CN112733421B - Task planning method for cooperation of unmanned aerial vehicle with ground fight - Google Patents

Abstract

The invention discloses a task planning method for a cooperation of unmanned aerial vehicles to combat the ground, which relates to the field of unmanned aerial vehicle task planning and comprises the following steps: s1, modeling a battlefield environment: the method comprises the steps of adopting a digital map information fusion principle to fuse original topography and threat information of a combat space into comprehensive topography information; s2, acquiring the number of man machines, the number of unmanned aerial vehicles and the target number: the set ground striking formation comprises 1 man-machine and m unmanned aerial vehicles (v ₁,v₂,…,v_m), and n targets (t ₁,t₂,…,t_n) needing striking are found; establishing a task allocation objective function: the establishment of the objective function comprehensively considers three factors including the hit value of the target, the track cost for the target and the threat cost in the cruising process. And the planar planning and the altitude planning are solved by adopting an A-scale algorithm, so that a task planning scheme meeting requirements can be provided for the cooperation of the unmanned aerial vehicle and the unmanned aerial vehicle.

Description

Task planning method for cooperation of unmanned aerial vehicle with ground fight

Technical Field

The invention relates to the technical field of unmanned aerial vehicle task planning, in particular to a task planning method for a manned unmanned aerial vehicle to perform collaborative ground combat.

Background

Compared with a man-machine, the unmanned aerial vehicle has obvious advantages in the aspects of cost, convenience, maneuverability, adaptability and the like, and is favored by multi-country military. The advantages of each fighter plane in the multi-plane collaborative combat are complementary in the aspects of airborne weapon equipment and tactics, and the fighter plane and the airborne weapon equipment are matched with each other, so that the efficiency of 1+1>2 can be achieved. In recent years, more theoretical researches are carried out on the cooperative formation of multiple unmanned aerial vehicles in China, but the intelligent development degree of the current unmanned aerial vehicles and combat systems is limited, the hardware capacity is not kept up with the theoretical height, the realization of the mutual cooperation between the pure unmanned aerial vehicles in a short time is not practical, and the combat process is controlled and supervised by human operation, so that the task completion and the use safety are ensured. Thus, there is still a need for preferential development of man/unmanned collaboration.

The unmanned/unmanned plane cooperative combat includes a plurality of key technologies including unmanned/unmanned plane cooperative control technology, situation awareness and assessment technology, cooperative task planning technology, formation flight and tracking control technology, battlefield intelligent decision technology, target hit efficiency assessment technology and the like. Many links are closely matched to obtain ideal combat effect. The mission planning technology serves as a center and an important link of the man/unmanned aerial vehicle collaborative combat technology, and the winning probability of combat is determined to a great extent.

There are few studies on mission planning of human/unmanned aerial vehicles at present, and many of the studies disclosed at present are on a two-dimensional plane. The invention researches the cooperative striking task planning of the unmanned/unmanned aerial vehicle in the three-dimensional scene, and aims at a limited centralized and distributed control mode of the unmanned/unmanned aerial vehicle cooperation. The specific scenario is described as follows: before the ground striking task is set, a man-machine acquires necessary information which is transmitted back by the reconnaissance and detection unmanned aerial vehicle and comprises information such as battlefield threat, environment, airspace, the number of targets of enemies and the like, the man-machine distributes targets which need to strike for each unmanned aerial vehicle, the process of the unmanned aerial vehicle for carrying out the task is monitored during the period, each unmanned aerial vehicle has inherent characteristics (such as flying performance, maneuvering performance and the like) and is provided with equipment such as an onboard computer, a payload, a sensor, a data chain and the like, and the onboard equipment of each aircraft is embedded with limited knowledge about the environment and other aircrafts. Given the list of targets, the drone needs to perform under certain constraints (flight, fuel, etc. constraints).

Disclosure of Invention

The invention aims to provide a task planning method for the collaborative ground combat of a unmanned aerial vehicle, so as to solve the problems in the background technology.

In order to achieve the above purpose, the present invention provides the following technical solutions:

a task planning method for the collaborative ground combat of unmanned aerial vehicles comprises the following steps:

S1, modeling a battlefield environment: the method comprises the steps of adopting a digital map information fusion principle to fuse original topography and threat information of a combat space into comprehensive topography information;

S2, acquiring the number of man machines, the number of unmanned aerial vehicles and the target number: the set ground striking formation comprises 1 man-machine and m unmanned aerial vehicles (v ₁,v₂,…,v_m), and n targets (t ₁,t₂,…,t_n) needing striking are found;

s3, establishing a task allocation objective function: the establishment of the objective function comprehensively considers three factors including the hit value of the target, the track cost for going to the target and the threat cost received in the cruising process;

① Impact value: the target hit value reflects the importance degree of the target, the unmanned aerial vehicle hits the target with high priority, and V (t _j) is used for representing the hit value of the target t _j;

V(t_j)＝P_jR_j

Wherein, P _j is the probability function of the target t _j being attacked, R _j is the attacked priority value of the target t _j, and P _ij is the efficiency of the unmanned plane v _i to hit the target t _j, namely the probability that v _i can hit t _j successfully;

② Track cost: according to three-dimensional terrain following and unmanned aerial vehicle flight maintenance, estimating the track distance, namely the track cost, from the unmanned aerial vehicle to the target;

③ Threat cost: the longer the unmanned aerial vehicle is exposed under the threat of enemy in the flight process, the larger the probability of being destroyed, and the threat probability of the unmanned aerial vehicle in a task scene along the L flight process can be expressed as:

Wherein, Representing threat probability of ith radar in unmanned aerial vehicle under task scene,/>Representing the threat probability of the jth ground-air missile in the task scene of the unmanned aerial vehicle, wherein the ③ Chinese character is expressed as an integral form of the track L, and in the actual situation, a plurality of points can be sampled on the track point according to the calculation time requirement to discretize the time;

S4, setting task allocation constraint conditions:

① Task allocation balancing constraints: the task allocation can allocate a plurality of targets for one unmanned aerial vehicle, and can allocate one target for the combined execution of a plurality of unmanned aerial vehicles, so that the execution efficiency of the task is improved, and the task allocation is uniformly restrained;

Where n _max represents the maximum number of targets allocated to the drone v _i, and m _max is the maximum number of drones allocated to the target t _j;

② Maximum range of unmanned aerial vehicle: the unmanned aerial vehicle range should be no greater than its maximum range:

L≤L_max

Wherein L _max is the maximum range of the single unmanned aerial vehicle;

③ Target of striking is coordinated and not repeated

Wherein, theta _i and theta _j are target sets allocated to the ith and jth unmanned aerial vehicles;

S5, solving task allocation based on a contract net algorithm;

After the battle targets are determined and battle field situation information is acquired, the bidding person randomly generates a bidding sequence for each unmanned aerial vehicle and publishes a target list to be beaten, auction starts, each unmanned aerial vehicle builds all possible target battle plans according to the target list, calculates profits of the plans, when bidding is completed, the unmanned aerial vehicle selects an optimal plan from a target attack plan list according to a greedy principle, and performs bidding as an own battle plan, all unmanned aerial vehicles bid according to the bidding sequence, after all unmanned aerial vehicles bid, the bidding is completed, a target allocation plan for unmanned aerial vehicle formation is obtained, if time or resource limitation is not met, a new bidding sequence generated by the bidding person randomly is performed for a new round of bidding, so that a better solution is sought, and when time out or resource exceeds limitation, the algorithm stops;

S6, considering population initialization gene coding: a real number coding mode is selected, and chromosome genes are represented by track point coordinates;

s7, each unmanned aerial vehicle performs track planning according to the task allocation result, and a track cost function is established:

Assuming that the unmanned aerial vehicle starts from the initial position O and reaches the target position G after passing through N-1 nodes, the track cost of the unmanned aerial vehicle passing through the track can be expressed as:

J＝ω₁f_l+ωf_h+ω₃f_d

Wherein, Representing the course cost, which is the sum of the distances of adjacent track segments, and d _i,i+1 represents the distance between two adjacent track points;

f _h = ≡hdl represents high Cheng Daijia, represents the integral of the altitude on the track, in order to make the unmanned aerial vehicle fly as close to the ground as possible, Representing threat cost, wherein the ith track point is subjected to threat cost of the jth threat point, and carrying out gene coding; a real number coding mode is selected, and chromosome genes are represented by track point coordinates;

s8, establishing a track planning constraint condition:

① Minimum track segment constraints: in order to avoid the waste of oil consumption caused by frequent turning and fluctuation of the unmanned aerial vehicle, the minimum track section, namely the distance that the unmanned aerial vehicle must keep flying straight before changing the current gesture, p _i is used for representing the ith track section, l _i is the length of the ith flight track section of the unmanned aerial vehicle, and l _min is the length of the minimum track section;

l_i≥l_min,i＝1,2,3…n

② Minimum fly height constraint: the unmanned plane should fly at a position close to the ground as much as possible, but cannot impact the ground due to the fact that the altitude is too low, h _i represents the ith flight path flying altitude, h _min represents the lowest flying altitude, and the unmanned plane has the following steps of

h_i≥h_min,i＝1,2,3…n

③ Maximum turning angle constraint: because the restriction of self mobility, can only change certain angle when unmanned aerial vehicle turns, so need to carry out the restriction of biggest turning angle, not more than biggest turning angle just can realize flying to next flight path point, the smaller the turning angle flies relatively steadily too, establishes the horizontal projection of ith section flight path and is a _i＝(x_i-x_i-1,y_i-y_i-1), unmanned aerial vehicle's biggest turning angle is θ, then:

④ Maximum climb angle constraint: similar to the maximum turning angle, the maximum angle is limited when the unmanned aerial vehicle climbs or descends due to the constraint limit of climbing and diving performances of the unmanned aerial vehicle, the height difference in the longitudinal direction of the track section i is set to be |z _i-z_i-1 |, and the maximum pitch angle of the unmanned aerial vehicle is set to be Then:

S9, selecting a parent individual by adopting a roulette method;

s10, intersecting operation: the crossover operator operation divides two individuals into two parts randomly, the front half part of the individual 1 is combined with the rear half part of the individual 2, and the rear half part of the individual 1 is combined with the front half part of the individual 1 to generate two brand new individuals;

S11, mutation operation: the mutation operation refers to randomly selecting single or multiple gene positions in a chromosome from filial generations after crossing, and carrying out mutation on gene values of the positions, wherein the mutation operation can improve the local searching capability of an algorithm, and when an individual approaches to an optimal solution of a problem after crossing, the mutation action is required to adjust partial gene position values of the individual so that the individual is closer to the optimal solution; secondly, the phenomenon of premature population is prevented, population diversity is maintained, new individual coding structures can be generated through mutation, premature is effectively avoided, and from the perspective of a track, the variation in the track, namely the change of coordinate values, can assist in generating new individuals, and affects the local searching capability of a genetic algorithm;

① Inserting operation operators: when the track section passes through a dangerous area or violates the lowest track height, a new track node is randomly inserted between two adjacent nodes in the track;

② Deleting operation operators: if the unmanned aerial vehicle track does not meet the flight constraint, deleting the intermediate node of the track;

③ Exchanging operation operators: the sequence of any two adjacent nodes in the track is exchanged, the turning angle can be reduced, the method is realized through a 2-opt algorithm of local search, and if the adaptability of a new path obtained after the exchanging operation is larger than that of an original path, the path is updated;

④ Disturbance operation operator: randomly changing the coordinate value of a track node, determining the disturbance range according to whether the original track is feasible or not, and if the original track is feasible, carrying out small-range disturbance to ensure that the track is still feasible after operation; otherwise, the disturbance range should be properly enlarged, and the track enters a feasible region through disturbance operation, so that the adaptability of the track can be improved;

⑤ Smoothing operator: smoothing the nodes of the track turning angle which do not meet the yaw angle constraint of the unmanned aerial vehicle, namely selecting a certain node in the track, inserting a new node into two track sections adjacent to the certain node to replace the original node, and removing sharp corners of the track through smoothing operation;

S12, carrying out population updating through steps S10 and S11:

Replacing parent individuals by offspring generated through cross mutation, and storing individuals with high adaptability in the parent, so as to complete population updating;

S13, circulating the steps S6-S12, and outputting an optimal track when the iteration times are met;

S14, carrying out local track planning on new threats in the environment;

Under the condition of encountering sudden threat, firstly, erasing a local track exposed in an environment change area, planning a two-dimensional track, then, carrying out height planning to process the track, and splicing the track with an original track to obtain an adjusted track;

s15, carrying out two-dimensional track planning on track re-planning by using an A-algorithm, and limiting a search space;

setting the minimum step length as l _min, the maximum turning angle as theta, the expansion area of the sparse A algorithm as a sector area, the expansion angle as 2 theta, the expansion radius as l _min, and if the expansion area is divided into N equal parts, the expansion point as n+1;

s16, carrying out two-dimensional track planning on track re-planning by using an A-algorithm, and establishing a cost function;

f(n)＝g(n)+h(n)

Wherein n is a node to be expanded, g (n) is a real cost from a starting point to a current node, h (n) is a heuristic function, and represents a cost estimation value from the current node n to a target node, f (n) is an estimation function of the node to be expanded, and represents an estimation of a cost required to be paid by a certain route passing through a track node n;

S17, carrying out two-dimensional track planning on track re-planning by using an algorithm A, and according to the next node with the minimum expansion cost in the steps S14 and S15, selecting a target as an expansion node;

s18, giving a proper height value to each track point in the re-planning two-dimensional track to obtain a feasible three-dimensional track;

s19: analyzing and contrasting the obtained three-dimensional track to obtain three groups of contrasting analysis charts, listing differences in the three groups of contrasting analysis charts, and analyzing the differences to obtain an optimal three-dimensional track;

Further, in the step S1, the digital map information fusion principle is adopted to fuse the original topography and the threat information of the combat space into the comprehensive topography information, which includes:

selecting a real terrain from a digital terrain elevation database by adopting a digital elevation map with a regular network structure, and obtaining an original digital elevation map through interpolation processing;

the method is equivalent to a three-dimensional threat source map aiming at radar, air-defense firepower and non-crossing areas;

information fusion is carried out on the original digital map and the threat equivalent digital map to generate an equivalent digital map;

further, in the step S3, estimating the track by using the terrain following and the unmanned aerial vehicle flight maintenance relay includes:

The flight limitation is not considered, the terrain information is utilized to follow the unmanned aerial vehicle and the vertical section where the target is located, the unmanned aerial vehicle is kept to obtain an approximate three-dimensional course, namely, on a three-dimensional comprehensive equivalent map, a section perpendicular to a horizontal plane is made through the position where the unmanned aerial vehicle v _i is located and the position where the target t _j is located, a line where the section intersects with the terrain is used as a reference for estimating the course, and a course meeting the terrain following and flight height limitation is planned, and the length of the course is the course cost.

Further, in the step S3, task allocation is performed by using a contract net algorithm, including:

The target attack plans of the unmanned aerial vehicle are defined as ordered sets of targets, which means that the unmanned aerial vehicle will attack the targets sequentially, since the track cost and threat cost between the unmanned aerial vehicle and different target points are different, and the gains of the unmanned aerial vehicle executing the tasks in different orders are also different, in the task allocation process, each computing node makes decisions based on local information, so each unmanned aerial vehicle orders the task orders according to the principle of maximizing the effectiveness of the unmanned aerial vehicle, for the initial target setting t= { t ₁,t₂…,t_n }, according to the limit of the maximum limit task, all target attack plans can be established, for example, if v _i can attack 2 targets at maximum, the attack plan sequence can be constructed as follows {{t₁},{t₂}…{t_n},{t₁,t₂},{t₂,t₁}…,{t_n-1,t_n}}.

In the process of constructing an attack plan, in order to avoid excessive calculation amount, some unrealistic plans are screened out, and the task ordering schemes are assumed to be assembled intoV _i the validity of scheme M _i＝{t_i1,t_i2,…,t_il selected based on the principles described above is:

In the auction process, after a certain unmanned aerial vehicle bids for its own attack plan according to the bidding sequence, the value of the allocated targets will decrease, other unmanned aerial vehicles need to update the current value of all targets and recalculate the bidding validity function, then according to the updated profit bidding attack plan, it is possible to avoid too many unmanned aerial vehicles attacking the same target and obtaining higher global target avails, and assuming that unmanned aerial vehicle v _i obtains target t _j in this round of auction, each unmanned aerial vehicle bidding thereafter will follow the formula:

Value_new(t_j)＝(1-P_ij)*Value_old(t_j)

The hit value of t _j is updated and bidding is conducted using the efficacy function calculated at the new target value when bidding is rolled to obtain a more rational attack plan.

Further, in S16, the g (n) design of the heuristic function includes:

g (n) represents the actual cost of the drone at the current node n of space:

g(n)＝ω_LL_n+ω_TT_n

The representative course is the sum of the distances of adjacent track segments, assuming that the starting point S is the 0 th starting point, the current node is the N-th course point, and d _i,i+1 is the distance between two adjacent track points.

Representing threat cost, wherein the ith track point is threatened by the jth threat point, omega _L and w _T represent the weight of the course cost and the threat cost, and omega _L+ω_T =1 is satisfied;

the heuristic function h (n) is set as the Euclidean distance from the current node n (x _n,y_n) of the unmanned plane to the target node G (x _G,y_G);

Further, in S18, the planning of the height includes:

Assuming that the reference track overlapping the environment change area is S _3D(s₁,s₂,…,s_N), the number of track points is N, the height corresponding to the ith track point is H _i, the two-dimensional track re-planned by bypassing the threat is P _2D＝(p₁,p₂,…,p_n), the number of track points is N, and the value of the height corresponding to the jth track point is H _j,h_j is as follows:

Wherein the method comprises the steps of Expressed as a whole,/>T=i% k represents the remainder, and in order to enable flight path flight, the maximum climb angle limit of the unmanned aerial vehicle is considered to ensure/>S is the step length of an algorithm A, and alpha is the maximum climbing angle limit of the unmanned aerial vehicle.

Further, in S19, the three-dimensional track is subjected to comparison analysis by software, marking is performed on the software, and the marked three-dimensional track is printed.

Compared with the prior art, the invention has the beneficial effects that:

Based on analyzing the advantages and disadvantages of three typical control modes of the unmanned aerial vehicle and the unmanned aerial vehicle, namely the full centralized type, the limited centralized type and the centerless type, the invention provides a control mode of the limited centralized type unmanned aerial vehicle and the unmanned aerial vehicle formation as a basic mode of the invention, and the unmanned aerial vehicle mission planning system are respectively researched in two aspects of mission allocation and flight path planning. According to the characteristics of a limited centralized distributed control mode, the invention designs an objective function of task allocation, and a decision of pilot participation objective priority is introduced into the objective function; constructing an equivalent three-dimensional digital map of the combat space fused with terrain and ground threats; using a three-dimensional estimated range mode, and obtaining a three-dimensional estimated range which is closer to a real battlefield environment by adopting a tangential surface topography following method; on the basis, a task allocation solving method based on a contract net algorithm is used, so that a man-machine is used as a host signer for task allocation to supervise the task allocation process and authorize the task, the practical combat mode is met, the track planning task is decomposed into two parts of reference track planning and online track re-planning, and the real-time requirement of the reference track planning is low.

The invention provides an unmanned aerial vehicle reference track planning method by introducing unmanned aerial vehicle maneuvering performance constraint into a mutation operator based on a genetic algorithm capable of realizing global optimization; aiming at the characteristic of higher real-time requirement in online track planning, the re-planning method divides the re-planning into two independent modules of two-dimensional track planning and height planning, and compared with the method for directly carrying out three-dimensional space planning, the re-planning method reduces the search space and greatly saves time.

Drawings

FIG. 1 illustrates a finite centralized distributed lower combat organizational structure according to the present invention;

FIG. 2 is a schematic view showing the estimated section of the voyage in step S3 of the present invention;

FIG. 3 shows a flow chart of the task allocation algorithm in step S2 of the present invention;

FIG. 4 is a schematic diagram showing the cross operation of step S10 in the embodiment of the present invention;

FIG. 5 is a schematic diagram showing the mutation operation of step S11 in the embodiment of the present invention;

FIG. 6 shows a schematic diagram of a genetic algorithm for path planning in an embodiment of the present invention;

FIG. 7 is an expanded schematic diagram of step S15 in an embodiment of the present invention;

FIG. 8 illustrates an original three-dimensional topographical view in an embodiment of the present invention;

FIG. 9 shows a threat equivalent map in an embodiment of the invention;

FIG. 10 shows a three-dimensional topographical view of the data fusion in an embodiment of the present invention;

FIG. 11 is a schematic diagram of task allocation in an embodiment of the present invention;

FIG. 12 is a diagram illustrating overall performance of task allocation in accordance with an embodiment of the present invention;

FIG. 13 shows a graph of overall performance change in an embodiment of the invention;

FIG. 14 shows an original schematic view of the track in an embodiment of the invention;

FIG. 15 illustrates a schematic diagram of a bursty threat in an embodiment of the invention;

FIG. 16 shows a schematic diagram of track re-planning in an embodiment of the invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

In order to implement a task planning method for the collaborative ground combat of the unmanned aerial vehicle, a PC with an operating environment of Intel Core i 5.7 GHz and 8G memory is matched, and a simulation platform is MATLAB2016a.

A task planning method for the collaborative ground combat of unmanned aerial vehicles comprises the following steps:

S1, modeling a battlefield environment: the method comprises the steps of adopting a digital map information fusion principle to fuse original topography and threat information of a combat space into comprehensive topography information;

S2, acquiring the number of man machines, the number of unmanned aerial vehicles and the target number: the set ground striking formation comprises 1 man-machine and m unmanned aerial vehicles (v ₁,v₂,…,v_m), and n targets (t ₁,t₂,…,t_n) needing striking are found;

s3, establishing a task allocation objective function: the establishment of the objective function comprehensively considers three factors including the hit value of the target, the track cost for going to the target and the threat cost received in the cruising process;

① Impact value: the target hit value reflects the importance degree of the target, the unmanned aerial vehicle hits the target with high priority, and V (t _j) is used for representing the hit value of the target t _j;

V(t_j)＝P_jR_j

Wherein, P _j is the probability function of the target t _j being attacked, R _j is the attacked priority value of the target t _j, and P _ij is the efficiency of the unmanned plane v _i to hit the target t _j, namely the probability that v _i can hit t _j successfully;

② Track cost: according to three-dimensional terrain following and unmanned aerial vehicle flight maintenance, estimating the track distance, namely the track cost, from the unmanned aerial vehicle to the target;

③ Threat cost: the longer the unmanned aerial vehicle is exposed under the threat of enemy in the flight process, the larger the probability of being destroyed, and the threat probability of the unmanned aerial vehicle in a task scene along the L flight process can be expressed as:

Wherein, Representing threat probability of ith radar in unmanned aerial vehicle under task scene,/>Representing the threat probability of the jth ground-air missile in the task scene of the unmanned aerial vehicle, wherein the ③ Chinese character is expressed as an integral form of the track L, and in the actual situation, a plurality of points can be sampled on the track point according to the calculation time requirement to discretize the time;

S4, setting task allocation constraint conditions:

① Task allocation balancing constraints: the task allocation can allocate a plurality of targets for one unmanned aerial vehicle, and can allocate one target for the combined execution of a plurality of unmanned aerial vehicles, so that the execution efficiency of the task is improved, and the task allocation is uniformly restrained;

Where n _max represents the maximum number of targets allocated to the drone v _i, and m _max is the maximum number of drones allocated to the target t _j;

② Maximum range of unmanned aerial vehicle: the unmanned aerial vehicle range should be no greater than its maximum range:

L≤L_max

Wherein L _max is the maximum range of the single unmanned aerial vehicle;

③ Target of striking is coordinated and not repeated

Wherein, theta _i and theta _j are target sets allocated to the ith and jth unmanned aerial vehicles;

S5, solving task allocation based on a contract net algorithm;

After the battle targets are determined and battle field situation information is acquired, the bidding person randomly generates a bidding sequence for each unmanned aerial vehicle and publishes a target list to be beaten, auction starts, each unmanned aerial vehicle builds all possible target battle plans according to the target list, calculates profits of the plans, when bidding is completed, the unmanned aerial vehicle selects an optimal plan from a target attack plan list according to a greedy principle, and performs bidding as an own battle plan, all unmanned aerial vehicles bid according to the bidding sequence, after all unmanned aerial vehicles bid, the bidding is completed, a target allocation plan for unmanned aerial vehicle formation is obtained, if time or resource limitation is not met, a new bidding sequence generated by the bidding person randomly is performed for a new round of bidding, so that a better solution is sought, and when time out or resource exceeds limitation, the algorithm stops;

S6, considering population initialization gene coding: a real number coding mode is selected, and chromosome genes are represented by track point coordinates;

s7, each unmanned aerial vehicle performs track planning according to the task allocation result, and a track cost function is established:

Assuming that the unmanned aerial vehicle starts from the initial position O and reaches the target position G after passing through N-1 nodes, the track cost of the unmanned aerial vehicle passing through the track can be expressed as:

J＝ω₁f_l+ωf_h+ω₃f_d

Wherein, Representing the course cost, which is the sum of the distances of adjacent track segments, and d _i,i+1 represents the distance between two adjacent track points;

f _h = ≡hdl represents high Cheng Daijia, represents the integral of the altitude on the track, in order to make the unmanned aerial vehicle fly as close to the ground as possible, Representing threat cost, wherein the ith track point is subjected to threat cost of the jth threat point, and carrying out gene coding; a real number coding mode is selected, and chromosome genes are represented by track point coordinates;

s8, establishing a track planning constraint condition:

① Minimum track segment constraints: in order to avoid the waste of oil consumption caused by frequent turning and fluctuation of the unmanned aerial vehicle, the minimum track section, namely the distance that the unmanned aerial vehicle must keep flying straight before changing the current gesture, p _i is used for representing the ith track section, l _i is the length of the ith flight track section of the unmanned aerial vehicle, and l _min is the length of the minimum track section;

l_i≥l_min,i＝1,2,3…n

② Minimum fly height constraint: the unmanned plane should fly at a position close to the ground as much as possible, but cannot impact the ground due to the fact that the altitude is too low, h _i represents the ith flight path flying altitude, h _min represents the lowest flying altitude, and the unmanned plane has the following steps of

h_i≥h_min,i＝1,2,3…n

③ Maximum turning angle constraint: because the restriction of self mobility, can only change certain angle when unmanned aerial vehicle turns, so need to carry out the restriction of biggest turning angle, not more than biggest turning angle just can realize flying to next flight path point, the smaller the turning angle flies relatively steadily too, establishes the horizontal projection of ith section flight path and is a _i＝(x_i-x_i-1,y_i-y_i-1), unmanned aerial vehicle's biggest turning angle is θ, then:

④ Maximum climb angle constraint: similar to the maximum turning angle, the maximum angle is limited when the unmanned aerial vehicle climbs or descends due to the constraint limit of climbing and diving performances of the unmanned aerial vehicle, the height difference in the longitudinal direction of the track section i is set to be |z _i-z_i-1 |, and the maximum pitch angle of the unmanned aerial vehicle is set to be Then:

S9, selecting a parent individual by adopting a roulette method;

s10, intersecting operation: the crossover operator operation divides two individuals into two parts randomly, the front half part of the individual 1 is combined with the rear half part of the individual 2, and the rear half part of the individual 1 is combined with the front half part of the individual 1 to generate two brand new individuals;

S11, mutation operation: the mutation operation refers to randomly selecting single or multiple gene positions in a chromosome from filial generations after crossing, and carrying out mutation on gene values of the positions, wherein the mutation operation can improve the local searching capability of an algorithm, and when an individual approaches to an optimal solution of a problem after crossing, the mutation action is required to adjust partial gene position values of the individual so that the individual is closer to the optimal solution; secondly, the phenomenon of premature population is prevented, population diversity is maintained, new individual coding structures can be generated through mutation, premature is effectively avoided, and from the perspective of a track, the variation in the track, namely the change of coordinate values, can assist in generating new individuals, and affects the local searching capability of a genetic algorithm;

① Inserting operation operators: when the track section passes through a dangerous area or violates the lowest track height, a new track node is randomly inserted between two adjacent nodes in the track;

② Deleting operation operators: if the unmanned aerial vehicle track does not meet the flight constraint, deleting the intermediate node of the track;

③ Exchanging operation operators: the sequence of any two adjacent nodes in the track is exchanged, the turning angle can be reduced, the method is realized through a 2-opt algorithm of local search, and if the adaptability of a new path obtained after the exchanging operation is larger than that of an original path, the path is updated;

④ Disturbance operation operator: randomly changing the coordinate value of a track node, determining the disturbance range according to whether the original track is feasible or not, and if the original track is feasible, carrying out small-range disturbance to ensure that the track is still feasible after operation; otherwise, the disturbance range should be properly enlarged, and the track enters a feasible region through disturbance operation, so that the adaptability of the track can be improved;

⑤ Smoothing operator: smoothing the nodes of the track turning angle which do not meet the yaw angle constraint of the unmanned aerial vehicle, namely selecting a certain node in the track, inserting a new node into two track sections adjacent to the certain node to replace the original node, and removing sharp corners of the track through smoothing operation;

S12, carrying out population updating through steps S10 and S11:

Replacing parent individuals by offspring generated through cross mutation, and storing individuals with high adaptability in the parent, so as to complete population updating;

S13, circulating the steps S6-S12, and outputting an optimal track when the iteration times are met;

S14, carrying out local track planning on new threats in the environment;

Under the condition of encountering sudden threat, firstly, erasing a local track exposed in an environment change area, planning a two-dimensional track, then, carrying out height planning to process the track, and splicing the track with an original track to obtain an adjusted track;

s15, carrying out two-dimensional track planning on track re-planning by using an A-algorithm, and limiting a search space;

setting the minimum step length as l _min, the maximum turning angle as theta, the expansion area of the sparse A algorithm as a sector area, the expansion angle as 2 theta, the expansion radius as l _min, and if the expansion area is divided into N equal parts, the expansion point as n+1;

s16, carrying out two-dimensional track planning on track re-planning by using an A-algorithm, and establishing a cost function;

f(n)＝g(n)+h(n)

Wherein n is a node to be expanded, g (n) is a real cost from a starting point to a current node, h (n) is a heuristic function, and represents a cost estimation value from the current node n to a target node, f (n) is an estimation function of the node to be expanded, and represents an estimation of a cost required to be paid by a certain route passing through a track node n;

S17, carrying out two-dimensional track planning on track re-planning by using an algorithm A, and according to the next node with the minimum expansion cost in the steps S14 and S15, selecting a target as an expansion node;

s18, giving a proper height value to each track point in the re-planning two-dimensional track to obtain a feasible three-dimensional track;

s19: analyzing and contrasting the obtained three-dimensional track to obtain three groups of contrasting analysis charts, listing differences in the three groups of contrasting analysis charts, and analyzing the differences to obtain an optimal three-dimensional track;

The terrain closing information includes:

selecting a real terrain from a digital terrain elevation database by adopting a digital elevation map with a regular network structure, and obtaining an original digital elevation map through interpolation processing;

the method is equivalent to a three-dimensional threat source map aiming at radar, air-defense firepower and non-crossing areas;

and carrying out information fusion on the original digital map and the threat equivalent digital map to generate the equivalent digital map.

In the step S3, the track estimation is carried out by utilizing the terrain following and the unmanned aerial vehicle flight maintenance relay, and the method comprises the following steps:

the flight limitation is not considered, the terrain information is utilized to follow the unmanned aerial vehicle and the vertical section where the target is located, the unmanned aerial vehicle is kept to obtain an approximate three-dimensional course, namely, on a three-dimensional comprehensive equivalent map, a section perpendicular to a horizontal plane is made through the position where the unmanned aerial vehicle vi is located and the position where the target t _j is located, a line where the section intersects with the terrain is used as a reference for estimating the course, and a course meeting the terrain following and flight height limitation is planned, and the length of the course is the course cost.

In the step S3, task allocation is performed by using a contract net algorithm, which comprises the following steps:

The target attack plans of the unmanned aerial vehicle are defined as ordered sets of targets, which means that the unmanned aerial vehicle will attack the targets sequentially, since the track cost and threat cost between the unmanned aerial vehicle and different target points are different, and the gains of the unmanned aerial vehicle executing the tasks in different orders are also different, in the task allocation process, each computing node makes decisions based on local information, so each unmanned aerial vehicle orders the task orders according to the principle of maximizing the effectiveness of the unmanned aerial vehicle, for the initial target setting t= { t ₁,t₂…,t_n }, according to the limit of the maximum limit task, all target attack plans can be established, for example, if v _i can attack 2 targets at maximum, the attack plan sequence can be constructed as follows {{t₁},{t₂}…{t_n},{t₁,t₂},{t₂,t₁}…,{t_n-1,t_n}}.

In the process of constructing an attack plan, in order to avoid excessive calculation amount, some unrealistic plans are screened out, and the task ordering schemes are assumed to be assembled intoV _i the validity of scheme M _i＝{t_i1,t_i2,…,t_il selected based on the principles described above is:

In the auction process, after a certain unmanned aerial vehicle bids for its own attack plan according to the bidding sequence, the value of the allocated targets will decrease, other unmanned aerial vehicles need to update the current value of all targets and recalculate the bidding validity function, then according to the updated profit bidding attack plan, it is possible to avoid too many unmanned aerial vehicles attacking the same target and obtaining higher global target avails, and assuming that unmanned aerial vehicle v _i obtains target t _j in this round of auction, each unmanned aerial vehicle bidding thereafter will follow the formula:

Value_new(t_j)＝(1-P_ij)*Value_old(t_j)

The hit value of t _j is updated and bidding is conducted using the efficacy function calculated at the new target value when bidding is rolled to obtain a more rational attack plan.

In S16, the g (n) design of the heuristic function includes:

g (n) represents the actual cost of the drone at the current node n of space:

g(n)＝ω_LL_n+ω_TT_n

The representative course is the sum of the distances of adjacent track segments, assuming that the starting point S is the 0 th starting point, the current node is the N-th course point, and d _i,i+1 is the distance between two adjacent track points.

Representing threat cost, wherein the ith track point is threatened by the jth threat point, omega _L and w _T represent the weight of the course cost and the threat cost, and omega _L+ω_T =1 is satisfied;

the heuristic function h (n) is set as the Euclidean distance from the current node n (x _n,y_n) of the unmanned plane to the target node G (x _G,y_G);

in S18, the planning of the height includes:

Assuming that the reference track overlapping the environment change area is S _3D(s₁,s₂,…,s_N), the number of track points is N, the height corresponding to the ith track point is H _i, the two-dimensional track re-planned by bypassing the threat is P _2D＝(p₁,p₂,…,p_n), the number of track points is N, and the value of the height corresponding to the jth track point is H _j,h_j is as follows:

Wherein the method comprises the steps of Expressed as a whole,/>T=i% k represents the remainder, and in order to enable flight path flight, the maximum climb angle limit of the unmanned aerial vehicle is considered to ensure/>S is the step length of an algorithm A, and alpha is the maximum climbing angle limit of the unmanned aerial vehicle.

In S19, the three-dimensional track is subjected to contrast analysis by software, marking is performed on the software, and the marked three-dimensional track is printed.

In this embodiment, selecting a real terrain from a digital terrain elevation database, planning a space of 200km by 200km, and performing interpolation processing to obtain fig. 8; the threat model of the radar and the ground-to-air missile is simplified and expressed by Gaussian distribution, and the expression is as follows:

Wherein x and y represent coordinate values of the threat projected onto a horizontal plane, and z _i is an elevation value corresponding to the threat; x _i and y _i represent coordinates of the ith threat center, x _si and y _si represent attenuation amounts of the ith threat related to directions along the x-axis and the y-axis, and h _i represents action intensity of the ith threat;

The air-defense cannon is equivalent to be hemispherical, and the threat equivalent map is shown in figure 9; information fusion is carried out on the original digital map and the threat equivalent digital map to generate a three-dimensional topographic map after data fusion, as shown in fig. 10;

acquiring the positions of the organic machine, the unmanned machine and the target according to the task,

Table 1 initial position coordinates of unmanned aerial vehicle

The threat cost of the unmanned aerial vehicle is calculated, the probability of the unmanned aerial vehicle being found when the unmanned aerial vehicle is away from the radar antenna distance R outside the no-fly zone R _min is attenuated at the speed of the 4 th side of the distance, and the probability is approximately as follows:

wherein K is a probability attenuation coefficient;

the effective killing probability formula of the ground-to-air missile is approximately expressed as:

P_M＝K₀(Δh/R)

Wherein K ₀ represents the killing probability of the missile under the condition of sunny weather, which is generally regarded as a constant, and Δh represents the height of the unmanned aerial vehicle relative to the ground-to-air missile battle array; r represents the radial distance between the unmanned aerial vehicle and the missile battle ground;

Calculating an objective function of task allocation;

Setting the maximum auction times of task allocation as 20 times, and carrying out contract net algorithm solution to obtain a task allocation schematic diagram shown in figure 11 and a task allocation overall efficiency schematic diagram shown in figure 12; the overall efficiency change curve of the task allocation scheme is shown in fig. 13, the overall efficiency is continuously increased along with the increase of bidding times, the overall efficiency is close to a fixed value after 15 bidding times, the whole process takes 2.651 seconds, each target can find the most suitable unmanned aerial vehicle to finish after task allocation, the overall efficiency is maximized through maximizing each local efficiency, the time is short, and the requirement of a rapid battlefield is met.

Setting the initial main population size as 100, the auxiliary population size as 100 and the propagation pool size as 40 in track planning; the crossover probability is 0.7, and the mutation probability is 0.1; the maximum number of evolutions is 500; the maximum turning angle is 45 degrees, and the maximum pitch angle is 45 degrees; the minimum fly height is 10 meters.

The evolution operator and the fitness function constructed according to the invention can meet the requirement that the unmanned aerial vehicle flight path is feasible, the flying height is not too high, the threat is avoided, the safety requirement is met, the objective function can be gradually converged along with the increase of iteration times, and the higher flight path precision is obtained due to the higher dimensionality of the initialized population.

In order to verify the performance of the re-planning of the track task, an unmanned plane is set to execute a certain target, the target flies along a reference track marked by genetic algorithm rules, two new threats are newly added on the track, threat 1 coordinates are (69,88) threat radius is 30, threat 2 coordinates are (143,177) threat radius is 40; the original track is shown in fig. 14, the burst track is shown in fig. 15, and the track re-planning is shown in fig. 16.

According to the flight path re-planning method provided by the invention, the unmanned aerial vehicle can avoid the threat, a flight path bypassing the threat from the side is re-planned, the unmanned aerial vehicle returns to the original reference flight path to continue flying to reach the final target point after reaching the target point of the local planning, and the online flight path planning is realized; and the planning takes 1.110 seconds, and under the same condition, the three-dimensional sparse A algorithm is used for track planning, and the time is 2.335 seconds, so that the time is saved, and the requirement of re-planning on the speed can be met.

It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.

Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.

Claims (7)

1. The task planning method for the collaborative ground combat of the unmanned aerial vehicle is characterized by comprising the following steps of:

S1, modeling a battlefield environment: the method comprises the steps of adopting a digital map information fusion principle to fuse original topography and threat information of a combat space into comprehensive topography information;

S2, acquiring the number of man machines, the number of unmanned aerial vehicles and the target number: the set ground striking formation comprises 1 man-machine and m unmanned aerial vehicles (v ₁,v₂,…,v_m), and n targets (t ₁,t₂,…,t_n) needing striking are found;

s3, establishing a task allocation objective function: the establishment of the objective function comprehensively considers three factors including the hit value of the target, the track cost for going to the target and the threat cost received in the cruising process;

① Impact value: the target hit value reflects the importance degree of the target, the unmanned aerial vehicle hits the target with high priority, and V (t _j) is used for representing the hit value of the target t _j;

V(t_j)＝P_jR_j

Wherein, P _j is the probability function of the target t _j being attacked, R _j is the attacked priority value of the target t _j, and P _ij is the efficiency of the unmanned plane v _i to hit the target t _j, namely the probability that v _i can hit t _j successfully;

② Track cost: according to three-dimensional terrain following and unmanned aerial vehicle flight maintenance, estimating the track distance, namely the track cost, from the unmanned aerial vehicle to the target;

③ Threat cost: the longer the unmanned aerial vehicle is exposed under the threat of enemy in the flight process, the larger the probability of being destroyed, and the threat probability of the unmanned aerial vehicle in a task scene along the L flight process can be expressed as:

Wherein, Representing threat probability of ith radar in unmanned aerial vehicle under task scene,/>Representing the threat probability of the jth ground-air missile in the task scene of the unmanned aerial vehicle, wherein the ③ Chinese character is expressed as an integral form of the track L, and in the actual situation, a plurality of points can be sampled on the track point according to the calculation time requirement to discretize the time;

S4, setting task allocation constraint conditions:

① Task allocation balancing constraints: the task allocation can allocate a plurality of targets for one unmanned aerial vehicle, and can allocate one target for the combined execution of a plurality of unmanned aerial vehicles, so that the execution efficiency of the task is improved, and the task allocation is uniformly restrained;

Where n _max represents the maximum number of targets allocated to the drone v _i, and m _max is the maximum number of drones allocated to the target t _j;

② Maximum range of unmanned aerial vehicle: the unmanned aerial vehicle range should be no greater than its maximum range:

L≤L_max

Wherein L _max is the maximum range of the single unmanned aerial vehicle;

③ Target of striking is coordinated and not repeated

Wherein, theta _i and theta _j are target sets allocated to the ith and jth unmanned aerial vehicles;

S5, solving task allocation based on a contract net algorithm;

After the battle targets are determined and battle field situation information is acquired, the bidding person randomly generates a bidding sequence for each unmanned aerial vehicle and publishes a target list to be beaten, auction starts, each unmanned aerial vehicle builds all possible target battle plans according to the target list, calculates profits of the plans, when bidding is completed, the unmanned aerial vehicle selects an optimal plan from a target attack plan list according to a greedy principle, and performs bidding as an own battle plan, all unmanned aerial vehicles bid according to the bidding sequence, after all unmanned aerial vehicles bid, the bidding is completed, a target allocation plan for unmanned aerial vehicle formation is obtained, if time or resource limitation is not met, a new bidding sequence generated by the bidding person randomly is performed for a new round of bidding, so that a better solution is sought, and when time out or resource exceeds limitation, the algorithm stops;

S6, considering population initialization gene coding: a real number coding mode is selected, and chromosome genes are represented by track point coordinates;

s7, each unmanned aerial vehicle performs track planning according to the task allocation result, and a track cost function is established:

Assuming that the unmanned aerial vehicle starts from the initial position O and reaches the target position G after passing through N-1 nodes, the track cost of the unmanned aerial vehicle passing through the track can be expressed as:

J＝ω₁f_l+ωf_h+ω₃f_d

Wherein, Representing the course cost, which is the sum of the distances of adjacent track segments, and d _i,i+1 represents the distance between two adjacent track points;

f _h = ≡hdl represents high Cheng Daijia, represents the integral of the altitude on the track, in order to make the unmanned aerial vehicle fly as close to the ground as possible, Representing threat cost, wherein the ith track point is subjected to threat cost of the jth threat point, and carrying out gene coding; a real number coding mode is selected, and chromosome genes are represented by track point coordinates;

s8, establishing a track planning constraint condition:

① Minimum track segment constraints: in order to avoid the waste of oil consumption caused by frequent turning and fluctuation of the unmanned aerial vehicle, the minimum track section, namely the distance that the unmanned aerial vehicle must keep flying straight before changing the current gesture, p _i is used for representing the ith track section, l _i is the length of the ith flight track section of the unmanned aerial vehicle, and l _min is the length of the minimum track section;

l_i≥l_min,i＝1,2,3…n

② Minimum fly height constraint: the unmanned plane should fly at a position close to the ground as much as possible, but cannot impact the ground due to the fact that the altitude is too low, h _i represents the ith flight path flying altitude, h _min represents the lowest flying altitude, and the unmanned plane has the following steps of

h_i≥h_min,i＝1,2,3…n

③ Maximum turning angle constraint: because the restriction of self mobility, can only change certain angle when unmanned aerial vehicle turns, so need to carry out the restriction of biggest turning angle, not more than biggest turning angle just can realize flying to next flight path point, the smaller the turning angle flies relatively steadily too, establishes the horizontal projection of ith section flight path and is a _i＝(x_i-x_i-1,y_i-y_i-1), unmanned aerial vehicle's biggest turning angle is θ, then:

④ Maximum climb angle constraint: similar to the maximum turning angle, the maximum angle is limited when the unmanned aerial vehicle climbs or descends due to the constraint limit of climbing and diving performances of the unmanned aerial vehicle, the height difference in the longitudinal direction of the track section i is set to be |z _i-z_i-1 |, and the maximum pitch angle of the unmanned aerial vehicle is set to be Then:

S9, selecting a parent individual by adopting a roulette method;

s10, intersecting operation: the crossover operator operation divides two individuals into two parts randomly, the front half part of the individual 1 is combined with the rear half part of the individual 2, and the rear half part of the individual 1 is combined with the front half part of the individual 1 to generate two brand new individuals;

S11, mutation operation: the mutation operation refers to randomly selecting single or multiple gene positions in a chromosome from filial generations after crossing, and carrying out mutation on gene values of the positions, wherein the mutation operation can improve the local searching capability of an algorithm, and when an individual approaches to an optimal solution of a problem after crossing, the mutation action is required to adjust partial gene position values of the individual so that the individual is closer to the optimal solution; secondly, the phenomenon of premature population is prevented, population diversity is maintained, new individual coding structures can be generated through mutation, premature is effectively avoided, and from the perspective of a track, the variation in the track, namely the change of coordinate values, can assist in generating new individuals, and affects the local searching capability of a genetic algorithm;

① Inserting operation operators: when the track section passes through a dangerous area or violates the lowest track height, a new track node is randomly inserted between two adjacent nodes in the track;

② Deleting operation operators: if the unmanned aerial vehicle track does not meet the flight constraint, deleting the intermediate node of the track;

③ Exchanging operation operators: the sequence of any two adjacent nodes in the track is exchanged, the turning angle can be reduced, the method is realized through a 2-opt algorithm of local search, and if the adaptability of a new path obtained after the exchanging operation is larger than that of an original path, the path is updated;

④ Disturbance operation operator: randomly changing the coordinate value of a track node, determining the disturbance range according to whether the original track is feasible or not, and if the original track is feasible, carrying out small-range disturbance to ensure that the track is still feasible after operation; otherwise, the disturbance range should be properly enlarged, and the track enters a feasible region through disturbance operation, so that the adaptability of the track can be improved;

⑤ Smoothing operator: smoothing the nodes of the track turning angle which do not meet the yaw angle constraint of the unmanned aerial vehicle, namely selecting a certain node in the track, inserting a new node into two track sections adjacent to the certain node to replace the original node, and removing sharp corners of the track through smoothing operation;

S12, carrying out population updating through steps S10 and S11:

Replacing parent individuals by offspring generated through cross mutation, and storing individuals with high adaptability in the parent, so as to complete population updating;

S13, circulating the steps S6-S12, and outputting an optimal track when the iteration times are met;

S14, carrying out local track planning on new threats in the environment;

Under the condition of encountering sudden threat, firstly, erasing a local track exposed in an environment change area, planning a two-dimensional track, then, carrying out height planning to process the track, and splicing the track with an original track to obtain an adjusted track;

s15, carrying out two-dimensional track planning on track re-planning by using an A-algorithm, and limiting a search space;

setting the minimum step length as l _min, the maximum turning angle as theta, the expansion area of the sparse A algorithm as a sector area, the expansion angle as 2 theta, the expansion radius as l _min, and if the expansion area is divided into N equal parts, the expansion point as n+1;

s16, carrying out two-dimensional track planning on track re-planning by using an A-algorithm, and establishing a cost function;

f(n)＝g(n)+h(n)

Wherein n is a node to be expanded, g (n) is a real cost from a starting point to a current node, h (n) is a heuristic function, and represents a cost estimation value from the current node n to a target node, f (n) is an estimation function of the node to be expanded, and represents an estimation of a cost required to be paid by a certain route passing through a track node n;

S17, carrying out two-dimensional track planning on track re-planning by using an algorithm A, and according to the next node with the minimum expansion cost in the steps S14 and S15, selecting a target as an expansion node;

s18, giving a proper height value to each track point in the re-planning two-dimensional track to obtain a feasible three-dimensional track;

S19: and (3) analyzing and contrasting the obtained three-dimensional track to obtain three groups of contrasting analysis charts, listing differences in the three groups of contrasting analysis charts, and analyzing the differences to obtain the optimal three-dimensional track.

2. The task planning method for the collaborative land battle of the unmanned aerial vehicle according to claim 1, wherein in S1, the digital map information fusion principle is adopted to fuse the original topography and the threat information of the battle space into the comprehensive topography information, which comprises the following steps:

selecting a real terrain from a digital terrain elevation database by adopting a digital elevation map with a regular network structure, and obtaining an original digital elevation map through interpolation processing;

the method is equivalent to a three-dimensional threat source map aiming at radar, air-defense firepower and non-crossing areas;

and carrying out information fusion on the original digital map and the threat equivalent digital map to generate the equivalent digital map.

3. The mission planning method for cooperation of unmanned aerial vehicles against ground according to claim 1, wherein in S3, estimating a track by using terrain following and unmanned aerial vehicle flight maintenance relay comprises:

The flight limitation is not considered, the terrain information is utilized to follow the unmanned aerial vehicle and the vertical section where the target is located, the unmanned aerial vehicle is kept to obtain an approximate three-dimensional course, namely, on a three-dimensional comprehensive equivalent map, a section perpendicular to a horizontal plane is made through the position where the unmanned aerial vehicle v _i is located and the position where the target t _j is located, a line where the section intersects with the terrain is used as a reference for estimating the course, and a course meeting the terrain following and flight height limitation is planned, and the length of the course is the course cost.

4. The task planning method for the collaborative ground combat of the unmanned aerial vehicle according to claim 1, wherein in S3, task allocation is performed by using a contractual network algorithm, comprising:

The unmanned aerial vehicle target attack plan is defined as an ordered set of targets, which means that the unmanned aerial vehicle will attack the targets sequentially, and because the track cost and threat cost between the unmanned aerial vehicle and different target points are different, the unmanned aerial vehicle executes the tasks in different orders, and in the task allocation process, each computing node makes a decision based on local information, so that each unmanned aerial vehicle sequences the tasks according to the principle of maximizing the effectiveness of the unmanned aerial vehicle, and for the initial target setting t= { t ₁,t₂…,t_n }, all target attack plans can be established according to the limitation of the maximum limit task;

in the process of constructing an attack plan, in order to avoid excessive calculation amount, some unrealistic plans are screened out, and the task ordering schemes are assumed to be assembled into V _i the validity of scheme M _i＝{t_i1,t_i2,…,t_il selected based on the principles described above is:

In the auction process, after a certain unmanned aerial vehicle bids for its own attack plan according to the bidding sequence, the value of the allocated targets will decrease, other unmanned aerial vehicles need to update the current value of all targets and recalculate the bidding validity function, then according to the updated profit bidding attack plan, it is possible to avoid too many unmanned aerial vehicles attacking the same target and obtaining higher global target avails, and assuming that unmanned aerial vehicle v _i obtains target t _j in this round of auction, each unmanned aerial vehicle bidding thereafter will follow the formula:

Value_new(t_j)＝(1-P_ij)*Value_old(t_j)

The hit value of t _j is updated and bidding is conducted using the efficacy function calculated at the new target value when bidding is rolled to obtain a more rational attack plan.

5. The mission planning method for collaborative ground combat of unmanned aerial vehicle according to claim 1, wherein in S16, designing g (n) of heuristic functions comprises:

g (n) represents the actual cost of the drone at the current node n of space:

g(n)＝ω_LL_n+ω_TT_n

representing the course, which is the sum of the distances of adjacent track segments, assuming that the starting point S is the 0 th starting point, the current node is the N-th course point, and d _i,i+1 is the distance between two adjacent track points;

Representing threat cost, wherein the ith track point is threatened by the jth threat point, omega _L and w _T represent the weight of the course cost and the threat cost, and omega _L+ω_T =1 is satisfied;

the heuristic function h (n) is set as the Euclidean distance from the current node n (x _n,y_n) of the unmanned plane to the target node G (x _G,y_G);

6. The mission planning method for collaborative ground combat of unmanned aerial vehicle according to claim 1, wherein in S18, the planning of the altitude comprises:

Assuming that the reference track overlapping the environment change area is S _3D(s₁,s₂,…,s_N), the number of track points is N, the height corresponding to the ith track point is H _i, the two-dimensional track re-planned by bypassing the threat is P _2D＝(p₁,p₂,…,p_n), the number of track points is N, and the value of the height corresponding to the jth track point is H _j,h_j is as follows:

Wherein the method comprises the steps of Expressed as a whole,/>T=i% k represents the remainder, and in order to enable flight path flight, the maximum climb angle limit of the unmanned aerial vehicle is considered to ensure/>S is the step length of an algorithm A, and alpha is the maximum climbing angle limit of the unmanned aerial vehicle.

7. The mission planning method for collaborative ground combat of unmanned aerial vehicle according to claim 1, wherein in S19, the three-dimensional trajectory is analyzed by software, marked on the software, and the marked three-dimensional trajectory is printed.