CN105469644A - Flight conflict resolution method and flight conflict resolution device - Google Patents

Abstract

The invention provides a flight conflict resolution method and a flight conflict resolution device. The method comprises steps: according to a four-dimensional trajectory of a to-be-optimized aerocraft, a flight conflict situation estimation model is acquired, wherein each individual in the model comprises the departure delay time for all to-be-optimized aerocrafts, and different individuals are different; according to a target function, a target function value corresponding to the above model is acquired, wherein the target function is built according to the flight conflict situations between all to-be-optimized aerocrafts; all individuals in the above model are divided into M groups, and as for each group, a memetic algorithm is adopted for preset times of variation; the M groups are arranged in a cycle, the optimized individual in each group is copied to the next group sequentially to replace the poorest individual in the next group, an updated model and the corresponding target function value are acquired, and the model corresponding to the larger value in the two target function values is kept. The computing efficiency is high, the flight conflict resolution efficiency is improved, and the average departure delay time is low.

Description

Flight conflict resolution method and equipment

Technical Field

The invention relates to an airspace traffic management technology, in particular to a flight conflict resolution method and equipment.

Background

In recent years, the air transportation industry has developed rapidly, and the civil aviation market is expected to continue to keep growing rapidly in the coming years. However, as the flight flow increases, the density of the aircrafts in the airspace correspondingly increases, the safety interval between the aircrafts is difficult to guarantee, the possibility of collision increases, and further the flight safety is seriously threatened. As one of the key technologies for ensuring flight safety, research on a flight conflict resolution method is necessary and urgent.

The traditional flight conflict resolution method mainly focuses on a local airspace, and a strategic level global resolution method is lacked. In addition, the single-point operation mode of the traditional flight conflict resolution method greatly limits the calculation efficiency.

Disclosure of Invention

The invention provides a method and equipment for flight conflict resolution, which plan airspace traffic according to a global strategy, reduce the calculated amount of flight conflict resolution and improve the flight conflict resolution efficiency.

In a first aspect, the present invention provides a flight conflict resolution method, including:

acquiring a flight conflict situation estimation model according to a four-dimensional track of an aircraft to be optimized, wherein the flight conflict situation estimation model comprises a plurality of individuals, each individual in the plurality of individuals comprises the take-off delay time of all the aircraft to be optimized, and different individuals are different;

obtaining an objective function value corresponding to the flight conflict situation estimation model according to an objective function, wherein the objective function is established according to flight conflict situations among all aircrafts to be optimized;

dividing all individuals in the flight conflict situation estimation model into M groups, and performing variation on each group for preset times by adopting a cultural genetic algorithm to optimize the takeoff delay time of the aircraft to be optimized in the group, wherein M is an integer greater than or equal to 2;

arranging the M groups to form a loop, taking any one group in the loop as an initial group, sequentially copying the optimal individual in each group to a next group, replacing the worst individual in the next group until the previous group in the initial group is finished, and obtaining an updated flight conflict situation estimation model;

and obtaining an objective function value corresponding to the updated flight conflict situation estimation model according to the objective function, and reserving the flight conflict situation estimation model corresponding to the larger value of the two objective function values.

In a first possible implementation manner of the first aspect, the obtaining a flight conflict situation estimation model according to a four-dimensional trajectory of an aircraft to be optimized includes:

obtaining flight Conflict Situations (CS) among all aircrafts to be optimized according to the following formula:

$\mathrm{CS}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{j>i}{\overset{n}{\mathrm{\Σ}}}\left[\frac{\left|\min (0,({\mathrm{dist}}_{\mathrm{ij}}\left(t\right)-{\mathrm{\ϵ}}_{\mathrm{ij}}))\right|}{\left|{\mathrm{\ϵ}}_{\mathrm{ij}}\right|}\right]$

wherein,_ijrepresenting the aircraft F to be optimized_iAnd F_jA safety interval therebetween; dist_ij(t) denotes the aircraft F to be optimized_iAnd F_jThe minimum distance between the aircraft and the aircraft to be optimized, ∑ is a summation symbol, min () represents the smaller of the two values in brackets, and n represents the total number of the aircraft to be optimized.

According to the first possible implementation manner of the first aspect, in a second possible implementation manner of the first aspect, the obtaining, according to an objective function, an objective function value corresponding to the flight conflict situation estimation model includes:

obtaining an objective function value corresponding to the flight conflict situation estimation model according to the following formula:

$F=\frac{1-\frac{1}{n}{\mathrm{\Σ}}_{i=1}^n\left(\frac{{\mathrm{\δ}}_i}{{\mathrm{\δ}}_{\max }}\right)}{1+\mathrm{CS}}$

wherein F represents individual fitness;_irepresenting the aircraft F to be optimized_iThe takeoff delay time of;_maxrepresenting a preset takeoff delay time.

According to the first aspect and any one of the first to the second possible implementation manners of the first aspect, in a third possible implementation manner of the first aspect, the local search in the cultural genetic algorithm is specifically:

obtaining local search frequency according to a Gaussian distribution model expressed asWherein gamma represents the local search frequency, G represents the preset times, mu represents the mean value of a Gaussian distribution model, sigma represents the standard deviation of the Gaussian distribution model, and η represents the number of individuals in each group;

obtaining individuals in the group for local search according to the local search frequency;

and carrying out local search on each individual in the individuals carrying out local search by adopting a preset local search strategy to obtain the individual with better self-fitness.

According to the first aspect and any one of the first to third possible implementation manners of the first aspect, in a fourth possible implementation manner of the first aspect, the individual further includes a number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual, and the global search in the cultural genetic algorithm is specifically:

obtaining self fitness of the aircraft to be optimized according to the number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual;

and carrying out global search by adopting a preset global search strategy according to the self fitness of the aircraft to be optimized to obtain an individual with better self fitness.

In a second aspect, the present invention provides a flight conflict resolution apparatus comprising:

the system comprises an acquisition module, a judgment module and a processing module, wherein the acquisition module is used for acquiring a flight conflict situation estimation model according to a four-dimensional track of an aircraft to be optimized, the flight conflict situation estimation model comprises a plurality of individuals, each individual in the plurality of individuals comprises the takeoff delay time of all the aircraft to be optimized, and different individuals are different;

the operation module is used for obtaining an objective function value corresponding to the flight conflict situation estimation model obtained by the obtaining module according to an objective function, wherein the objective function is established according to flight conflict situations among all aircrafts to be optimized;

the variation module is used for dividing all individuals in the flight conflict situation estimation model acquired by the acquisition module into M groups, and for each group, performing variation for preset times by adopting a cultural genetic algorithm to optimize the takeoff delay time of the aircraft to be optimized in the group, wherein M is an integer greater than or equal to 2;

the updating module is used for arranging the M groups obtained by the variation module after variation to form a cycle, taking any one group in the cycle as an initial group, sequentially copying the optimal individual in each group to a next group, replacing the worst individual in the next group until the previous group in the initial group is finished, and obtaining an updated flight conflict situation estimation model;

the operation module is further configured to obtain, according to the objective function, an objective function value corresponding to the updated flight conflict situation estimation model obtained by the update module, and retain the flight conflict situation estimation model corresponding to the larger of the two objective function values.

In a first possible implementation manner of the second aspect, the obtaining module is specifically configured to:

obtaining flight Conflict Situations (CS) among all aircrafts to be optimized according to the following formula:

$\mathrm{CS}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{j>i}{\overset{n}{\mathrm{\Σ}}}\left[\frac{\left|\min (0,({\mathrm{dist}}_{\mathrm{ij}}\left(t\right)-{\mathrm{\ϵ}}_{\mathrm{ij}}))\right|}{\left|{\mathrm{\ϵ}}_{\mathrm{ij}}\right|}\right]$

wherein,_ijrepresenting the aircraft F to be optimized_iAnd F_jA safety interval therebetween; dist_ij(t) denotes the aircraft F to be optimized_iAnd F_jThe minimum distance between the aircraft and the aircraft to be optimized, ∑ is a summation symbol, min () represents the smaller of the two values in brackets, and n represents the total number of the aircraft to be optimized.

According to the first possible implementation manner of the second aspect, in a second possible implementation manner of the second aspect, the operation module is specifically configured to:

obtaining an objective function value corresponding to the flight conflict situation estimation model according to the following formula:

$F=\frac{1-\frac{1}{n}{\mathrm{\Σ}}_{i=1}^n\left(\frac{{\mathrm{\δ}}_i}{{\mathrm{\δ}}_{\max }}\right)}{1+\mathrm{CS}}$

wherein F represents individual fitness;_irepresenting the aircraft F to be optimized_iThe takeoff delay time of;_maxrepresenting a preset takeoff delay time.

In a third possible implementation manner of the second aspect, the mutation module is specifically configured to, when the local search in the cultural genetic algorithm is adopted:

obtaining local search frequency according to a Gaussian distribution model expressed asWherein gamma represents the local search frequency, G represents the preset times, mu represents the mean value of a Gaussian distribution model, sigma represents the standard deviation of the Gaussian distribution model, and η represents the number of individuals in each group;

obtaining individuals in the group for local search according to the local search frequency;

and carrying out local search on each individual in the individuals carrying out local search by adopting a preset local search strategy to obtain the individual with better self-fitness.

According to the second aspect and any one of the first to third possible implementation manners of the second aspect, in a fourth possible implementation manner of the second aspect, the individual further includes a number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual, and the variation module is specifically configured to, when the global search in the cultural genetic algorithm is adopted:

obtaining self fitness of the aircraft to be optimized according to the number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual;

and carrying out global search by adopting a preset global search strategy according to the self fitness of the aircraft to be optimized to obtain an individual with better self fitness.

The flight conflict situation estimation models are grouped, a multi-island genetic algorithm is adopted for parallel processing, the operation efficiency is high, and the lower average flight delay time is achieved while the flight conflict resolution efficiency is improved; through a cultural genetic algorithm, the requirement of flight conflict resolution under global optimization is met, and the global strategic planning of airspace traffic is realized.

Drawings

FIG. 1 is a schematic flow chart of a first embodiment of a flight conflict resolution method of the present invention;

FIG. 2 is a diagram illustrating an example of individual codes in a first embodiment of a flight conflict resolution method according to the present invention;

FIG. 3 is a diagram illustrating an exemplary migration exchange among groups according to an embodiment of the flight conflict resolution method of the present invention;

FIG. 4 is a schematic flow chart of a second embodiment of a flight conflict resolution method of the present invention;

FIG. 5 is a schematic diagram of a cross-over method in a second embodiment of the flight conflict resolution method of the present invention;

FIG. 6 is a schematic diagram illustrating a variation method in a second embodiment of the flight conflict resolution method according to the present invention;

fig. 7 is a schematic structural diagram of a third embodiment of the flight conflict resolution device according to the present invention.

Detailed Description

The flight conflict resolution problem is a complex and large-scale optimization problem of multivariable (including continuous and discrete variables), multi-target, multi-constraint, nonlinear, multi-extreme, objective function and constraint condition non-analytic function. The traditional optimization algorithms, such as the gradient-based optimization algorithm and the Powell method, have the following disadvantages:

(1) the traditional optimization algorithm cannot be directly used for processing the optimization problem of continuous/discrete hybrid design variables;

(2) the traditional optimization algorithm is sensitive to initial values and is easy to fall into local optimal points;

(3) the single-point operation mode of the traditional optimization algorithm greatly limits the calculation efficiency;

(4) conventional optimization algorithms often require that the objective function and constraints are continuously differentiable analytical functions.

The multi-island genetic algorithm has the advantages that a large-scale problem can be decomposed into small-scale problems, and then the small-scale problems are solved one by one, so that the multi-island genetic algorithm is used as a main method for improving the calculation efficiency.

First, the concept of the present invention will be explained as follows.

1. Flight conflict and flight collision: when the distance between two aircraft is less than the collision threshold (collision threshold), the two aircraft are considered to be at risk of collision (collision risk).

2. Four-dimensional track (4D-track, 4DT for short): four-dimensional track is an air traffic control concept proposed by the Federal aviation administration (Federal aviation administration, abbreviated as FAA) in NextGen 2007. It describes the four-dimensional spatio-temporal information of an aircraft from takeoff to landing, including the spatial path and the time of flight.

3. The intelligent optimization algorithm comprises the following steps: an intelligent optimization algorithm (intelligentized optimization algorithm), also known as intelligent computing (intelligentized computing), is an optimization algorithm developed by simulating or revealing certain natural phenomena or processes, the idea and content of the algorithm relate to subjects such as mathematics, physics, biology, computer science and the like, the algorithm does not depend on gradient information, has global, parallel and efficient optimization performance, high robustness and universality, and provides a new idea and means for solving large-scale nonlinear problems.

4. Culture gene algorithm: the cultural genetic algorithm (Memetic Algorithm, MA for short) simulates the variation process supported by a large amount of professional knowledge by using local heuristic search, and is a combination of global search based on population and local heuristic search based on individuals. The culture genetic algorithm provides a framework which is a concept, different culture genetic algorithms can be formed by adopting different search strategies under the framework, for example, a global search strategy can adopt a genetic algorithm, an evolution strategy, an evolution plan and the like, and a local search strategy can adopt mountain climbing search, simulated annealing, a greedy algorithm, tabu search, guided local search and the like.

Example one

Fig. 1 is a schematic flow chart of a first embodiment of a flight conflict resolution method according to the present invention. The present embodiment provides a flight conflict resolution method, which may be executed by a flight conflict resolution device, and the device may be integrated in any terminal device, such as a computer, a notebook computer, or a Personal Digital Assistant (PDA). As shown in fig. 1, an embodiment of the present invention provides a flight conflict resolution method, including:

s101, acquiring a flight conflict situation estimation model according to the four-dimensional track of the aircraft to be optimized, wherein the flight conflict situation estimation model comprises a plurality of individuals, each individual in the plurality of individuals comprises the takeoff delay time of all the aircraft to be optimized, and different individuals are different.

Because the track information sharing is realized under the four-dimensional track operation condition, the track of the aircraft has the advantage of high precision on the basis of mastering the intention information of the aircraft, particularly, the aircraft can realize the straight-line flight between any two points in the short-term track within twenty minutes, and therefore, a deterministic method is adopted for the short-term flight conflict situation estimation, namely, the position of the aircraft is supposed to be extrapolated based on the current velocity vector, and the track extrapolation error is not considered, so that the method can realize the optimal estimation in a short distance.

The invention realizes the flight conflict resolution by the takeoff delay time of the aircraft to be optimized. And coding all the aircrafts to be optimized, wherein the whole piece of coding comprises the takeoff delay time of each aircraft and the conflict situation existing among all the aircrafts to be optimized, and each piece of coding is taken as an individual, as shown in fig. 2, wherein A1, A2, A3 and … … respectively represent parameters of each aircraft, such as the takeoff delay time and the like. For different individuals, the takeoff delay time of at least one aircraft to be optimized in the takeoff delay times of the aircraft to be optimized is different from that of other individuals.

And S102, obtaining an objective function value corresponding to the flight conflict situation estimation model according to the objective function.

The objective function is established according to flight conflict situations among all aircrafts to be optimized. The objective function may be preset or may be generated as needed. In view of the cost savings associated with flights (aircraft), the objective function is set to take into account at least two factors: flight conflicts between aircraft are eliminated as much as possible while reducing takeoff delay time. For different flight conflict situation estimation models, the larger the function value of the objective function is, the better the result is, namely, the fewer conflicts among aircrafts are.

S103, dividing all individuals in the flight conflict situation estimation model into M groups, and performing variation on each group for preset times by adopting a cultural genetic algorithm to optimize the takeoff delay time of the aircraft to be optimized in the group, wherein M is an integer greater than or equal to 2.

Specifically, the flight conflict situation estimation model includes a plurality of individuals, and in order to improve the calculation efficiency, all the individuals in the model are divided into M groups for processing, where the division manner is not limited, and for example, an equal division manner may be adopted.

Then, a cultural genetic algorithm is used for mutation of each group, and the operation is circulated for a preset number of times. In addition, the treatments between different groups are independent. The operation time and the increase trend of the preset times are the same, and the preset times can be set according to actual requirements, for example, the preset times is 1.

The cultural genetic algorithm in this step is implemented according to the takeoff delay time of the aircraft to be optimized.

And S104, arranging the M groups to form a loop, taking any group in the loop as an initial group, sequentially copying the optimal individual in each group to a next group, replacing the worst individual in the next group until the previous group in the initial group is finished, and obtaining the updated flight conflict situation estimation model.

Fig. 3 shows an example of inter-group migration exchange, where M ═ 5 in this example, and is set 1, set 2, set 3, set 4, and set 5, respectively, and these 5 groups are arranged in a ring. For example, starting with group 1, the best individual in group 1 is copied to group 2, replacing the worst individual in group 2, and so on, until the best individual in group 5 is copied to group 1, ending, thus improving the overall quality of the individuals of each group. This step is used to achieve the sharing of the best individuals between the different groups.

And S105, calculating an objective function value corresponding to the updated flight conflict situation estimation model according to the objective function, and reserving the flight conflict situation estimation model corresponding to the larger value of the two objective function values.

The invention aims to reduce flight conflicts, so that the objective function values corresponding to the front and rear flight conflict situation estimation models in the embodiment are compared, and the better one of the objective function values is reserved.

The flight conflict situation estimation models are grouped, a multi-island genetic algorithm is adopted for parallel processing, the operation efficiency is high, and the lower average flight delay time is achieved while the flight conflict resolution efficiency is improved; through a cultural genetic algorithm, the requirement of flight conflict resolution under global optimization is met, and the global strategic planning of airspace traffic is realized.

Example two

Fig. 4 is a flowchart illustrating a second embodiment of a flight conflict resolution method according to the present invention. As shown in fig. 4, the method includes:

s401, establishing a flight conflict situation estimation model.

In this embodiment, first allThe takeoff delay time of the aircraft to be optimized is set to the set Wherein,_ithe takeoff delay time of the ith aircraft to be optimized is represented, and n represents the total number of all the aircraft to be optimized. For each variable_iThe constraints of the value range must be satisfied:_i∈[0,_max]，_maxis the maximum takeoff delay time that is objectively allowed. The method takes the takeoff delay time of the aircraft to be optimized as an adjustment parameter, and optimizes the flight conflict situation estimation model by combining the takeoff delay time of all the aircraft to be optimized, namely a flight conflict resolution scheme. Here, the coding of the aircraft to be optimized is also shown in fig. 2, and is not described here in detail.

If two aircrafts F to be optimized on the same altitude level are considered₁And F₂As shown in FIG. 5, wherein F₁The takeoff airport of (a) is located on the navigation section (W1, W3), and the current position coordinate is (x)₁,y₁) Angle of direction theta₁The velocity is v; f₂The takeoff airport of (1) is B, is positioned on a navigation road section (W2, W3), and the current position coordinate is (x)₂,y₂) Angle of direction theta₂The velocity magnitude is v.

At time t, F₁And F₂The coordinates of (a) are:

(x₁′,y₁′)＝(x₁+vtcosθ₁,y₁+vtsinθ₁)(1)

(x₂′,y₂′)＝(x₂+vtcosθ₂,y₂+vtsinθ₂)(2)

wherein (x)₁′,y₁') is F₁Position coordinates at time t, (x)₂′,y₂') is F₂Bit at time tAnd (5) setting coordinates. Then F at this time₁And F₂The distance between (a) and (b) is:

${\mathrm{dist}}_{12}=\sqrt{{({x_2}^{\′}-{x_1}^{\′})}^2+{({y_2}^{\′}-{y_1}^{\′})}^2}---\left(3\right)$

in the formula, dist₁₂Is represented by F₁And F₂Is measured.

Substituting equations (1) and (2) into equation (3), deriving t on both sides of the equation, and letting ddist₁₂(dt is 0) to obtain dist₁₂Moment of taking minimum:

$t_{\min }=\frac{(x_2-x_1)(\cos {\mathrm{\θ}}_2-\cos {\mathrm{\θ}}_1)}{2v[\cos ({\mathrm{\θ}}_2-{\mathrm{\θ}}_1)-1]}+\frac{(y_2-y_1)(\sin {\mathrm{\θ}}_2-\sin {\mathrm{\θ}}_1)}{2v[\cos ({\mathrm{\θ}}_2-{\mathrm{\θ}}_1)-1]}---\left(4\right)$

if t_min>0, the existence moment t is represented to enable the distance between the two aircrafts to be minimum; then t is put_minThe specific value of (a) is substituted into the formula (3), and the minimum distance dist between the two aircrafts can be obtained₁₂(t_min)。

Due to the spacing conditions between aircraft, i.e., safety, collision, dangerous proximity and collision, etc., are defined based on the extent of the aircraft safety distance violation. Therefore, the invention adopts the safety interval violation degree between the aircrafts to describe the flight conflict situation in the airway network in a fine mode. Knowing that the set of aircraft to be optimized is F, assuming that the aircraft to be optimized is F_iAnd F_jA safety interval of_ijThen F is_iAnd F_jDegree of security gap violation b between_ijIs defined as:

$b_{\mathrm{ij}}=\frac{\left|\min (0,({\mathrm{dist}}_{\mathrm{ij}}\left(t_{\min }\right)-{\mathrm{\ϵ}}_{\mathrm{ij}}))\right|}{\left|{\mathrm{\ϵ}}_{\mathrm{ij}}\right|}---\left(5\right)$

wherein, b_ijIs a real number from 0 to 1, dist_ij(t_min) Representing the aircraft F to be optimized_iAnd F_jThe minimum distance between them, ∑ is the sum sign, min () means taking the smaller of the two values in brackets, if b_ijThe value is 0, the safety interval is kept between the aircrafts to be optimized, and no flight conflict exists between the aircrafts to be optimized; if b is_ijThe larger the value is, the deeper the safety interval violation degree between the aircrafts to be optimized is, and the more violent the flight conflict between the aircrafts to be optimized is; when b is_ij1, two aircrafts to be optimized collide.

The flight conflict situation among all the aircraft to be optimized in the set F can be defined as:

$\mathrm{CS}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{j>i}{\overset{n}{\mathrm{\Σ}}}{b}_{\mathrm{ij}}---\left(6\right)$

that is to say, the obtaining of the flight conflict situation estimation model according to the four-dimensional trajectory of the aircraft to be optimized may specifically include: obtaining flight conflict situations among all aircrafts to be optimized according to the following formula:

$\mathrm{CS}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{j>i}{\overset{n}{\mathrm{\Σ}}}\left[\frac{\left|\min (0,({\mathrm{dist}}_{\mathrm{ij}}\left(t\right)-{\mathrm{\ϵ}}_{\mathrm{ij}}))\right|}{\left|{\mathrm{\ϵ}}_{\mathrm{ij}}\right|}\right]---\left(7\right)$

s402, establishing an objective function.

In view of the cost savings in flight (of the aircraft to be optimized), the objective function is set with a view to eliminating conflicts between the aircraft to be optimized and to reducing the takeoff delay time, and therefore the objective function is established according to the following formula:

$F=\frac{1-\frac{1}{n}{\mathrm{\Σ}}_{i=1}^n\left(\frac{{\mathrm{\δ}}_i}{{\mathrm{\δ}}_{\max }}\right)}{1+\mathrm{CS}}---\left(8\right)$

wherein F represents individual fitness;_irepresenting the aircraft F to be optimized_iThe takeoff delay time of;_maxindicating a preset takeoff delay time, e.g.,_maxthe maximum departure delay time allowed objectively. The significance of the objective function is that the smaller the flight conflicts between the aircrafts to be optimized, the smaller the average takeoff delay of the aircrafts to be optimized, the larger the objective function value, the better the result, i.e. the better the scheme.

Then, the obtaining an objective function value corresponding to the flight conflict situation estimation model according to the objective function may specifically include: and obtaining an objective function value corresponding to the flight conflict situation estimation model according to a formula (8).

It should be noted that when the flight conflict situation estimation model and the objective function are given, S401 to S402 may be omitted, and it is not necessary to repeatedly calculate the flight conflict situations between all the aircraft to be optimized, and according to the flight conflict situations and the objective function, the objective function values corresponding to the flight conflict situation estimation model before and after updating are obtained. Wherein, the updating of the flight conflict situation estimation model is obtained according to S403-S407.

In addition, the invention also effectively improves the algorithm efficiency and saves the calculation cost through a self-adaptive mechanism.

And S403, dividing all individuals into M groups equally.

To cope with the enormous computational effort of large-scale optimization, all individuals are divided into equal groups of M, where M is an integer greater than or equal to 2, e.g., M-5.

And (3) for each group in the M groups, performing variation for a preset number of times by adopting a cultural genetic algorithm so as to optimize the takeoff delay time of the aircraft to be optimized in the group. Wherein, the groups are optimized simultaneously, in the respective optimization process of each group, different groups are independent, and after each mutation is performed for a preset number of times (refer to S404-S406), migration and replacement are performed between different groups, see S407.

S404, selecting individuals in each group.

In this step, the local search frequency in the parallel MA search process is used as a normal or gaussian distribution model:

$\mathrm{\γ}(G;\mathrm{\μ},\mathrm{\σ},\mathrm{\η})=\frac{1}{\sqrt{2\mathrm{\π}}\mathrm{\σ}}\exp (-\frac{1}{2}{\left(\frac{G-\mathrm{\μ}}{\mathrm{\σ}}\right)}^2)---\left(9\right)$

wherein γ represents the local search frequency; g represents the preset times; μ represents the mean of the gaussian distribution model; σ represents the standard deviation of the Gaussian distribution model; η represents the number of individuals in each group. The local search frequency is obtained according to equation (9). The present invention provides classification through an adaptive evolutionary algorithm, which may be referred to as a diversity-based static adaptive strategy.

And obtaining the individuals in the group for local search according to the local search frequency. Specifically, the individuals in the group for local search are obtained according to formula (9) and the number of individuals included in the group:

φ(G,ξ；μ,σ,η)＝γ*ξ(10)

wherein φ represents the number of individuals of the local search; ξ represents the number of individuals contained within the group. Further, the individuals in the group used for the local search are determined according to the number of the individuals of the local search. Wherein, the individuals in the group for local search can be realized in various ways, and the invention is not limited thereto.

S405, local search in each group.

And carrying out local search on each individual in the individuals carrying out the local search by adopting a preset local search strategy to obtain the individual with better individual fitness.

Specifically, for all selected individuals, the MA local search is sequentially performed, and the method includes: for each individual, traversing all the aircrafts to be optimized, if the self-adaptability value (the self-adaptability of the airplane is explained in the later global search) of a certain aircraft to be optimized is smaller than a judgment value, executing a local search method on the aircraft to be optimized, namely giving the aircraft to be optimized a random takeoff delay time again; if the number of the aircraft to be optimized is larger than or equal to the judgment value, skipping the aircraft to be optimized, and continuing to perform local search on the aircraft to be optimized; and comparing the individuals after the local search operation with the individuals before the local search operation, if the individual fitness value is larger, replacing the individuals before the local search operation, otherwise, discarding the individuals after the local search operation, and still reserving the individuals which are not operated before. The value of the judgment value needs to ensure the effectiveness of the method and the timeliness of the operation, so the value of the judgment value cannot be too large or too small, and the judgment value is set to be 0.3 through theoretical analysis and experimental tests.

And S406, global search in each group.

And respectively optimizing in each group by adopting a rapid genetic algorithm. Wherein the individual may further comprise the number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual. The global search in the MA may specifically be: obtaining self fitness of the aircraft to be optimized according to the number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual; and carrying out global search by adopting a preset global search strategy according to the self fitness of the aircraft to be optimized to obtain an individual with better self fitness.

A. And (4) selecting a method.

A fast genetic algorithm is determined.

It should be noted that the method used in the global search may be implemented in various ways, for example, the above-mentioned fast genetic algorithm, or a random selection method, or sorting flight conflicts, selecting a better/worse split, and so on. Here, the tournament selection method is exemplified.

B. And (4) a crossing method.

The crossing method used by the invention is designed aiming at the 4DT flight conflict resolution, and the population optimizing speed can be effectively improved at the initial optimization stage.

And the self-adaptability of each airplane in the special group is set for the airplane. Aircraft F to be optimized_iThe self-adaptive degree of (a) is shown as follows:

$S_i=\frac{1-\frac{{\mathrm{\δ}}_i}{{\mathrm{\δ}}_{\max }}}{1+{\mathrm{NG}}_i}---\left(11\right)$

wherein NC is_iRepresenting the aircraft F to be optimized_iNumber of conflicts, S, with other aircraft to be optimized_iRepresenting the aircraft F to be optimized_iSelf-fitness of (2).

Fig. 5 is a schematic diagram of a crossing method in the second embodiment of the flight conflict resolution method of the present invention. As shown in fig. 5, in any two individuals a and B in the group, the self-fitness of each aircraft to be optimized is correspondingly compared, where the aircraft to be optimized in the individual a is respectively represented by Ai, the aircraft to be optimized in the individual B is respectively represented by Bi, and i represents all the aircraft to be optimized in the individual in a traversal manner. If S_Ai>S_BiThen the corresponding variables in the two sub-individuals inherit the genes of the individual Ai; if S_Ai<S_BiIn other cases, the corresponding variable in the two children is a linear combination of the individuals Ai and Bi, for example, according to the random parameter α, the corresponding variable C1 in child individual 1 is (1- α) × Ai + α × Bi, the corresponding variable C2 in child individual 2 is α × Ai + (1- α) × Bi., wherein the crossover operation is pc by probability, i.e., C1 crosses pc by (1- α) × Ai + α × Bi and α × Ai + (1- α) × Bi, and C2 crosses accordingly.

C. A mutation method.

Calculating self-fitness corresponding to each aircraft to be optimized, and if the self-fitness is smaller than the variation parameter, randomly generating a takeoff delay time of the aircraft to be optimized againReplacing the original variable value with the individual value, namely carrying out mutation; otherwise, no mutation is performed. As shown in FIG. 6, the reason S₁<、S₅<And S₇<And carrying out variation on the aircrafts to be optimized, namely A1, A5 and A7. The mutation probability is pm. The advantage of this variation method is similar to the crossover method described above, and better and faster optimization is possible.

And S407, performing migration exchange between the groups every preset times.

In order to ensure that each group can share the best individual evolved, communication among the groups is carried out at preset intervals. The method comprises the following steps: in the scheme, the preset times are 10, and the communication is performed between each group every 10 times, wherein each group forms a ring shape, as shown in fig. 3; the best individual in each group is copied to the next group, replacing the worst individual, thus improving the overall quality of the individuals of each group.

And S408, optimizing the flight conflict situation estimation model.

According to the formula (8), obtaining the objective function value corresponding to the flight conflict situation estimation model updated in the step S407 and the objective function value corresponding to the flight conflict situation estimation model in the step S401, and reserving the flight conflict situation estimation model corresponding to the larger value of the two objective function values as a better flight conflict situation estimation model.

EXAMPLE III

Fig. 7 is a schematic structural diagram of a third embodiment of the flight conflict resolution device according to the present invention. The invention provides a flight conflict resolution device which can be integrated into any terminal device such as a computer, a notebook computer or a PDA. As shown in fig. 7, the flight conflict solution apparatus 70 includes: an acquisition module 71, an operation module 72, a mutation module 73, and an update module 74.

The obtaining module 71 is configured to obtain a flight conflict situation estimation model according to a four-dimensional trajectory of an aircraft to be optimized, where the flight conflict situation estimation model includes a plurality of individuals, each of the plurality of individuals includes takeoff delay times of all aircraft to be optimized, and different individuals are different from each other; the operation module 72 is configured to obtain an objective function value corresponding to the flight conflict situation estimation model obtained by the obtaining module 71 according to an objective function, where the objective function is established according to flight conflict situations among all aircraft to be optimized; the variation module 73 is configured to divide all individuals in the flight conflict situation estimation model acquired by the acquisition module 72 into M groups, and for each group, perform variation for a preset number of times by using a cultural genetic algorithm to optimize the takeoff delay time of the aircraft to be optimized in the group, where M is an integer greater than or equal to 2; the updating module 74 is configured to arrange the M groups after variation obtained by the variation module 73 into a loop, use any one group in the loop as an initial group, sequentially copy the optimal individuals in each group to a next group, replace the worst individuals in the next group, and obtain an updated flight conflict situation estimation model until the previous group in the initial group is finished; the operation module 72 may also be configured to obtain, according to the objective function, an objective function value corresponding to the updated flight conflict situation estimation model obtained by the updating module 74, and retain the flight conflict situation estimation model corresponding to the larger value of the two objective function values.

The apparatus of this embodiment may be used to implement the technical solutions of the method embodiments shown in fig. 1 or fig. 4, and the implementation principles and technical effects are similar, which are not described herein again.

In the above embodiment, the obtaining module 71 may specifically be configured to: obtaining flight Conflict Situations (CS) among all aircrafts to be optimized according to the following formula:

$\mathrm{CS}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\underset{j>i}{\overset{n}{\mathrm{\&Sigma;}}}\left[\frac{\left|\min (0,({\mathrm{dist}}_{\mathrm{ij}}\left(t\right)-{\mathrm{\&epsiv;}}_{\mathrm{ij}}))\right|}{\left|{\mathrm{\&epsiv;}}_{\mathrm{ij}}\right|}\right]$

wherein,_ijrepresenting the aircraft F to be optimized_iAnd F_jA safety interval therebetween; dist_ij(t) denotes the aircraft F to be optimized_iAnd F_jThe minimum distance between the aircraft and the aircraft to be optimized, ∑ is a summation symbol, min () represents the smaller of the two values in brackets, and n represents the total number of the aircraft to be optimized.

On the basis of the above embodiment, the operation module 72 can be specifically configured to: obtaining an objective function value corresponding to the flight conflict situation estimation model according to the following formula:

$F=\frac{1-\frac{1}{n}{\mathrm{\&Sigma;}}_{i=1}^n\left(\frac{{\mathrm{\&delta;}}_i}{{\mathrm{\&delta;}}_{\max }}\right)}{1+\mathrm{CS}}$

wherein F represents individual fitness;_irepresenting the aircraft F to be optimized_iThe takeoff delay time of;_maxrepresenting a preset takeoff delay time.

In the above embodiment, when the mutation module 73 employs a local search in a cultural genetic algorithm, it is specifically configured to: obtaining local search frequency according to a Gaussian distribution model expressed asWherein gamma represents the local search frequency, G represents the preset times, mu represents the mean value of a Gaussian distribution model, sigma represents the standard deviation of the Gaussian distribution model, η represents the number of individuals in each group, individuals for local search in the group are obtained according to the local search frequency, and each individual for local search is locally searched by adopting a preset local search strategy to obtain an individual with better self-fitness.

Further, the individuals may further include the number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual, and when the variation module 73 adopts global search in the cultural genetic algorithm, the variation module is specifically configured to: obtaining self fitness of the aircraft to be optimized according to the number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual; and carrying out global search by adopting a preset global search strategy according to the self fitness of the aircraft to be optimized to obtain an individual with better self fitness.

Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. A flight conflict resolution method, comprising:

acquiring a flight conflict situation estimation model according to a four-dimensional track of an aircraft to be optimized, wherein the flight conflict situation estimation model comprises a plurality of individuals, each individual in the plurality of individuals comprises the take-off delay time of all the aircraft to be optimized, and different individuals are different;

obtaining an objective function value corresponding to the flight conflict situation estimation model according to an objective function, wherein the objective function is established according to flight conflict situations among all aircrafts to be optimized;

dividing all individuals in the flight conflict situation estimation model into M groups, and performing variation on each group for preset times by adopting a cultural genetic algorithm to optimize the takeoff delay time of the aircraft to be optimized in the group, wherein M is an integer greater than or equal to 2;

arranging the M groups to form a loop, taking any one group in the loop as an initial group, sequentially copying the optimal individual in each group to a next group, replacing the worst individual in the next group until the previous group in the initial group is finished, and obtaining an updated flight conflict situation estimation model;

and obtaining an objective function value corresponding to the updated flight conflict situation estimation model according to the objective function, and reserving the flight conflict situation estimation model corresponding to the larger value of the two objective function values.

2. The method according to claim 1, wherein the obtaining a flight conflict situation estimation model from the four-dimensional trajectory of the aircraft to be optimized comprises:

obtaining flight Conflict Situations (CS) among all aircrafts to be optimized according to the following formula:

$\mathrm{CS}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\underset{j>i}{\overset{n}{\mathrm{\&Sigma;}}}\left[\frac{\left|\min (0,({\mathrm{dist}}_{\mathrm{ij}}\left(t\right)-{\mathrm{\&epsiv;}}_{\mathrm{ij}}))\right|}{\left|{\mathrm{\&epsiv;}}_{\mathrm{ij}}\right|}\right]$

wherein,_ijrepresenting the aircraft F to be optimized_iAnd F_jA safety interval therebetween; dist_ij(t) denotes the aircraft F to be optimized_iAnd F_jThe minimum distance between the aircraft and the aircraft to be optimized, ∑ is a summation symbol, min () represents the smaller of the two values in brackets, and n represents the total number of the aircraft to be optimized.

3. The method according to claim 2, wherein the obtaining an objective function value corresponding to the flight conflict situation estimation model according to an objective function comprises:

obtaining an objective function value corresponding to the flight conflict situation estimation model according to the following formula:

$F=\frac{1-\frac{1}{n}{\mathrm{\&Sigma;}}_{i=1}^n\left(\frac{{\mathrm{\&delta;}}_i}{{\mathrm{\&delta;}}_{\max }}\right)}{1+\mathrm{CS}}$

wherein F represents individual fitness;_irepresenting the aircraft F to be optimized_iThe takeoff delay time of;_maxrepresenting a preset takeoff delay time.

4. The method according to any one of claims 1 to 3, wherein the local search in the cultural genetic algorithm is specifically:

obtaining local search frequency according to a Gaussian distribution model expressed asWherein gamma represents the local search frequency, G represents the preset times, mu represents the mean value of a Gaussian distribution model, sigma represents the standard deviation of the Gaussian distribution model, and η represents the number of individuals in each group;

obtaining individuals in the group for local search according to the local search frequency;

and carrying out local search on each individual in the individuals carrying out local search by adopting a preset local search strategy to obtain the individual with better self-fitness.

5. The method according to any one of claims 1 to 4, wherein the individual further comprises the number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual, and the global search in the cultural genetic algorithm is specifically:

obtaining self fitness of the aircraft to be optimized according to the number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual;

and carrying out global search by adopting a preset global search strategy according to the self fitness of the aircraft to be optimized to obtain an individual with better self fitness.

6. A flight conflict resolution apparatus, comprising:

the system comprises an acquisition module, a judgment module and a processing module, wherein the acquisition module is used for acquiring a flight conflict situation estimation model according to a four-dimensional track of an aircraft to be optimized, the flight conflict situation estimation model comprises a plurality of individuals, each individual in the plurality of individuals comprises the takeoff delay time of all the aircraft to be optimized, and different individuals are different;

the operation module is used for obtaining an objective function value corresponding to the flight conflict situation estimation model obtained by the obtaining module according to an objective function, wherein the objective function is established according to flight conflict situations among all aircrafts to be optimized;

the variation module is used for dividing all individuals in the flight conflict situation estimation model acquired by the acquisition module into M groups, and for each group, performing variation for preset times by adopting a cultural genetic algorithm to optimize the takeoff delay time of the aircraft to be optimized in the group, wherein M is an integer greater than or equal to 2;

the updating module is used for arranging the M groups obtained by the variation module after variation to form a cycle, taking any one group in the cycle as an initial group, sequentially copying the optimal individual in each group to a next group, replacing the worst individual in the next group until the previous group in the initial group is finished, and obtaining an updated flight conflict situation estimation model;

the operation module is further configured to obtain, according to the objective function, an objective function value corresponding to the updated flight conflict situation estimation model obtained by the update module, and retain the flight conflict situation estimation model corresponding to the larger of the two objective function values.

7. The device of claim 6, wherein the obtaining module is specifically configured to:

obtaining flight Conflict Situations (CS) among all aircrafts to be optimized according to the following formula:

$\mathrm{CS}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\underset{j>i}{\overset{n}{\mathrm{\&Sigma;}}}\left[\frac{\left|\min (0,({\mathrm{dist}}_{\mathrm{ij}}\left(t\right)-{\mathrm{\&epsiv;}}_{\mathrm{ij}}))\right|}{\left|{\mathrm{\&epsiv;}}_{\mathrm{ij}}\right|}\right]$

wherein,_ijrepresenting the aircraft F to be optimized_iAnd F_jA safety interval therebetween; dist_ij(t) denotes the aircraft F to be optimized_iAnd F_jThe minimum distance between the aircraft and the aircraft to be optimized, ∑ is a summation symbol, min () represents the smaller of the two values in brackets, and n represents the total number of the aircraft to be optimized.

8. The device of claim 7, wherein the operation module is specifically configured to:

obtaining an objective function value corresponding to the flight conflict situation estimation model according to the following formula:

$F=\frac{1-\frac{1}{n}{\mathrm{\&Sigma;}}_{i=1}^n\left(\frac{{\mathrm{\&delta;}}_i}{{\mathrm{\&delta;}}_{\max }}\right)}{1+\mathrm{CS}}$

wherein F represents individual fitness;_irepresenting the aircraft F to be optimized_iThe takeoff delay time of;_maxrepresenting a preset takeoff delay time.

9. The apparatus according to any of the claims 6-8, wherein the mutation module, when using a local search in a cultural genetic algorithm, is specifically configured to:

obtaining local search frequency according to a Gaussian distribution model expressed asWherein gamma represents the local search frequency, G represents the preset times, mu represents the mean value of a Gaussian distribution model, sigma represents the standard deviation of the Gaussian distribution model, and η represents the number of individuals in each group;

obtaining individuals in the group for local search according to the local search frequency;

and carrying out local search on each individual in the individuals carrying out local search by adopting a preset local search strategy to obtain the individual with better self-fitness.

10. The apparatus according to any one of claims 6 to 9, wherein the individual further comprises a number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual, and the variation module, when using a global search in a cultural genetic algorithm, is configured to:

obtaining self fitness of the aircraft to be optimized according to the number of conflicts between each aircraft to be optimized and other aircraft to be optimized in the individual;

and carrying out global search by adopting a preset global search strategy according to the self fitness of the aircraft to be optimized to obtain an individual with better self fitness.