

will Z^gThe fitness values corresponding to the middle L individuals and the fitness values corresponding to the L ' individuals after cross variation are subjected to descending order arrangement according to the fitness values to obtain L + L ' individuals with the fitness values in descending order arrangement, then the front L individuals are selected from the L + L ' individuals with the fitness values in descending order arrangement, g is added by 1, the selected front L individuals are used as the population of N unmanned aerial vehicles when the number of individuals after the g iteration is L, and the substep is returned to 4.3;

4.9 repeating substeps 4.3 to 4.8 until a population Z of N unmanned aerial vehicles is obtained when the number of individuals after G iteration is L^GTaking the group Z of N unmanned aerial vehicles when the number of individuals after the G-th iteration is L^GThe gene carried by the individual with the maximum medium fitness value serves as the optimal value x of the yaw angle variation of the N unmanned aerial vehicles at the (k +1) th time t within the single-step flight path planning time interval t_opt， Representing the optimal yaw angle variation of the ith unmanned aerial vehicle within the single-step track planning time interval t at the (k +1) th time t;

obtaining the optimal course vector of the N unmanned aerial vehicles at the (k +1) th t moment by calculating the following course updating relational expression The yaw angle of the ith unmanned aerial vehicle at kt moment is shown,representing the yaw angle of the ith unmanned aerial vehicle at the moment (k +1) t;

calculating the position coordinate matrix P of the N unmanned aerial vehicles in the flyable area A at the (k +1) th time t through the following position updating relational expression^k+1And the position coordinate matrix P of the N unmanned aerial vehicles in the flyable area A at the (k +1) th time point^k+1The optimal track position of N unmanned aerial vehicles in the flyable area A at the (k +1) th time t;

wherein the content of the first and second substances,represents the flight path position of the ith unmanned plane in the flyable area A at the (k +1) th time point t,x-axis coordinates representing the track position of the ith drone within the flyable area a at time (k +1) t,y-axis coordinates representing the track position of the ith drone within the flyable area a at time (k +1) t,an x-axis coordinate representing a track position of the ith drone within the flyable area a at a time kt,y-axis coordinate, v, representing track position of ith unmanned aerial vehicle in flyable area A at kt moment_pRepresenting the average flying speed of the unmanned aerial vehicle, cos representing the operation of finding a cosine, and superscript T representing the transposition operation;

the optimal course vector of N unmanned aerial vehicles at the (k +1) th time tAnd a position coordinate matrix P of N unmanned aerial vehicles in the flyable area A at the (k +1) th time point^k+1And the N unmanned aerial vehicles which are the S-th single step complete can be used for realizing the real-time maximum monitoring coverage of the designated task monitoring area S.

6. The method for planning a flight path by unmanned aerial vehicle monitoring coverage single step optimization based on genetic algorithm as claimed in claim 5, wherein the process of 4.7 is as follows:

4.71 determination of superior individuals Using wheel selection, the second in each roundThe probability that an individual is selected is

4.72 after the q-th roundThe probability of an individual being selected is recorded as Is any one of L individuals and can be selected repeatedly, q belongs to {1,2, …, L ' }, L ' is a set total number of rounds, and L ' and L have equal values; and selecting the q-th roundThe individual is marked asThe number of the excellent individuals is small,is 1, andand the q th round selectedIndividual correspondence is then orderedAdding 1;

4.73 repeating the sequence q from 1 to L 'for 4.72 times to obtain 1 st to L' excellent individuals, respectively, as Z^gL' excellent individuals of (1) Is shown asGenes carried by excellent individuals;l' represents Z^gTotal number of excellent individuals contained in (1); l 'is equal to L'.

7. A genetic algorithm based unmanned aerial vehicle surveillance coverage single step optimization flight path planning method according to claim 5 or 6, characterized in that the process of 4.8 is:

4.82 let j' take 1 toRepeat 4.81 while traversing Z^gThe L' excellent individuals in the series are respectively obtained, and then the genes x carried by the 1 st new individual after the cross_{c_1}Gene x carried by the L' th new individual after crossover_{c_L”}Marked as L' new individuals after the intersection,the gene carried by each new individual after the crossing comprises N' positions,l' is an even number greater than 0;

4.83 the crossover L' new individuals were processedThe wheel mutation operation, L' andequal in value, producing one [0,1 ] per round]Random number in the interval, ifThe random number generated by the round is not more than the set genetic algorithm variation probability P_mutationThen, the gene x carried by the j' th new individual after the crossover_{c_j”}As gene x carried by the m-th individual after cross mutation_mThe initial value of m is 1, and m is added with 1; if it is firstThe random number generated by the round is larger than the set genetic algorithm variation probability P_mutationThen, after the crossing, the j "th new individual carries gene x_{c_j”}Randomly selecting a position pos for gene variation, namely selecting the gene x carried by the jth new individual after crossing_{c_j”}Replacement of an element at mid-position posIs [ -2 θ ]_max,2θ_max]A random value in the range is obtained, and then the gene x carried by the m' th individual after cross variation is obtained_m'The initial value of m 'is 1, and m' is added with 1; the specific operation is as follows:

gene x carried by jth new individual after crossing_{c_j”}The N' positions are marked as And then the gene x carried by the m' th individual after the cross mutation_m'Comprises the following steps:

where pos ∈ {1,2, …, N' },denotes the gene x carried by the j' th new individual after crossover_{c_j”}The element at the position of the second pos,represents the gene x carried by the m' th individual after the cross mutation_m'An element at the position of the second pos;

4.84 orderRespectively take 1 toRepeat 4.83 to obtain the genes carried by the 1 st individual after cross mutationGenes carried by the L-th individual after cross mutationL' individuals Z marked as after cross variation_mutation(ii) a Wherein, the gene carried by each individual after the cross mutation comprises N 'positions, L' andthe values are equal and correspond to each other one by one;

wherein x is_m”Represents the gene carried by the m' th individual after the cross mutation, representing the yaw angle variation of the ith unmanned aerial vehicle subjected to linear coding and the mth individual after cross mutation in the single-step track planning time interval t, wherein m is belonged to {1,2, …, L' };

4.85 calculation of L "individuals Z after Cross mutation_mutationFeasible positions of the m' th individual N unmanned aerial vehicles at the (k +1) th time t Representing the feasible position of the ith unmanned aerial vehicle of the m' individual at the (k +1) th time t, the expression is as follows:

wherein the content of the first and second substances,x-axis coordinates representing the feasible location of the ith drone for the m "th individual at time (k +1) t,the ith unmanned plane representing the m' individual is at the (k +1) th) the y-axis coordinate of the feasible locations at time t,the x-axis coordinate, v, of the track position of the ith unmanned aerial vehicle in the flyable area A at the kt moment_pRepresents the average flying speed of the unmanned aerial vehicle,the yaw angle of the ith unmanned aerial vehicle at kt moment is shown,shows the yaw angle variation of the ith unmanned plane and the mth individual after linear coding in the k +1 step single-step flight path planning time interval t,representing the y-axis coordinate of the track position of the ith unmanned aerial vehicle in the flyable area A at the kt moment, wherein cos represents cosine solving operation, sin represents sine solving operation, and superscript T represents transposition operation;

4.86 get m 'from 1 to L respectively, repeat 4.85, and then get L' individual Z after cross variation_mutationFeasible positions of the 1 st individual N unmanned aerial vehicles at the (k +1) th timeTo the cross-mutated L' individuals Z_mutationFeasible positions of the middle L' individual N unmanned aerial vehicles at the (k +1) th t momentMarking as feasible positions of N unmanned aerial vehicles corresponding to L' individuals after cross mutation at (k +1) th timeThe expression is as follows:

l' individuals Z after cross mutation_mutationFeasible positions of the m' th individual N unmanned aerial vehicles at the (k +1) th time tFitness function of genetic algorithmIs specifically represented as follows:

4.87 Cross-mutated L' individuals Z_mutationFeasible positions of the m' th individual N unmanned aerial vehicles at the (k +1) th time tSubstituting into the L' individuals Z after cross mutation_mutationFeasible positions of the m' th individual N unmanned aerial vehicles at the (k +1) th time tFitness function of genetic algorithmIn the method, L' individual Z after cross mutation is calculated_mutationM 'th individual fitness value f'_m”；

4.88 taking 1 to L ' from m ', repeating for 4.87 to obtain L ' individual Z after cross mutation_mutationFitness value f 'of the 1 st individual'₁To the cross-mutated L' individuals Z_mutationFitness of the L-th individualValue f'_L”The fitness value f, f ═ f ' corresponding to L ' individuals after cross mutation '₁,…,f'_m”,…,f'_L”]；

4.88 mixing of Z^gAnd performing descending order arrangement on the fitness values corresponding to the middle L individuals and the fitness values corresponding to the L ' individuals after cross variation according to the fitness values to obtain L + L ' individuals with the fitness values in descending order arrangement, then selecting the front L individuals from the L + L ' individuals with the fitness values in descending order arrangement, adding 1 to g, taking the selected front L individuals as the population of the N unmanned aerial vehicles when the number of the individuals after the g iteration is L, and returning to the substep 4.3.