CN117647997B - Knowledge bidirectional migration unmanned aerial vehicle collaborative track local re-planning method and system - Google Patents

Abstract

The invention belongs to the field of unmanned path planning, in particular relates to a method and a system for planning a cooperative track local re-plan of an unmanned plane with knowledge bidirectional migration, and aims to solve the problems that the existing unmanned plane path planning has insufficient capability for coping with complex dynamic environments and cannot meet the real-time performance of coping with emergencies. The invention comprises the following steps: when the coarse position and threat value of the mobile threat source are detected to meet the unmanned aerial vehicle angle obstacle avoidance judging condition, constructing a reference point space based on the unmanned aerial vehicle and the task point tangent plane, acquiring multiple separation spaces through a space separation method based on the reference point space, and further determining the fine position of the mobile threat source; and generating a plurality of unmanned aerial vehicle distribution initial schemes through a differential evolution algorithm, and acquiring an unmanned aerial vehicle optimal planning scheme through a knowledge bidirectional migration method. According to the distribution scheme with better convergence in the initial distribution scheme, the invention provides a strategy of knowledge bidirectional migration, and improves the planning efficiency of the planning system.

Description

Knowledge bidirectional migration unmanned aerial vehicle collaborative track local re-planning method and system

Technical Field

The invention belongs to the field of unmanned path planning, and particularly relates to a method and a system for collaborative track local re-planning of an unmanned plane with knowledge bidirectional migration.

Background

Along with the rapid development of unmanned aerial vehicle technology, unmanned aerial vehicle has lower cost, stronger intelligence and better flexibility for unmanned aerial vehicle has obtained very extensive application, for example: enemy information detection, target persistence detection, electronic adversary of enemy and other fields. But at the same time comes with two important difficulties: 1) The unmanned aerial vehicle processes some complex tasks together through a cooperative technology, so that the execution efficiency of the unmanned aerial vehicle planning system is lower and lower; 2) The working scene of the unmanned aerial vehicle is more and more complex, and particularly the influence of the dynamic environment on the unmanned aerial vehicle track causes the initial track of the unmanned aerial vehicle to fail. Aiming at the collaborative flight path planning of multiple unmanned aerial vehicles in a dynamic environment, the prior art often considers the influence of a two-dimensional plane threat source on the flight path planning of the multiple unmanned aerial vehicles, and ignores the real-time influence of ordinate data on the unmanned aerial vehicles; meanwhile, in the planning process, the unmanned aerial vehicles are often regarded as one particle, the change of the relative position and the relative speed between the unmanned aerial vehicles in the three-dimensional space along with time is ignored, and then the unmanned aerial vehicles cannot perform real-time obstacle avoidance detection; finally, the change of the dynamic environment greatly affects the collaborative track planning of the unmanned aerial vehicle, and the initial track cannot be used as a reference in a transformation time period. Therefore, the method has important significance in researching the problem of local re-planning of the collaborative flight path of the multiple unmanned aerial vehicles in the three-dimensional dynamic environment.

Disclosure of Invention

In order to solve the problems in the prior art, namely that the existing unmanned aerial vehicle path planning has insufficient capability of coping with complex dynamic environments and cannot meet the problem of real-time coping with emergencies, the invention provides a method for collaborative flight path local re-planning of an unmanned aerial vehicle with knowledge bidirectional migration, which comprises the following steps:

step S1, acquiring the position of each unmanned aerial vehicle, and constructing a unmanned aerial vehicle discrete model;

monitoring the coarse position and threat value of the mobile threat source in real time;

s2, constructing unmanned aerial vehicle angle obstacle avoidance judging conditions based on the unmanned aerial vehicle discrete model, further determining a rapid reaction type obstacle avoidance strategy, and setting constraint conditions and an objective function;

step S3, when the coarse position and threat value of the mobile threat source are detected to meet the unmanned plane angle obstacle avoidance judging condition, constructing a reference point space based on the unmanned plane and the task point tangent plane, acquiring multiple separation spaces through a space separation method based on the reference point space, and further determining the fine position of the mobile threat source;

step S4, constructing a road-finding vertical section based on the fine position of the mobile threat source and the position of the current unmanned aerial vehicle, and generating a plurality of unmanned aerial vehicle distribution initial schemes meeting an objective function and a quick response obstacle avoidance strategy through a differential evolution algorithm and under constraint conditions;

And S5, acquiring an optimal planning scheme of the unmanned aerial vehicle by a knowledge bidirectional migration method based on the initial scheme of the unmanned aerial vehicle distribution, and taking the optimal planning scheme as a local re-planning path.

Further, the unmanned aerial vehicle discrete model comprises:

the unmanned aerial vehicle position, the unmanned aerial vehicle speed, the unmanned aerial vehicle relative position and the unmanned aerial vehicle relative included angle;

wherein, unmanned aerial vehicle position is:

L _i (t)＝[x _i (t)，y _i (t)，z _i (t)] ^T ；

L _i (t) represents the position of the ith unmanned aerial vehicle at any t moment, x _i (t) represents the position of the ith unmanned aerial vehicle on the x axis in the arbitrary t moment space, y _i (t) represents the position of the ith unmanned aerial vehicle on the y-axis in the arbitrary t-moment space, z _i (T) represents the z-axis position of the ith unmanned aerial vehicle in any T moment space, and T represents transposition;

the speed of the unmanned aerial vehicle is as follows:

V _i (t)＝[v _x，i (t)，v _y，i (t)，v _z，i (t)] ^T ；

V _i (t) represents the speed, v, of the ith unmanned aerial vehicle at any time t _x，i (t) represents the speed of the ith unmanned aerial vehicle in the x direction at any time t, v _y，i (t) represents the speed of the ith unmanned aerial vehicle in the y direction at any time t，v _z，i (t) represents the speed of the ith unmanned aerial vehicle in the z direction at any time t;

the unmanned aerial vehicle position at the next time t+1 is:

L _i (t+1)＝A·L _i (t)+B·u _i (t)；

L _i (t+1) represents the position of the ith unmanned aerial vehicle at the time t+1, A represents the real-time state matrix of the unmanned aerial vehicle, B represents the input matrix, u _i (t) represents the control amount received by the ith unmanned aerial vehicle at the time t; the real-time state matrix of the unmanned aerial vehicle comprises speed and acceleration; the input matrix is an instruction of the control quantity to the unmanned aerial vehicle;

the relative positions of the unmanned aerial vehicle are as follows:

t ₀ indicating the initial time, L _ij (t ₀ ) Indicating that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle are at the initial time t ₀ V of the relative positions of (2) _ij (t ₀ ) Indicating that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle are at the initial time t ₀ Relative velocity, L _i (t ₀ ) Represents the initial time t of the ith unmanned aerial vehicle ₀ The position L of (2) _j (t ₀ ) Represents the initial time t of the jth unmanned plane ₀ V positions of (2) _i (t ₀ ) Represents the initial time t of the ith unmanned aerial vehicle ₀ V of (2) _j (t ₀ ) Represents the initial time t of the jth unmanned plane ₀ Is a speed of (2);

the unmanned aerial vehicle position at which the running time approaches 0 infinitely is:

θ represents the amount of time infinitely approaching 0, t ₀ →t ₀ +θ represents that the running time approaches 0, L _i (t ₀ +θ) represents the position of the ith unmanned aerial vehicle when the running time approaches 0 infinitely, L _j (t ₀ +θ) represents the position of the jth unmanned aerial vehicle when the running time approaches 0 infinitely, and the unmanned aerial vehicle flies at a constant speed when the running time approaches 0 infinitely;

the relative positions of unmanned aerial vehicles with the running time approaching 0 infinitely are as follows:

L _ij (t ₀ +θ)＝||L _ij (t ₀ )+V _ij (t ₀ )·θ||；

L _ij (t ₀ +θ) represents the relative positions of the ith and jth robots when the run time approaches 0 infinitely;

The included angle between unmanned aerial vehicle is:

γ _ij (t) represents an included angle between the ith unmanned aerial vehicle and the jth unmanned aerial vehicle at any t moment, and I represent norms.

Further, the unmanned aerial vehicle angle obstacle avoidance judging condition specifically comprises:

when gamma (t) ∈ [0, pi/2 ], if L _ij (t)·V _ij (t) > 0, determining that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle are far away in different directions;

when gamma (t) ∈ [ pi/2, pi ], if L _ij (t)·V _ij (t) is less than or equal to 0, judging that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle approach in the same direction, and triggering an obstacle avoidance strategy;

setting a safety distance:

the safe distance between the ith unmanned aerial vehicle and the jth unmanned aerial vehicle at any moment t is represented, and N represents the number of unmanned aerial vehicles.

Further, the fast reaction type obstacle avoidance strategy specifically comprises the following steps:

the fast reaction type obstacle avoidance strategy triggered according to the unmanned aerial vehicle angle obstacle avoidance judging condition is as follows:

when the safety distance is satisfiedWhen the collision risk exists, the collision risk is indicated;

in the flight process, the control rule of each unmanned aerial vehicle is the same and is in the flight state with inertia, and any two unmanned aerial vehicles jointly adjust the sum gamma (t) of the flight angles _total ：

The design of the fast reactive obstacle avoidance strategy of the ith unmanned aerial vehicle is as follows:

ρ represents an unmanned aerial vehicle heading control factor:

When the relative position of any two unmanned aerial vehicles is smaller than the safe distanceAnd changing the original flight track by the fast reactive obstacle avoidance strategy of the ith unmanned aerial vehicle until the safety distance is met.

Further, the objective function specifically includes:

wherein C is _fuel Indicating the oil consumption of all unmanned aerial vehicle tracks C _time C, representing time for completing flight path of unmanned aerial vehicle _planning Representing the voyage cost of the unmanned aerial vehicle, N represents the number of unmanned aerial vehicles, k represents the time period between the unmanned aerial vehicle reaching the task point, and l _i (. Cndot.) represents the fuel consumption of the unmanned aerial vehicle flying from the position at the time T to the position at the time t+1, T _i (. Cndot.) shows the actual time from the flight at the time t to the time t+1, d _i (. Cndot.) represents the actual flight distance from the position at time t to the position at time t+1.

Further, the constraint condition includes:

speed control constraint of unmanned aerial vehicle in three-dimensional direction:

|V _xyz，i (t) | represents the integrated speed of the ith unmanned aerial vehicle in the three-dimensional environment at time t, |V _x，i (t) | represents the absolute value of the speed of the ith unmanned aerial vehicle in the x direction at time t, |v _y，i (t) | represents the absolute value of the speed of the ith unmanned aerial vehicle in the y direction at time t, |V _z，i (t) i represents the absolute value of the speed of the ith drone in the z direction at time t,representing the speed limit of the unmanned aerial vehicle in the x-direction, < > >Representing the speed limit of the unmanned aerial vehicle in the y-direction, < >>Representing the speed limit of the drone in the z direction.

Further, the multiple separation space is obtained by a method comprising:

a1, constructing a rectangular space perpendicular to the ground as a reference point space, wherein two opposite planes perpendicular to the ground of the reference point space are marked as a first tangent plane and a second tangent plane, the position of the first tangent plane is determined by the position of the unmanned aerial vehicle, the position of a position task point of the second tangent plane is determined, the reference point space is { reference point 1, reference points 2, …, reference point 8}, and the reference points are 8 vertexes of the reference point space;

step A2, based on the reference point space, performing halving from the longer side of the reference point space to obtain 2 first separation spaces:

dis () represents the euclidean distance between two reference points;

step A3, determining threat values of two first separation spaces according to the coarse position of the mobile threat source, and judging the first separation space where the mobile threat source is located;

the number of times of separation of the space n is 2 at this time;

step A4, selecting an nth-1 separation space where the mobile threat source is located, and carrying out dichotomy on the longer side of the nth-1 separation space to obtain 2 nth separation spaces:

Step A5, determining threat values of two nth separation spaces according to the coarse position of the mobile threat source, and judging the nth separation space where the mobile threat source is located;

repeating the steps A4 to A5 until the separation cannot be continued or the maximum division times are reached;

a mobile threat source fine location is obtained.

Further, the method comprises the steps of,

the road searching vertical section specifically comprises:

based on the multiple separation spaces and the fine positions of the mobile threat sources, the current position and the task point position of the unmanned aerial vehicle are passed through to make an initial path-finding vertical tangent plane CutV _ertical ＝[A，B，C，D]Determining key nodes _point ；

Judging the initial road-finding vertical tangent plane Cut _Vertical =[A,B,C,D]Whether overlap exists with the nth separation space where the mobile threat source fine position is located;

if there is overlap, the n-th separation space where the fine position of the mobile threat source is located is separated from the initial road-finding vertical tangent Cut _Vertical =[A,B,C,D]The nearest reference point is used as a stage node, the current position and the task point position of the unmanned aerial vehicle are divided into beta stages by the stage node, the beta stage node is used as the end point of the beta stage and the starting point of the beta+1st stage, and a plurality of road searching vertical sections are constructed;

constructing a rotatable coordinate system o 'x' y 'z' based on each of the road-finding vertical Cut surfaces, wherein the road-finding vertical Cut surfaces Cut _Vertical ＝[A，B，C，o′]The rotation angle is changed in the rotatable coordinate system.

Further, the knowledge bidirectional migration specifically includes:

generating a plurality of multi-unmanned aerial vehicle distribution initial schemes through a differential evolution algorithm;

determining multiple unmanned aerial vehicle allocation initial scheme setsWherein p represents the serial number of each execution scheme in the preliminary track planning execution scheme set;

the method comprises the steps of distributing three candidate track schemes randomly from a plurality of unmanned aerial vehicles according to an initial scheme, and according to unmanned aerial vehicle sequences: UAV= [ UAV ] ₁ ，uav ₂ ，...，uav _i ]Task point sequence assignment ₁ ＝[mission ₁ ，mission ₂ ，...，mission _m ]、Mission ₂ ＝[mission ₃ ，mission ₁ ，...，mission ₆ ]And mixing ₃ ＝[mission ₈ ，mission ₃ ，...，mission ₁ ]The sequences are crossed, mutated, differenced and selected to obtain a new task point sequence after crossing

According to the new task point sequence of the assignment _new Judging an allocation identifier; the identifier represents the rationality of the post-difference allocation sequence; an identifier of 0 indicates that the allocation scheme is unreasonable, and an identifier of 1 indicates that the allocation scheme is reasonable;

if the identifier is 0, the allocation scheme is unreasonable, and at the moment, unreasonable task point sequences are covered in sequence by adopting a positive sequence ordering mutation operator to obtain a new multi-unmanned aerial vehicle allocation scheme;

representing the new multi-unmanned aerial vehicle allocation scheme as knowledge K;

the knowledge K comprises the shortest track of the initial scheme of the multi-unmanned aerial vehicle distribution Shortest planning timeAnd minimal fuel consumption->

The task of the collaborative track local re-planning of the multiple unmanned aerial vehicles is represented as delta ₃ ＝{C _fuel ，C _time ，C _planning 3 represents the number of tasks;

the initial scheme is distributed to each unmanned aerial vehicle through the knowledge K and task representation of the multi-unmanned aerial vehicle collaborative track local re-planning to carry out scheme quality assessment, and a quality assessment result is obtained;

based on the quality evaluation result, adjusting an evolution strategy of a differential evolution algorithm, and distributing tasks to a plurality of task solvers;

performing multi-task calculation through a plurality of task solvers to construct a new allocation scheme;

the multitasking computation includes:

solving the task of the multi-unmanned aerial vehicle collaborative track local re-planning of each multi-unmanned aerial vehicle allocation initial scheme through a plurality of task solvers, and simultaneously calculating the crowding distance Dis:

wherein, minDis (K) _p ，K _q ) Representing the p-th knowledge K _p And the q-th knowledge K _q Euclidean distance minimum of (c); max Dis (K) _p ，K _q ) Representing the p-th knowledge K _p And the q-th knowledge K _q Euclidean distance maximum value (x);

based on the crowding distance, the minimum knowledge of Dis is reserved in the current task solver, and the rest knowledge is uploaded to a central processor through a knowledge migration method;

the knowledge migration comprises knowledge extraction, adaptive knowledge generation and knowledge bidirectional migration;

The bidirectional migration of the knowledge comprises the steps that each reserved knowledge K is uploaded to a cloud server, the cloud server performs balance solver and knowledge migration, balances a plurality of task sources, determines the migration strength of each knowledge source, and migrates the corresponding knowledge to each task solver;

after the knowledge migration is completed, updating the knowledge transfer probability of the current task and the source task selection probability;

and solving by the current task solver to obtain an optimal planning scheme.

In another aspect of the present invention, a system for collaborative track local re-planning for an unmanned aerial vehicle with knowledge bi-directional migration is provided, the system comprising:

the discrete model construction module is used for acquiring the position of each unmanned aerial vehicle and constructing a unmanned aerial vehicle discrete model;

monitoring the coarse position and threat value of the mobile threat source in real time;

the obstacle avoidance strategy construction module is used for constructing unmanned aerial vehicle angle obstacle avoidance judging conditions based on the unmanned aerial vehicle discrete model so as to determine a quick response type obstacle avoidance strategy, and setting constraint conditions and an objective function;

the mobile threat source space reconstruction module is used for constructing a reference point space based on the plane tangent to the unmanned plane and the task point when detecting that the coarse position and the threat value of the mobile threat source meet the unmanned plane angle obstacle avoidance judging condition, acquiring a plurality of separation spaces through a space separation method based on the reference point space, and further determining the fine position of the mobile threat source;

The initial allocation scheme generation module is used for constructing a road-finding vertical section based on the fine position of the mobile threat source and the position of the current unmanned aerial vehicle, and generating a plurality of unmanned aerial vehicle allocation initial schemes meeting an objective function and a quick response obstacle avoidance strategy through a differential evolution algorithm and under constraint conditions;

and the local re-planning path generation module is used for acquiring an optimal planning scheme of the unmanned aerial vehicle by a knowledge bidirectional migration method based on the initial scheme allocated by the unmanned aerial vehicles, and taking the optimal planning scheme as a local re-planning path.

The invention has the beneficial effects that:

(1) According to the unmanned aerial vehicle angle obstacle avoidance judging method and the rapid reaction type obstacle avoidance method, through an unmanned aerial vehicle discrete model, the change of the relative position and the relative speed between unmanned aerial vehicles along with the change of time is fully considered.

(2) According to the invention, the initial allocation scheme of the multi-unmanned aerial vehicle collaborative track planning is obtained through iterative updating of the differential evolution algorithm, so that the planning efficiency of the planning system is improved.

(3) According to the distribution scheme with better convergence in the initial distribution scheme, the invention is defined as high-quality reference knowledge in the re-planning process, a reference knowledge updating method is determined, a strategy of knowledge bidirectional migration is provided, redundant calculation is avoided, and the planning efficiency of a planning system is improved.

Drawings

Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the following drawings, in which:

FIG. 1 is a flow diagram of a method for knowledge bi-directional migration of unmanned aerial vehicle collaborative track local re-planning;

fig. 2 is a schematic diagram of acquiring an optimal planning scheme of an unmanned aerial vehicle by a knowledge bidirectional migration method in the embodiment of the invention;

FIG. 3 is a flow chart of a method for knowledge migration in an embodiment of the invention;

FIG. 4 is a schematic diagram of a method for determining a fine position of a mobile threat source by obtaining multiple separation spaces through a spatial separation method in an embodiment of the invention;

FIG. 5 is a schematic diagram of the construction of a vertical section of a road-finding in accordance with the present invention.

Detailed Description

The present application is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be noted that, for convenience of description, only the portions related to the present invention are shown in the drawings.

It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

In order to more clearly describe the method for the collaborative track local re-planning of the unmanned aerial vehicle with the knowledge bi-directional migration of the present invention, each step in the embodiment of the present invention is described in detail below with reference to fig. 1.

The method for planning the unmanned aerial vehicle collaborative track local re-planning by the knowledge bi-directional migration in the first embodiment of the invention comprises the following steps S1-S5, wherein the detailed description of each step is as follows:

step S1, acquiring the position of each unmanned aerial vehicle, and constructing a unmanned aerial vehicle discrete model;

in this embodiment, the unmanned aerial vehicle discrete model includes:

the unmanned aerial vehicle position, the unmanned aerial vehicle speed, the unmanned aerial vehicle relative position and the unmanned aerial vehicle relative included angle;

wherein, unmanned aerial vehicle position is:

L _i (t)＝[x _i (t)，y _i (t)，z _i (t)] ^T ；

L _i (t) represents the position of the ith unmanned aerial vehicle at any t moment, x _i (t) represents the position of the ith unmanned aerial vehicle on the x axis in the arbitrary t moment space, y _i (t) represents the position of the ith unmanned aerial vehicle on the y-axis in the arbitrary t-moment space, z _i (T) represents the z-axis position of the ith unmanned aerial vehicle in any T moment space, and T represents transposition;

the speed of the unmanned aerial vehicle is as follows:

V _i (t)＝[v _x，i (t)，v _y，i (t)，v _z，i (t)] ^T ；

V _i (t) represents the speed, v, of the ith unmanned aerial vehicle at any time t _x，i (t) represents the speed of the ith unmanned aerial vehicle in the x direction at any time t, v _y，i (t) represents the speed of the ith unmanned aerial vehicle in the y direction at any time t, v _z，i (t) represents the speed of the ith unmanned aerial vehicle in the z direction at any time t;

the unmanned aerial vehicle position at the next time t+1 is:

L _i (t+1)＝A·L _i (t)+B·u _i (t)；

L _i (t+1) represents the position of the ith unmanned aerial vehicle at the time t+1, A represents the real-time state matrix of the unmanned aerial vehicle, B represents the input matrix, u _i (t) represents the control amount received by the ith unmanned aerial vehicle at the time t; the real-time state matrix of the unmanned aerial vehicle comprises speed and acceleration; the input matrix is an instruction of the control quantity to the unmanned aerial vehicle; the input matrix refers to an instruction of the control quantity to the unmanned aerial vehicle, and if the all-zero matrix indicates that the unmanned aerial vehicle is in a static state, the flight state of the unmanned aerial vehicle in the flight process can be changed (similar to interference) by changing the input matrix;

the relative positions of the unmanned aerial vehicle are as follows:

t ₀ indicating the initial time, L _ij (t ₀ ) Indicating that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle are at the initial time t ₀ V of the relative positions of (2) _ij (t ₀ ) Indicating that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle are at the initial time t ₀ Relative velocity, L _i (t ₀ ) Represents the initial time t of the ith unmanned aerial vehicle ₀ The position L of (2) _j (t ₀ ) Represents the initial time t of the jth unmanned plane ₀ V positions of (2) _i (t ₀ ) Represents the initial time t of the ith unmanned aerial vehicle ₀ V of (2) _j (t ₀ ) Represents the initial time t of the jth unmanned plane ₀ Is a speed of (2);

the unmanned aerial vehicle position at which the running time approaches 0 infinitely is:

θ represents the amount of time infinitely approaching 0, t ₀ →t ₀ +θ represents that the running time approaches 0, L _i (t ₀ +θ) represents the position of the ith unmanned aerial vehicle when the running time approaches 0 infinitely, L _j (t ₀ +θ) represents the position of the jth unmanned aerial vehicle when the running time approaches 0 infinitely, and the unmanned aerial vehicle flies at a constant speed when the running time approaches 0 infinitely;

the relative positions of unmanned aerial vehicles with the running time approaching 0 infinitely are as follows:

L _ij (t ₀ +θ)＝||L _ij (t ₀ )+V _ij (t ₀ )·θ||；

L _ij (t ₀ +θ) represents the relative positions of the ith and jth robots when the run time approaches 0 infinitely;

the included angle between unmanned aerial vehicle is:

γ _ij (t) represents any of the ith unmanned aerial vehicle and the jth unmanned aerial vehicle at the time tThe included angle between the unmanned aerial vehicles is that, I.I representing the norm.

Monitoring the coarse position and threat value of the mobile threat source in real time;

s2, constructing unmanned aerial vehicle angle obstacle avoidance judging conditions based on the unmanned aerial vehicle discrete model, further determining a rapid reaction type obstacle avoidance strategy, and setting constraint conditions and an objective function;

in this embodiment, the individuals in the multi-unmanned aerial vehicle collaborative track planning algorithm are represented by constructing individual four-tuple information, where the four-tuple information includes an unmanned aerial vehicle sequence uav= [ UAV ] ₁ ，uav ₂ ，...，uav _i ]I represents a serial number of the unmanned aerial vehicle; task point sequence assignment= [ assignment = [ assignment) ₁ ，mission ₂ ，...，mission _m ]M represents a serial number of a task point; unmanned aerial vehicle and task point estimated voyage cost Flight _approximate ＝[F ₁ ，F ₂ ，...，F _f ]F represents the sequence number of the approximate voyage cost; threat source Weight _threat ＝[W ₁ ，W ₂ ，...，W _w ]W represents the serial number of threat source weight, and the individual expression is: individual =<UAV，Mission，Flight _approximate ，Weight _threat >。

According to the invention, through the fact that the information of the unmanned aerial vehicle cannot be fully considered by an individual in a cooperative algorithm, four tuple information of unmanned aerial vehicle track planning is constructed, each tuple information comprises the allocation sequences of the unmanned aerial vehicle and task points, and important information such as estimated course cost is estimated.

In this embodiment, the conditions for determining the obstacle avoidance angle of the unmanned aerial vehicle are specifically:

when gamma (t) ∈ [0, pi/2 ], if L _ij (t)·V _ij (t) > 0, determining that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle are far away in different directions;

when gamma (t) ∈ [ pi/2, pi ], if L _ij (t)·V _ij (t) is less than or equal to 0, judging that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle approach in the same direction, and triggering an obstacle avoidance strategy;

setting a safety distance:

the safe distance between the ith unmanned aerial vehicle and the jth unmanned aerial vehicle at any moment t is represented, and N represents the number of unmanned aerial vehicles.

When the flight distance of the two unmanned aerial vehicles is smaller than the safe distanceThe unmanned aerial vehicle control system can trigger angle judgment, and the flight angle is changed rapidly by the unmanned aerial vehicle rapid reaction obstacle avoidance mechanism, so that collision between unmanned aerial vehicles is avoided.

In this embodiment, the fast reaction type obstacle avoidance strategy specifically includes:

the fast reaction type obstacle avoidance strategy triggered according to the unmanned aerial vehicle angle obstacle avoidance judging condition is as follows:

when the safety distance is satisfiedWhen the collision risk exists between unmanned aerial vehicles, the collision risk is indicated;

in the flight process, the control rule of each unmanned aerial vehicle is the same and is in the flight state with inertia, and any two unmanned aerial vehicles jointly adjust the sum gamma (t) of the flight angles _total ：

The design of the fast reactive obstacle avoidance strategy of the ith unmanned aerial vehicle is as follows:

ρ represents an unmanned aerial vehicle heading control factor:

when the relative position of any two unmanned aerial vehicles is smaller than the safe distanceAnd changing the original flight track by the fast reactive obstacle avoidance strategy of the ith unmanned aerial vehicle until the safety distance is met.

In this embodiment, the objective function is specifically:

wherein C is _fuel Indicating the oil consumption of all unmanned aerial vehicle tracks C _time C, representing time for completing flight path of unmanned aerial vehicle _planning Representing the voyage cost of the unmanned aerial vehicle, N represents the number of unmanned aerial vehicles, k represents the time period between the unmanned aerial vehicle reaching the task point, and l _i (. Cndot.) represents the fuel consumption of the unmanned aerial vehicle flying from the position at the time T to the position at the time t+1, T _i (. Cndot.) shows the actual time from the flight at the time t to the time t+1, d _i (. Cndot.) represents the actual flight distance from the position at time t to the position at time t+1.

In this embodiment, the constraint condition includes:

speed control constraint of unmanned aerial vehicle in three-dimensional direction:

|V _xyz，i (t) | represents the integrated speed of the ith unmanned aerial vehicle in the three-dimensional environment at time t, |V _x，i (t) | represents the absolute value of the speed of the ith unmanned aerial vehicle in the x direction at time t, |v _y，i (t) | represents the absolute value of the speed of the ith unmanned aerial vehicle in the y direction at time t, |V _z，i (t) i represents the absolute value of the speed of the ith drone in the z direction at time t,representing the speed limit of the unmanned aerial vehicle in the x-direction, < >>Representing the speed limit of the unmanned aerial vehicle in the y-direction, < >>Representing the speed limit of the drone in the z direction. The limiting speed of the unmanned aerial vehicle in all directions can be set according to the requirements.

Step S3, when the coarse position and threat value of the mobile threat source are detected to meet the unmanned plane angle obstacle avoidance judging condition, constructing a reference point space based on the unmanned plane and the task point tangent plane, acquiring multiple separation spaces through a space separation method based on the reference point space, and further determining the fine position of the mobile threat source;

in this embodiment, the multiple separation spaces, as shown in fig. 4, are obtained by a method including:

A1, constructing a rectangular space perpendicular to the ground as a reference point space, wherein two opposite planes perpendicular to the ground of the reference point space are marked as a first tangent plane and a second tangent plane, the position of the first tangent plane is determined by the position of the unmanned aerial vehicle, the position of a position task point of the second tangent plane is determined, the reference point space is { reference point 1, reference points 2, …, reference point 8}, and the reference points are 8 vertexes of the reference point space;

step A2, based on the reference point space, performing halving from the longer side of the reference point space to obtain 2 first separation spaces:

dis () represents the euclidean distance between two reference points;

step A3, determining threat values of two first separation spaces according to the coarse position of the mobile threat source, and judging the first separation space where the mobile threat source is located;

the number of times of separation of the space n is 2 at this time;

step A4, selecting an nth-1 separation space where the mobile threat source is located, and carrying out dichotomy on the longer side of the nth-1 separation space to obtain 2 nth separation spaces:

step A5, determining threat values of two nth separation spaces according to the coarse position of the mobile threat source, and judging the nth separation space where the mobile threat source is located;

repeating the steps A4 to A5 until the separation cannot be continued or the maximum division times are reached;

A mobile threat source fine location is obtained.

The space separation method adopted in the embodiment is different from the common simulation space grid method, and grid processing is not needed for the whole simulation space, so that the calculation efficiency is improved.

Step S4, constructing a road-finding vertical section based on the fine position of the mobile threat source and the position of the current unmanned aerial vehicle, and generating a plurality of unmanned aerial vehicle distribution initial schemes meeting an objective function and a quick response obstacle avoidance strategy through a differential evolution algorithm and under constraint conditions;

in the present embodiment of the present invention, in the present embodiment,

the road searching vertical section specifically comprises:

based on the multiple separation spaces and the fine positions of the mobile threat sources, the current position and the task point position of the unmanned aerial vehicle are passed through to make an initial path-finding vertical tangent plane Cut _Vertical ＝[A，B，C，D]Determining keyNode key _point ；

Judging the initial road-finding vertical tangent plane Cut _Vertical ＝[A，B，C，D]Whether overlap exists with the nth separation space where the mobile threat source fine position is located;

if there is overlap, the n-th separation space where the fine position of the mobile threat source is located is separated from the initial road-finding vertical tangent Cut _Vertical ＝[A，B，C，D]The nearest reference point is used as a stage node, the current position and the task point position of the unmanned aerial vehicle are divided into beta stages by the stage node, the beta stage node is used as the end point of the beta stage and the starting point of the beta+1st stage, and a plurality of road searching vertical sections are constructed;

Constructing a rotatable coordinate system o 'x' y 'z' based on each of the road-finding vertical Cut surfaces, wherein the road-finding vertical Cut surfaces Cut _Vertical ＝[A，B，C，o′]The rotation angle is changed in the rotatable coordinate system. A schematic of the vertical section constructed is shown in FIG. 5.

And S5, acquiring an optimal planning scheme of the unmanned aerial vehicle by a knowledge bidirectional migration method based on the initial scheme of the unmanned aerial vehicle distribution, and taking the optimal planning scheme as a local re-planning path.

In this embodiment, the knowledge bidirectional migration, as shown in fig. 2, specifically includes:

generating a plurality of multi-unmanned aerial vehicle distribution initial schemes through a differential evolution algorithm;

determining multiple unmanned aerial vehicle allocation initial scheme setsWherein p represents the serial number of each execution scheme in the preliminary track planning execution scheme set;

the method comprises the steps of distributing three candidate track schemes randomly from a plurality of unmanned aerial vehicles according to an initial scheme, and according to unmanned aerial vehicle sequences: UAV= [ UAV ] ₁ ，uav ₂ ，…uav _i ]Task point sequence assignment ₁ ＝[mision ₁ ，mission ₂ ，…，mission _m ]、Mission ₂ ＝[mission ₃ ，mission ₁ ，...，mission ₆ ]And mixing ₃ ＝[mission ₈ ，mission ₃ ，...，mission ₁ ]The sequences are crossed, mutated, differenced and selected to obtain a new task point sequence after crossing

According to the new task point sequence of the assignment _new Judging an allocation identifier; the identifier represents the rationality of the post-difference allocation sequence; an identifier of 0 indicates that the allocation scheme is unreasonable, and an identifier of 1 indicates that the allocation scheme is reasonable; the cross-talk in this segment can produce unreasonable allocation sequences such as: the task point sequence 1 and the task point sequence 2 differ by-1, and it is obvious that the task point sequence is a value that is not allowed to appear-1, so that it is reasonable to use an allocation identifier of 0 to represent the result of cross-difference.

If the identifier is 0, the allocation scheme is unreasonable, and at the moment, unreasonable task point sequences are covered in sequence by adopting a positive sequence ordering mutation operator to obtain a new multi-unmanned aerial vehicle allocation scheme;

representing the new multi-unmanned aerial vehicle allocation scheme as knowledge K;

the knowledge K comprises the shortest track of the initial scheme of the multi-unmanned aerial vehicle distributionShortest planning timeAnd minimal fuel consumption->

The task of the collaborative track local re-planning of the multiple unmanned aerial vehicles is represented as delta ₃ ＝{C _fuel ，C _time ，C _planning 3 represents the number of tasks; according to task needs, the number of tasks and the observation indexes can be increased.

The initial scheme is distributed to each unmanned aerial vehicle through the knowledge K and task representation of the multi-unmanned aerial vehicle collaborative track local re-planning to carry out scheme quality assessment, and a quality assessment result is obtained;

based on the quality evaluation result, adjusting an evolution strategy of a differential evolution algorithm, and distributing tasks to a plurality of task solvers;

performing multi-task calculation through a plurality of task solvers to construct a new allocation scheme;

the multitasking computation includes:

solving the task of the multi-unmanned aerial vehicle collaborative track local re-planning of each multi-unmanned aerial vehicle allocation initial scheme through a plurality of task solvers, and simultaneously calculating the crowding distance Dis:

Wherein, minDis (K) _p ，K _q ) Representing the p-th knowledge K _p And the q-th knowledge K _q Euclidean distance minimum of (c); maxDIs (K) _p ，K _q ) Representing the p-th knowledge K _p And the q-th knowledge K _q Euclidean distance maximum value (x);

based on the crowding distance, the minimum knowledge of Dis is reserved in the current task solver, and the rest knowledge is uploaded to a central processor through a knowledge migration method; smaller Dis means that the corresponding knowledge provides similar but redundant information for evolution in subsequent environments, resulting in poorer diversity of the knowledge base.

Carrying out knowledge migration on the reserved knowledge K by a current task solver; the knowledge migration comprises knowledge extraction, adaptive knowledge generation and knowledge bidirectional migration;

the bidirectional migration of the knowledge comprises uploading each unreserved knowledge K to a central processor, and the central processor performs balance solver and knowledge migration, balances a plurality of task sources, determines the migration strength of each knowledge source and migrates the corresponding knowledge to each task solver;

after the knowledge migration is completed, updating the knowledge transfer probability of the current task and the source task selection probability;

and solving by the current task solver to obtain an optimal planning scheme.

The invention carries out bidirectional migration of knowledge, as shown in fig. 3, specifically comprising:

the solution of the task of the multi-unmanned aerial vehicle collaborative track local re-planning obtained by calculation of each round of multi-task solver is recorded as a generation of knowledge; for example, if the number of iterations is 1000 generations, 1000 generations of knowledge will be generated;

storing each generation of knowledge and algorithm run time in a library;

initially, a new multi-unmanned aerial vehicle distribution scheme is obtained through a differential evolution algorithm; the method only relates to a reference allocation scheme, and does not relate to knowledge migration among tasks, and the knowledge migration probability of an initial task solver is 0;

after a new multi-unmanned aerial vehicle allocation scheme is obtained, extracting knowledge of the new multi-unmanned aerial vehicle allocation scheme and corresponding algorithm running time from a library to generate self-adaptive knowledge; the adaptive knowledge is determined according to the objective function, the unmanned plane range cost and the flight time, for example, the range distance of the optimal solution in the 1 st generation is 100km, the time is 50s, the range distance of the optimal solution in the second generation is 120km, and the time is 20s. And if the target function pursues that the unmanned aerial vehicle course distance is shortest, calling a first generation optimal solution to enter an iterative loop, and if the target function pursues that the unmanned aerial vehicle time is shortest, calling a second generation optimal solution to enter the iterative loop.

The bi-directional migration of knowledge herein involves knowledge in the library and knowledge of the current iteration update process into the next generation, and is therefore referred to as knowledge bi-directional migration. The selection probability of the update source task in fig. 3 is 1, which refers to both the reference library and the self knowledge migration.

Although the steps are described in the above-described sequential order in the above-described embodiments, it will be appreciated by those skilled in the art that in order to achieve the effects of the present embodiments, the steps need not be performed in such order, and may be performed simultaneously (in parallel) or in reverse order, and such simple variations are within the scope of the present invention.

The system for collaborative track local re-planning for unmanned aerial vehicle with knowledge bidirectional migration according to the second embodiment of the present invention includes:

the discrete model construction module is used for acquiring the position of each unmanned aerial vehicle and constructing a unmanned aerial vehicle discrete model;

monitoring the coarse position and threat value of the mobile threat source in real time;

the obstacle avoidance strategy construction module is used for constructing unmanned aerial vehicle angle obstacle avoidance judging conditions based on the unmanned aerial vehicle discrete model so as to determine a quick response type obstacle avoidance strategy, and setting constraint conditions and an objective function;

The mobile threat source space reconstruction module is used for constructing a reference point space based on the plane tangent to the unmanned plane and the task point when detecting that the coarse position and the threat value of the mobile threat source meet the unmanned plane angle obstacle avoidance judging condition, acquiring a plurality of separation spaces through a space separation method based on the reference point space, and further determining the fine position of the mobile threat source;

the initial allocation scheme generation module is used for constructing a road-finding vertical section based on the fine position of the mobile threat source and the position of the current unmanned aerial vehicle, and generating a plurality of unmanned aerial vehicle allocation initial schemes meeting an objective function and a quick response obstacle avoidance strategy through a differential evolution algorithm and under constraint conditions;

and the local re-planning path generation module is used for acquiring an optimal planning scheme of the unmanned aerial vehicle by a knowledge bidirectional migration method based on the initial scheme allocated by the unmanned aerial vehicles, and taking the optimal planning scheme as a local re-planning path.

It will be clear to those skilled in the art that, for convenience and brevity of description, the specific working process of the system described above and the related description may refer to the corresponding process in the foregoing method embodiment, which is not repeated here.

It should be noted that, in the system for collaborative track local re-planning of an unmanned aerial vehicle with knowledge bidirectional migration provided in the foregoing embodiment, only the division of each functional module is illustrated, in practical application, the foregoing functional allocation may be completed by different functional modules according to needs, that is, the modules or steps in the foregoing embodiment of the present invention are further decomposed or combined, for example, the modules in the foregoing embodiment may be combined into one module, or may be further split into a plurality of sub-modules, so as to complete all or part of the functions described above. The names of the modules and steps related to the embodiments of the present invention are merely for distinguishing the respective modules or steps, and are not to be construed as unduly limiting the present invention.

It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the storage device and the processing device described above and the related description may refer to the corresponding process in the foregoing method embodiment, which is not repeated herein.

Those of skill in the art will appreciate that the various illustrative modules, method steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the program(s) corresponding to the software modules, method steps, may be embodied in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. To clearly illustrate this interchangeability of electronic hardware and software, various illustrative components and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as electronic hardware or software depends upon the particular application and design constraints imposed on the solution. Those skilled in the art may implement the described functionality using different approaches for each particular application, but such implementation is not intended to be limiting.

The terms "first," "second," and the like, are used for distinguishing between similar objects and not for describing a particular sequential or chronological order.

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/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/apparatus.

Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.

Claims (6)

1. A method for collaborative track local re-planning for unmanned aerial vehicle with knowledge bi-directional migration, the method comprising:

step S1, acquiring the position of each unmanned aerial vehicle, and constructing a unmanned aerial vehicle discrete model;

Monitoring the coarse position and threat value of the mobile threat source in real time;

s2, constructing unmanned aerial vehicle angle obstacle avoidance judging conditions based on the unmanned aerial vehicle discrete model, further determining a rapid reaction type obstacle avoidance strategy, and setting constraint conditions and an objective function;

the unmanned aerial vehicle angle obstacle avoidance judging condition specifically comprises:

when gamma (t) ∈ [0, pi/2 ], if L _ij (t)·V _ij (t) > 0, determining that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle are far away in different directions;

when gamma (t) ∈ [ pi/2, pi ], if L _ij (t)·V _ij (t) is less than or equal to 0, judging that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle approach in the same direction, and triggering an obstacle avoidance strategy;

setting a safety distance:

the safety distance between the ith unmanned aerial vehicle and the jth unmanned aerial vehicle at any moment t is represented, and N represents the number of unmanned aerial vehicles;

the fast reaction type obstacle avoidance strategy specifically comprises the following steps:

the fast reaction type obstacle avoidance strategy triggered according to the unmanned aerial vehicle angle obstacle avoidance judging condition is as follows:

when the safety distance is satisfied

When the collision risk exists between unmanned aerial vehicles, the collision risk is indicated;

in the flight process, the control rule of each unmanned aerial vehicle is the same and is in the flight state with inertia, and any two unmanned aerial vehicles jointly adjust the sum gamma (t) of the flight angles _total ：

The design of the fast reactive obstacle avoidance strategy of the ith unmanned aerial vehicle is as follows:

ρ represents an unmanned aerial vehicle heading control factor:

when any two unmanned aerial vehicles are in relative positionIs smaller than the safety distanceChanging the original flight track by the rapid reactive obstacle avoidance strategy of the ith unmanned aerial vehicle until the safety distance is met;

step S3, when the coarse position and threat value of the mobile threat source are detected to meet the unmanned plane angle obstacle avoidance judging condition, constructing a reference point space based on the unmanned plane and the task point tangent plane, acquiring multiple separation spaces through a space separation method based on the reference point space, and further determining the fine position of the mobile threat source;

the multiple separation space is obtained by the method comprising the following steps:

a1, constructing a rectangular space perpendicular to the ground as a reference point space, wherein two opposite planes perpendicular to the ground of the reference point space are marked as a first tangent plane and a second tangent plane, the position of the first tangent plane is determined by the position of the unmanned aerial vehicle, the position of a position task point of the second tangent plane is determined, the reference point space is { reference point 1, reference points 2, …, reference point 8}, and the reference points are 8 vertexes of the reference point space;

step A2, based on the reference point space, performing halving from the longer side of the reference point space to obtain 2 first separation spaces:

dis () represents the euclidean distance between two reference points;

step A3, determining threat values of two first separation spaces according to the coarse position of the mobile threat source, and judging the first separation space where the mobile threat source is located;

the number of times of separation of the space n is 2 at this time;

step A4, selecting an nth-1 separation space where the mobile threat source is located, and carrying out dichotomy on the longer side of the nth-1 separation space to obtain 2 nth separation spaces:

step A5, determining threat values of two nth separation spaces according to the coarse position of the mobile threat source, and judging the nth separation space where the mobile threat source is located;

repeating the steps A4 to A5 until the separation cannot be continued or the maximum division times are reached;

obtaining a mobile threat source fine position;

step S4, constructing a road-finding vertical section based on the fine position of the mobile threat source and the position of the current unmanned aerial vehicle, and generating a plurality of unmanned aerial vehicle distribution initial schemes meeting an objective function and a quick response obstacle avoidance strategy through a differential evolution algorithm and under constraint conditions;

the road searching vertical section specifically comprises:

based on the multiple separation spaces and the fine positions of the mobile threat sources, the current position and the task point position of the unmanned aerial vehicle are passed through to make an initial path-finding vertical tangent plane Cut _Vertical ＝[A，B，C，D]Determining key nodes _point ；

Judging the initial road-finding vertical tangent plane Cut _Vertical ＝[A，B，C，D]Whether overlap exists with the nth separation space where the mobile threat source fine position is located;

if there is overlap, the n-th separation space where the fine position of the mobile threat source is located is separated from the initial road-finding vertical tangent Cut _Vertical ＝[A，B，C，D]The nearest reference point is used as a stage node, the current position and the task point position of the unmanned aerial vehicle are divided into beta stages by the stage node, the beta stage node is used as the end point of the beta stage and the starting point of the beta+1st stage, and a plurality of road searching vertical sections are constructed;

constructing a rotatable coordinate system O 'x' y 'z' based on each road-finding vertical section, wherein the road-finding vertical section Cut _Vertical ＝[A，B，C，o′]Changing a rotation angle in the rotatable coordinate system;

and S5, acquiring an optimal planning scheme of the unmanned aerial vehicle by a knowledge bidirectional migration method based on the initial scheme of the unmanned aerial vehicle distribution, and taking the optimal planning scheme as a local re-planning path.

2. The method for collaborative track local re-planning for a knowledge bi-directional migration of a unmanned aerial vehicle of claim 1, wherein the unmanned aerial vehicle discrete model comprises:

the unmanned aerial vehicle position, the unmanned aerial vehicle speed, the unmanned aerial vehicle relative position and the unmanned aerial vehicle relative included angle;

Wherein, unmanned aerial vehicle position is:

L _i (t)＝[x _i (t)，y _i (t)，z _i (t)] ^T ；

L _i (t) represents the position of the ith unmanned aerial vehicle at any t moment, x _i (t) represents the position of the ith unmanned aerial vehicle on the x axis in the arbitrary t moment space, y _i (t) represents the position of the ith unmanned aerial vehicle on the y-axis in the arbitrary t-moment space, Z _i (T) represents the z-axis position of the ith unmanned aerial vehicle in any T moment space, and T represents transposition;

the speed of the unmanned aerial vehicle is as follows:

V _i (t)＝[v _x，i (t)，v _y，i (t)，v _z，i (t)] ^T ；

V _i (t) represents the speed, v, of the ith unmanned aerial vehicle at any time t _x，i (t) represents the speed of the ith unmanned aerial vehicle in the x direction at any time t, v _y，i (t) represents the speed of the ith unmanned aerial vehicle in the y direction at any time t, v _z，i (t) represents the speed of the ith unmanned aerial vehicle in the z direction at any time t;

the unmanned aerial vehicle position at the next time t+1 is:

L _i (t+1)＝A·L _i (t)+B·u _i (t)；

L _i (t+1) represents the position of the ith unmanned aerial vehicle at the time t+1, A represents the real-time state matrix of the unmanned aerial vehicle, B represents the input matrix, u _i (t) represents the control amount received by the ith unmanned aerial vehicle at the time t; the real-time state matrix of the unmanned aerial vehicle comprises speed and acceleration; the input matrix is an instruction of the control quantity to the unmanned aerial vehicle;

the relative positions of the unmanned aerial vehicle are as follows:

t ₀ indicating the initial time, L _ij (t ₀ ) Indicating that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle are at the initial time t ₀ V of the relative positions of (2) _ij (t ₀ ) Indicating that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle are at the initial time t ₀ Relative velocity, L _i (t ₀ ) Represents the initial time t of the ith unmanned aerial vehicle ₀ The position L of (2) _j (t ₀ ) Represents the initial time t of the jth unmanned plane ₀ V positions of (2) _i (t ₀ ) Represents the initial time t of the ith unmanned aerial vehicle ₀ V of (2) _j (t ₀ ) Represents the initial time t of the jth unmanned plane ₀ Is a speed of (2);

the unmanned aerial vehicle position at which the running time approaches 0 infinitely is:

θ represents the amount of time infinitely approaching 0, t ₀ →t ₀ +θ represents that the running time approaches 0, L _i (t ₀ +θ) represents the position of the ith unmanned aerial vehicle when the running time approaches 0 infinitely, L _j (t ₀ +θ) represents the position of the jth unmanned aerial vehicle when the running time approaches 0 infinitely, and the unmanned aerial vehicle flies at a constant speed when the running time approaches 0 infinitely;

the relative positions of unmanned aerial vehicles with the running time approaching 0 infinitely are as follows:

L _ij (t ₀ +θ)＝||L _ij (t ₀ )+V _ij (t ₀ )·θ||；

L _ij (t ₀ +θ) represents the relative positions of the ith and jth robots when the run time approaches 0 infinitely;

the included angle between unmanned aerial vehicle is:

γ _ij (t) represents an included angle between the ith unmanned aerial vehicle and the jth unmanned aerial vehicle at any t moment, and I represent norms.

3. The method for collaborative flight path local re-planning for a knowledge bi-directional migration unmanned aerial vehicle according to claim 1, wherein the objective function is specifically:

Wherein C is _fuel Indicating the oil consumption of all unmanned aerial vehicle tracks C _time C, representing time for completing flight path of unmanned aerial vehicle _planning Representing the voyage cost of the unmanned aerial vehicle, N represents the number of unmanned aerial vehicles, k represents the time period between the unmanned aerial vehicle reaching the task point, and l _i (. Cndot.) represents the fuel consumption of the unmanned aerial vehicle flying from the position at the time T to the position at the time t+1, T _i (. Cndot.) shows the actual time from the flight at the time t to the time t+1, d _i (. Cndot.) represents the actual flight distance from the position at time t to the position at time t+1.

4. The method for collaborative flight path local re-planning for a knowledge bi-directional migration according to claim 1, wherein the constraints include:

speed control constraint of unmanned aerial vehicle in two-dimensional direction:

|V _xyz，i (t) | represents the integrated speed of the ith unmanned aerial vehicle in the three-dimensional environment at time t, |V _x，i (t) | represents the absolute value of the speed of the ith unmanned aerial vehicle in the x direction at time t, |v _y，i (t) | represents the absolute value of the speed of the ith unmanned aerial vehicle in the y direction at time t, |V _z，i (t) i represents the absolute value of the speed of the ith drone in the z direction at time t,representing the speed limit of the unmanned aerial vehicle in the x-direction, < >>Representing the speed limit of the unmanned aerial vehicle in the y-direction, < >>Representing the speed limit of the drone in the z direction.

5. The method for collaborative flight path local re-planning for knowledge bi-directional migration of unmanned aerial vehicles according to claim 1, wherein the knowledge bi-directional migration is specifically:

generating a plurality of multi-unmanned aerial vehicle distribution initial schemes through a differential evolution algorithm;

determining multiple unmanned aerial vehicle allocation initial scheme sets

Wherein p represents the serial number of each execution scheme in the preliminary track planning execution scheme set;

the method comprises the steps of distributing three candidate track schemes randomly from a plurality of unmanned aerial vehicles according to an initial scheme, and according to unmanned aerial vehicle sequences: UAV= [ UAV ] ₁ ，uav ₂ ，...，uav _i ]Task point sequence

Mission ₁ ＝[mission ₁ ，mission ₂ ，...，mission _m ]、Mission ₂ ＝[mission ₃ ，mission ₁ ，...，mission ₆ ]And mixing ₃ ＝[mission ₈ ，mission ₃ ，...，mission ₁ ]

The sequences are crossed, mutated, differenced and selected to obtain a new task point sequence after crossing

According to the new task point sequence of the assignment _new Judging an allocation identifier; the identifier represents the rationality of the post-difference allocation sequence; an identifier of 0 indicates that the allocation scheme is unreasonable, and an identifier of 1 indicates that the allocation scheme is reasonable;

if the identifier is 0, the allocation scheme is unreasonable, and at the moment, unreasonable task point sequences are covered in sequence by adopting a positive sequence ordering mutation operator to obtain a new multi-unmanned aerial vehicle allocation scheme;

representing the new multi-unmanned aerial vehicle allocation scheme as knowledge K;

the knowledge K comprises the shortest track of the initial scheme of the multi-unmanned aerial vehicle distribution Shortest planning time->And minimal fuel consumption->

The task of the multi-unmanned aerial vehicle collaborative track local re-planning is expressed as

δ ₃ ＝{C _fuel ，C _time ，C _planning 3 represents a task number;

the initial scheme is distributed to each unmanned aerial vehicle through the knowledge K and task representation of the multi-unmanned aerial vehicle collaborative track local re-planning to carry out scheme quality assessment, and a quality assessment result is obtained;

based on the quality evaluation result, adjusting an evolution strategy of a differential evolution algorithm, and distributing tasks to a plurality of task solvers;

performing multi-task calculation through a plurality of task solvers to construct a new allocation scheme;

the multitasking computation includes:

solving the task of the multi-unmanned aerial vehicle collaborative track local re-planning of each multi-unmanned aerial vehicle allocation initial scheme through a plurality of task solvers, and simultaneously calculating the crowding distance Dis:

wherein, minDis (K) _p ，K _q ) Representing the p-th knowledge K _p And the q-th knowledge K _q Euclidean distance minimum of (c); maxDIs (K) _p ，K _q ) Representing the p-th knowledge K _p And the q-th knowledge K _q Euclidean distance maximum value (x);

based on the crowding distance, the minimum knowledge of Dis is reserved in the current task solver, and the rest knowledge is uploaded to a central processor through a knowledge migration method;

the knowledge migration comprises knowledge extraction, adaptive knowledge generation and knowledge bidirectional migration;

The bidirectional migration of the knowledge comprises uploading each unreserved knowledge K to a central processor, and the central processor performs balance solver and knowledge migration, balances a plurality of task sources, determines the migration strength of each knowledge source and migrates the corresponding knowledge to each task solver;

after the knowledge migration is completed, updating the knowledge transfer probability of the current task and the source task selection probability;

and solving by the current task solver to obtain an optimal planning scheme.

6. A system for collaborative track local re-planning for unmanned aerial vehicle with knowledge bi-directional migration, the system comprising:

the discrete model construction module is used for acquiring the position of each unmanned aerial vehicle and constructing a unmanned aerial vehicle discrete model;

monitoring the coarse position and threat value of the mobile threat source in real time;

the obstacle avoidance strategy construction module is used for constructing unmanned aerial vehicle angle obstacle avoidance judging conditions based on the unmanned aerial vehicle discrete model so as to determine a quick response type obstacle avoidance strategy, and setting constraint conditions and an objective function;

the unmanned aerial vehicle angle obstacle avoidance judging condition specifically comprises:

when gamma (t) ∈ [0, pi/2 ], if L _ij (t)·V _ij (t) > 0, determining that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle are far away in different directions;

When gamma (t) ∈ [ pi/2, pi ], if L _ij (t)·V _ij (t) is less than or equal to 0, judging that the ith unmanned aerial vehicle and the jth unmanned aerial vehicle approach in the same direction, and triggering an obstacle avoidance strategy;

setting a safety distance:

the safety distance between the ith unmanned aerial vehicle and the jth unmanned aerial vehicle at any moment t is represented, and N represents the number of unmanned aerial vehicles;

the fast reaction type obstacle avoidance strategy specifically comprises the following steps:

the fast reaction type obstacle avoidance strategy triggered according to the unmanned aerial vehicle angle obstacle avoidance judging condition is as follows:

when the safety distance is satisfiedWhen the collision risk exists between unmanned aerial vehicles, the collision risk is indicated;

in the flying process, the control rule of each unmanned plane is the same and all are in a flying state with inertiaIn a state, any two unmanned aerial vehicles jointly adjust the sum gamma (t) of flight angles _total ：

The design of the fast reactive obstacle avoidance strategy of the ith unmanned aerial vehicle is as follows:

ρ represents an unmanned aerial vehicle heading control factor:

when the relative position of any two unmanned aerial vehicles is smaller than the safe distanceChanging the original flight track by the rapid reactive obstacle avoidance strategy of the ith unmanned aerial vehicle until the safety distance is met;

the mobile threat source space reconstruction module is used for constructing a reference point space based on the plane tangent to the unmanned plane and the task point when detecting that the coarse position and the threat value of the mobile threat source meet the unmanned plane angle obstacle avoidance judging condition, acquiring a plurality of separation spaces through a space separation method based on the reference point space, and further determining the fine position of the mobile threat source;

The multiple separation space is obtained by the method comprising the following steps:

a1, constructing a rectangular space perpendicular to the ground as a reference point space, wherein two opposite planes perpendicular to the ground of the reference point space are marked as a first tangent plane and a second tangent plane, the position of the first tangent plane is determined by the position of the unmanned aerial vehicle, the position of a position task point of the second tangent plane is determined, the reference point space is { reference point 1, reference points 2, …, reference point 8}, and the reference points are 8 vertexes of the reference point space;

step A2, based on the reference point space, performing halving from the longer side of the reference point space to obtain 2 first separation spaces:

dis () represents the euclidean distance between two reference points;

step A3, determining threat values of two first separation spaces according to the coarse position of the mobile threat source, and judging the first separation space where the mobile threat source is located;

the number of times of separation of the space n is 2 at this time;

step A4, selecting an nth-1 separation space where the mobile threat source is located, and carrying out dichotomy on the longer side of the nth-1 separation space to obtain 2 nth separation spaces:

step A5, determining threat values of two nth separation spaces according to the coarse position of the mobile threat source, and judging the nth separation space where the mobile threat source is located;

Repeating the steps A4 to A5 until the separation cannot be continued or the maximum division times are reached;

obtaining a mobile threat source fine position;

the initial allocation scheme generation module is used for constructing a road-finding vertical section based on the fine position of the mobile threat source and the position of the current unmanned aerial vehicle, and generating a plurality of unmanned aerial vehicle allocation initial schemes meeting an objective function and a quick response obstacle avoidance strategy through a differential evolution algorithm and under constraint conditions;

the road searching vertical section specifically comprises:

based on the multiple separation spaces and the fine positions of the mobile threat sources, the current position and the task point position of the unmanned aerial vehicle are passed through to make an initial path-finding vertical tangent plane Cut _Vertical ＝[A，B，C，D]Determining key nodes _point ；

Judging the initial road-finding vertical tangent plane Cut _Vertical ＝[A，B，C，D]Whether overlap exists with the nth separation space where the mobile threat source fine position is located;

if there is overlap, the n-th separation space where the fine position of the mobile threat source is located is separated from the initial road-finding vertical tangent Cut _Vertical ＝[A，B，C，D]The nearest reference point is used as a stage node, the current position and the task point position of the unmanned aerial vehicle are divided into beta stages by the stage node, the beta stage node is used as the end point of the beta stage and the starting point of the beta+1st stage, and a plurality of road searching vertical sections are constructed;

Constructing a rotatable coordinate system O 'x' y 'z' based on each road-finding vertical section, wherein the road-finding vertical section Cut _Vertical ＝[A，B，C，o′]Changing a rotation angle in the rotatable coordinate system;

and the local re-planning path generation module is used for acquiring an optimal planning scheme of the unmanned aerial vehicle by a knowledge bidirectional migration method based on the initial scheme allocated by the unmanned aerial vehicles, and taking the optimal planning scheme as a local re-planning path.