A method for optimizing the dispatch of multiple gliding aircraft launched from a carrier

Abstract

The present application relates to a kind of carrier release multi-gliding aircraft optimization scheduling method.The present application aims to solve the problem that the existing method less considers resource scheduling problem, when the number of gliding aircraft is large, the timeliness of task cannot be considered;And crow search algorithm mainly solves continuous problem, it is easy to fall into local optimum, and optimal solution can not be found under the problem of shorter planning time constraint.Process is as follows:one, constructs multi-gliding aircraft airborne release scheduling model;Two, the time t of the i-th gliding aircraft from the j-th aircraft flies to the k-th target is solved ijk , t ijk It is brought into multi-gliding aircraft airborne release scheduling model to calculate distribution index;Three, based on distribution index and improved hybrid crow optimization algorithm, the optimal position in all crows is output, that is, optimal distribution strategy is obtained.The present application is used in the field of carrier release multi-gliding aircraft optimization scheduling.

Description

Technical Field

[0001] The invention relates to an optimization scheduling method for launching multiple gliding aircraft from a carrier aircraft. Background Art

[0002] A glider is a medium- to long-range unmanned transport aircraft that can carry and airdrop important items such as cargo, medicine, food, or strategic resources. Specifically, a single aircraft can carry multiple gliders. Once the aircraft reaches its target, it releases the glider at a specific altitude, speed, and angle. The glider then flies unpowered to deliver the cargo to the target. After the cargo is released, it can be recovered and reused. Therefore, compared to direct delivery from an aircraft, using a glider is more efficient, requires fewer pilots, and faces less pressure on airspace planning.

[0003] Disadvantages of existing technology

[0004] (1) Reducing the time it takes for multiple gliders to deliver cargo to multiple destinations is of great significance. However, existing glider collaborative planning methods rarely consider resource scheduling issues. When the number of gliders is large, it is difficult to take into account the timeliness of the mission.

[0005] (2) The crow search algorithm has played an important role in the field of task planning, but it mainly solves continuous problems. The crow search algorithm is prone to fall into local optimality and may not be able to find the optimal solution under shorter planning time constraints. Summary of the Invention

[0006] The purpose of the present invention is to solve the problems that the existing collaborative planning methods for gliding aircraft pay little attention to resource scheduling issues and cannot take into account the timeliness of tasks when the number of gliding aircraft is large; and the crow search algorithm mainly solves continuous problems and is prone to falling into local optimality, and may not be able to find the optimal solution under short planning time constraints. Therefore, a method for optimizing the scheduling of multiple gliding aircraft deployed by carrier aircraft is proposed.

[0007] A method for optimizing the scheduling of multi-glider aircraft deployed by a carrier aircraft includes the following steps:

[0008] Step 1: Construct a multi-glider aircraft airborne delivery scheduling model;

[0009] Step 2: Calculate the time t for the i-th glider to fly from the j-th aircraft to the k-th target ijk , t ijk Bring in multiple glider aircraft onboard delivery scheduling models to calculate allocation indicators;

[0010] Step 3: Based on the allocation index and the improved hybrid crow optimization algorithm, the optimal position of all crows is output, that is, the optimal allocation strategy is obtained.

[0011] The beneficial effects of the present invention are:

[0012] The present invention proposes an optimized scheduling method for airborne delivery of gliding aircraft, which adopts an improved hybrid crow optimization algorithm and considers multiple aircraft with different altitudes, speeds, positions and headings to simultaneously deliver multiple gliding aircraft from different angles, so as to achieve the shortest total time for multiple gliding aircraft to deliver cargo to different targets.

[0013] The crow search algorithm plays an important role in the field of mission planning. Since the glider aircraft airborne delivery scheduling problem of the present invention is a discrete problem, it cannot be directly processed by the hybrid crow optimization algorithm. The hybrid crow optimization algorithm needs to be improved. The improved hybrid crow optimization algorithm is used to solve the glider aircraft airborne delivery scheduling problem of the present invention and find the optimal solution under the short planning time constraint. The improved hybrid crow optimization algorithm of the present invention considers the resource scheduling problem and takes into account the timeliness of the task when the number of glider aircraft is large.

[0014] The present invention solves the problems that existing collaborative planning methods for gliding aircraft pay little attention to resource scheduling issues and cannot take into account the timeliness of tasks when the number of gliding aircraft is large; and the problem that the crow search algorithm mainly solves continuous problems, is prone to falling into local optimality, and may not be able to find the optimal solution under short planning time constraints.

[0015] Key technical points of the present invention

[0016] (1) Multi-glider airborne delivery scheduling model

[0017] Considering that multiple aircraft with different altitudes, speeds, positions and headings launch multiple gliders simultaneously from different angles, and considering the flight time of the gliders from launch to landing, with the goal of minimizing the total completion time and maximizing the efficiency of multiple gliders performing attack missions on multiple targets, an airborne launch scheduling model for gliders is established.

[0018] (2) Improved hybrid crow optimization algorithm

[0019] An improved hybrid crow optimization algorithm is proposed. A high-quality initial crow population is obtained by initializing twice. At the same time, a chaotic perturbation strategy is introduced to perturb the positions of a few crows with outstanding performance to avoid falling into local optimality.

[0020] The present invention relates to a method for optimizing the scheduling of multiple glider aircraft airborne delivery. This method considers the simultaneous delivery of multiple glider aircraft from different angles by multiple aircraft with different altitudes, speeds, positions, and headings. The method combines the flight sequence and delivery conditions of each glider aircraft, the flight status of each carrier aircraft, and the location of each target. The method also considers the flight time from delivery to landing of each glider aircraft. With the goal of minimizing the total completion time and maximizing the efficiency of multiple glider aircraft attacking multiple targets, an optimization model for scheduling glider aircraft airborne delivery is established. Furthermore, an improved hybrid crow optimization algorithm is proposed to improve the quality and diversity of the initial population, facilitate the rapid optimization of the crow algorithm, and ultimately obtain the optimal allocation strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Optimal scheduling model diagram for multi-glider airborne delivery;

[0022] Figure 2 Flowchart of the improved hybrid crow optimization algorithm. DETAILED DESCRIPTION

[0023] Specific embodiment 1: This embodiment is a carrier aircraft to launch a multi-glider aircraft optimization scheduling method. The specific process is as follows:

[0024] Step 1: Construct a multi-glider aircraft airborne delivery scheduling model;

[0025] Step 2: Calculate the time t for the i-th glider to fly from the j-th aircraft to the k-th target ijk , t ijk Bring in multiple glider aircraft onboard delivery scheduling models to calculate allocation indicators;

[0026] Step 3: Based on the allocation index and the improved hybrid crow optimization algorithm, the optimal position of all crows is output, that is, the optimal allocation strategy is obtained.

[0027] Specific embodiment 2: This embodiment differs from specific embodiment 1 in that: in step 1, a multiple glider aircraft airborne delivery scheduling model is constructed; the specific process is as follows:

[0028] Step 1: Assume that N gliders are dropped from M aircraft onto Q targets.

[0029] The time it takes for the i-th glider to fly from the j-th aircraft to the k-th target is t ijk ;

[0030] i=1,2,…,N, j=1,2,…,M, k=1,2,…,Q;

[0031] N, M, and Q are all positive integers;

[0032] Step 1 and 2: Set the constraints for the multi-glider airborne delivery scheduling model; the conditions are as follows:

[0033] 1) N>Q, that is, the number of gliders is greater than the number of targets. Some targets may need to be assigned at least two gliders;

[0034] 2) Setting the initial launch conditions for the carrier aircraft to limit the turning radius of the glider aircraft, that is, the glider aircraft cannot turn towards the target with an excessive amplitude;

[0035] The initial launch conditions of the carrier aircraft include altitude, speed, heading, position, and angle;

[0036] 3) A glider can only reach one target;

[0037] 4) A target can be assigned to any glider;

[0038] Step 13: Based on steps 11 and 12, construct a multi-glider aircraft airborne delivery scheduling model.

[0039] Other steps and parameters are the same as those in the first embodiment.

[0040] Specific embodiment three: This embodiment differs from specific embodiment one or two in that: in step one-three, a multiple glider aircraft airborne delivery scheduling model is constructed based on steps one-one and one-two;

[0041] The expression of the multiple glider aircraft airborne delivery scheduling model is:

[0042]

[0043] Where: J is the allocation index, and the allocation index is used as the fitness;

[0044] k1 and k2 represent dimensionless weight coefficients, respectively;

[0045] t ijk is the time it takes for the i-th glider to fly from the j-th aircraft to the k-th target;

[0046] V ik is the decision variable. When the i-th glider does not fly to the k-th target, V ik =0, otherwise V ik =1;

[0047] The optimization variable is the decision variable V ik , that is, whether to perform allocation for glider aircraft i (i=1,2…,N) and target k (k=1,2…,Q), allocation is V ik =1, otherwise V ik =0;

[0048] R ik It represents the turning radius of the i-th glider flying to the k-th target. Obviously, R ik The smaller the better;

[0049] I k is the important attribute of the kth target, with a value range of [1,10]. More aircraft should be allocated to key targets during scheduling;

[0050] D k is the difficulty attribute of the kth target, such as terrain, weather, etc., with a value range of [1,10]. The greater the difficulty, the more difficult it is to transport, and more aircraft need to be allocated to it;

[0051] Time t ijk It mainly depends on the distance between the jth aircraft and the kth target.

[0052] Other steps and parameters are the same as those in the first or second embodiment.

[0053] Specific embodiment 4: This embodiment differs from any one of specific embodiments 1 to 3 in that: in step 2, the time t required for the i-th glider to fly from the j-th aircraft to the k-th target is calculated. ijk , t ijk Bring in multiple glider aircraft onboard deployment scheduling models to calculate allocation indicators; the specific process is as follows:

[0054] Step 2.1: Assume that the glider aircraft gradually lowers its altitude after being launched. The flight altitude h of the glider aircraft changes with the remaining range R of the glider aircraft. L Change, the expression is:

[0055]

[0056] Among them, R L represents the remaining range of the glider aircraft, a1, a2, a3, and a4 are polynomial coefficients, and h is the flight altitude of the glider aircraft;

[0057] Step 22: Flight altitude h and remaining range R of the glider L Satisfies the following relationship:

[0058] h′(t)=Vsinγ

[0059] R L ′(t)=-Vcosγ (3)

[0060] Where h′(t) represents the derivative of the flight altitude h of the glider with respect to time t;

[0061] R L ′(t) is the remaining range R of the glider aircraft L The derivative with respect to time t;

[0062] V and γ are the current speed and inclination angle of the glider aircraft, respectively;

[0063] Step 2: From formula (2), we get:

[0064]

[0065] Step 24: From formula (3), we get:

[0066]

[0067] Step 25: Combining equations (4) and (5), we get:

[0068]

[0069] Step 26: Set the target point altitude and the remaining range R of the glider when it reaches the target. L are all zero, and the inclination angle of the glider aircraft when it reaches the target is -80°;

[0070] Combined with the current flight altitude h of the glider aircraft and the current remaining range R of the glider aircraft L and the current inclination angle γ of the glider, the combined equations (2) and (6):

[0071]

[0072] Based on formula (7), the four unknown coefficients a1 to a4 are solved;

[0073] e -h Indicates the Take the logarithm,

[0074] Step 27: From formula (3), we get:

[0075]

[0076] Considering the uniform motion of a gliding aircraft, integrate both sides of equation (8) and introduce the first equation of equation (7): Get time t;

[0077] Step 28: Take time t as the time t for the i-th glider to fly from the j-th aircraft to the k-th target ijk , t ijk Substitute into formula (1) to calculate the allocation index J.

[0078] The other steps and parameters are the same as those in the first to third embodiments.

[0079] Specific embodiment 5: This embodiment differs from any one of specific embodiments 1 to 4 in that: in step 27, the uniform motion of the gliding aircraft is considered, both sides of equation (8) are integrated, and the first equation of equation (7) is introduced to obtain the time t; the expression is:

[0080]

[0081] Among them, δR L Indicates that the range interval [0,R L ] is divided into equal intervals, here we take δR L =5km; n represents a positive integer;

[0082] At this point, V, γ, h, and R can be determined based on the glider launch position and target position. L , combined with formula (7) to obtain coefficients a1~a4, and finally use formula (9) to obtain t ijk , which makes it easier to calculate the planning indicators in formula (1).

[0083] Excellent initial population helps to improve the performance of the algorithm. The present invention uses the important attributes of each target I k and difficulty attribute D k , combined with the time obtained by formula (9), the obviously unreasonable results are preliminarily screened out, and the chaotic perturbation strategy is introduced in the iteration.

[0084] The other steps and parameters are the same as those in the first to fourth embodiments.

[0085] Specific embodiment 6: This embodiment differs from any one of specific embodiments 1 to 5 in that: in step 3, the optimal position of all crows is output based on the allocation index and the improved hybrid crow optimization algorithm, that is, the optimal allocation strategy is obtained;

[0086] The optimal position for the crow is the allocation scheme, for example, target 1 is allocated to the first glider, which is equal to 1, and then target 2 is allocated to the second glider, which is equal to 2; Figure 2 ;

[0087] Crows are highly intelligent birds, with strong abilities for memory, communication, storage, following, and stealing. Inspired by this, the crow optimization algorithm relies on four basic principles for its optimization: ① Crows live in groups; ② Crows know where food is; ③ Crows follow and steal food from each other; and ④ Crows protect food with a certain probability.

[0088] The specific process is:

[0089] Step 3: 1. Set the number of iterations l = 1 and randomly initialize the positions of all crows;

[0090] Step 32: Randomly initialize the locations where all crows store food;

[0091] Step 3. Each crow randomly chooses an object to follow;

[0092] Step 3 and 4: Calculate the fitness J of each crow's position according to the multiple glider aircraft airborne delivery scheduling model (1);

[0093] Step 35: Update the current location of all crows and the location of all crows' food storage;

[0094] Step 36: Add chaos disturbance to the current positions of all crows updated in step 35 to obtain the current positions of all crows updated again;

[0095] Step 37: Repeat steps 33-36 until the maximum number of iterations is reached, and output the optimal position of all crows, that is, the optimal allocation strategy is obtained.

[0096] Position refers to which glider is assigned to which target, which is an allocation relationship and allocation scheme;

[0097] For example, 1 is paired with A, 2 is paired with B, 3 is paired with C, and 4 is paired with D. This is a distribution scheme, defined as a position. Similarly, if 1 is paired with Z, 2 is paired with W, and 3 is paired with X, this is also another correspondence, which is also another position.

[0098] The position is related to the time t in formula 1, and the time t is obtained from the position;

[0099] The positions are represented as numbers, for example, 1234, which is equivalent to crows 1 to 4 being assigned to the target, 1234 or 4321 or 1423, and so on. With the assignment results, the assignment relationship between the aircraft and the target can be determined, and then the aircraft can be further obtained; the time t is calculated based on the glider's altitude, speed, inclination angle, remaining range, etc.

[0100] The other steps and parameters are the same as those in the first to fifth embodiments.

[0101] Specific embodiment 7: This embodiment differs from any one of specific embodiments 1 to 6 in that in step 31, the number of iterations l is set to 1, and the positions of all crows are randomly initialized; the specific process is as follows:

[0102] 1a) Randomly select 20% of the crows;

[0103] 1b), randomly initialize the position of the crows extracted in step 1a);

[0104] 1c) Calculate the fitness value J of each crow's position according to the multi-glider aircraft airborne delivery scheduling model (1), and sort the fitness values ​​from low to high;

[0105] Position refers to which glider is assigned to which target, which is an allocation relationship and allocation scheme;

[0106] For example, 1 is paired with A, 2 is paired with B, 3 is paired with C, and 4 is paired with D. This is a distribution scheme, defined as a position. Similarly, if 1 is paired with Z, 2 is paired with W, and 3 is paired with X, this is also another correspondence, which is also another position.

[0107] The position is related to the time t in formula 1, and the time t is obtained from the position;

[0108] The positions are represented as numbers, for example, 1234, which is equivalent to crows 1 to 4 being assigned to the target, 1234 or 4321 or 1423, and so on. With the assignment results, the assignment relationship between the aircraft and the target can be determined, and then the aircraft can be further obtained; the time t is calculated based on the glider's altitude, speed, inclination angle, remaining range, etc.

[0109] 1d) Repeat steps 1b) to 1c) for β times to obtain the position of each crow sorted from low to high fitness value;

[0110] 1e) Select the positions of the bottom 15% of crows in 1d) and delete them;

[0111] 1f) Take the positions of the remaining crows as the positions of all randomly initialized crows.

[0112] The other steps and parameters are the same as those in the first to sixth embodiments.

[0113] Specific embodiment eight: This embodiment differs from any one of specific embodiments one to seven in that, in 1d), β=10.

[0114] The other steps and parameters are the same as those in the first to seventh embodiments.

[0115] Specific embodiment 9: This embodiment differs from any one of specific embodiments 1 to 8 in that in step 35, the current positions of all crows and the positions of all crows storing food are updated; the specific process is as follows:

[0116] Let the position of crow i in the lth iteration be x i l (optimization variable),

[0117] as follows:

[0118]

[0119] in, The position of crow i in 1-D space in the lth iteration, The position of crow i in 2-D space in the lth iteration, The position of crow i in the lth iteration; i = 1, 2,..., δ, δ is a positive integer;

[0120] Suppose crow j visits the position where crow j stores food in the lth iteration And crow i follows crow j to steal the food stored by crow j; j ≠ i;

[0121] If crow i is not discovered by crow j, crow i updates its position in the (l+1)th iteration according to the following expression The expression is:

[0122]

[0123] Where, r i is a random number between 0 and 1;

[0124] L is the distance from crow j to the position m l j where crow j stores food;

[0125] The position of crow i in the (l+1)th iteration;

[0126] The position of crow i in the lth iteration;

[0127] If crow i has been discovered by crow j, crow j adopts a deceptive strategy and flies to a random position that is irrelevant to the position where crow j stores food The expression is:

[0128]

[0129] Where, The position of crow j in the (l+1)th iteration, The position of crow j in the lth iteration, R j is a random number between -1 and 1;

[0130] And the position of crow i after being deceived is:

[0131]

[0132] Where, The position of crow i in the (l+1)th iteration, The position of crow i in the lth iteration, R i is a random number between -1 and 1;

[0133] Crow i also stores food, and crow i will compare the location of crow i's food storage and crow i follows crow j to find the new position According to the allocation index J, a better location (the location corresponding to the smaller value of the allocation index J) is selected as the new food storage location for crow i The expression is:

[0134]

[0135] The other steps and parameters are the same as those in Specific Embodiments 1 to 8.

[0136] Specific embodiment 10: This embodiment differs from any one of specific embodiments 1 to 9 in that, in step 36, chaotic disturbance is added to the current positions of all crows updated in step 35 to obtain the current positions of all crows updated again; the specific process is as follows:

[0137] Use Skew Tent mapping to generate chaotic sequences:

[0138]

[0139] Where x is a random number between [0,1];

[0140] i corresponds to crow number i;

[0141] y i is a random number between [0,1] corresponding to the i-th crow, y i+1 is a random number between [0,1] corresponding to the i+1th crow;

[0142] Think when 0 <x<1且0<y<1时系统处于混沌状态;

[0143] Generate chaotic sequence y1,y2... according to formula (15);

[0144] Use Logistic mapping to generate chaotic random numbers and apply them to the first 15% of crows:

[0145]

[0146] in, is the position of crow j in the lth iteration, y j is a random number between [0,1] corresponding to crow j.

[0147] The other steps and parameters are the same as those in Specific Embodiments 1 to 9.

[0148] The present invention may have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

Claims

1. A method for optimizing the dispatching of multiple glider aircraft by a carrier aircraft, characterized by: The specific process of the method is: Step 1: Construct a multi-glider aircraft airborne delivery scheduling model; Step 2: Calculate the time t for the i-th glider to fly from the j-th aircraft to the k-th target ijk , t ijk Bring in multiple glider aircraft onboard delivery scheduling models to calculate allocation indicators; Step 3: Based on the allocation index and the improved hybrid crow optimization algorithm, the optimal position of all crows is output, that is, the optimal allocation strategy is obtained; The expression of the multiple glider aircraft airborne delivery scheduling model is: Where: J is the allocation index, and the allocation index is used as the fitness; k1 and k2 represent weight coefficients respectively; t ijk is the time it takes for the i-th glider to fly from the j-th aircraft to the k-th target; V ik is the decision variable. When the i-th glider does not fly to the k-th target, V ik =0, otherwise V ik =1; R ik represents the turning radius of the i-th glider flying towards the k-th target; I k is the important attribute of the k-th target, with a value range of [1,10]; D k is the difficulty attribute of the kth target, with a value range of [1,10]; In step 2, the time t required for the i-th glider to fly from the j-th aircraft to the k-th target is calculated. ijk , t ijk Bring in multiple glider aircraft onboard deployment scheduling models to calculate allocation indicators; the specific process is as follows: Step 21: The flight altitude h of the glider aircraft changes with the remaining range R of the glider aircraft. L Change, the expression is: Among them, R L represents the remaining range of the glider aircraft, a1, a2, a3, and a4 are polynomial coefficients, and h is the flight altitude of the glider aircraft; Step 22: Flight altitude h and remaining range R of the glider L The following relationship is satisfied: h′(t)=Vsinγ R L ′(t)=-Vcosγ (3) Where h′(t) represents the derivative of the flight altitude h of the glider with respect to time t; R L ′(t) is the remaining range R of the glider aircraft L The derivative with respect to time t; V and γ are the current speed and inclination angle of the glider aircraft, respectively; Step 2: From formula (2), we get: Step 24: From formula (3), we get: Step 25: Combining equations (4) and (5), we get: Step 26: Set the target point altitude and the remaining range R of the glider when it reaches the target. L are all zero, and the inclination angle of the glider aircraft when it reaches the target is -80°; Combined with the current flight altitude h of the glider aircraft and the current remaining range R of the glider aircraft L and the current inclination angle γ of the glider, the combined equations (2) and (6): Based on formula (7), the four unknown coefficients a1 to a4 are solved; Step 27: From formula (3), we get: Considering the uniform motion of a gliding aircraft, integrate both sides of equation (8) and introduce the first equation of equation (7): Get time t; Step 28: Take time t as the time t for the i-th glider to fly from the j-th aircraft to the k-th target ijk , t ijk Substitute into formula (1) to calculate the allocation index J; In step 3, the optimal position of all crows is output based on the allocation index and the improved hybrid crow optimization algorithm, that is, the optimal allocation strategy is obtained; the specific process is: Step 3:

1. Set the number of iterations to 1 and randomly initialize the positions of all crows. Step 32: Randomly initialize the locations where all crows store food; Step 3. Each crow randomly chooses an object to follow; Step 3 and 4: Calculate the fitness J of each crow's position according to the multiple glider aircraft airborne delivery scheduling model (1); Step 35: Update the current location of all crows and the location of all crows' food storage; Step 36: Add disturbance to the current positions of all crows updated in step 35 to obtain the current positions of all crows updated again; Step 37: Repeat steps 33-36 until the maximum number of iterations is reached, and output the optimal position of all crows, that is, the optimal allocation strategy is obtained.

2. The method for optimizing the deployment of multiple gliders by a carrier aircraft according to claim 1, characterized in that: In the step 1, a multiple glider aircraft airborne delivery scheduling model is constructed; the specific process is as follows: Step 1: Assume that N gliders are dropped from M aircraft onto Q targets. The time it takes for the i-th glider to fly from the j-th aircraft to the k-th target is t ijk ; i=1,2,…,N, j=1,2,…,M, k=1,2,…,Q; Step 1 and 2: Set the constraints for the multi-glider airborne delivery scheduling model; the conditions are as follows: 1) N>Q, that is, the number of glider aircraft is greater than the target number; 2) Set the initial launch conditions for the carrier aircraft and limit the turning radius of the glider aircraft; The initial launch conditions of the carrier aircraft include altitude, speed, heading, position, and angle; 3) A glider can only reach one target; 4) A target can be assigned to any glider; Step 13: Based on steps 11 and 12, construct a multi-glider aircraft airborne delivery scheduling model.

3. The method for optimizing the deployment of multiple gliders by a carrier aircraft according to claim 2, characterized in that: In step 27, the uniform motion of the glider is considered, and the time t is obtained by integrating both sides of equation (8) and introducing the first equation of equation (7). The expression is: Among them, δR L Indicates that the range interval [0,R L ] is an equally divided interval; n represents a positive integer.

4. The method for optimizing the deployment of multiple gliders by a carrier aircraft according to claim 3, characterized in that: In step 31, let the number of iterations l = 1 and randomly initialize the positions of all crows; the specific process is: 1a) Randomly select 20% of the crows; 1b), randomly initialize the position of the crows extracted in step 1a); 1c) Calculate the fitness value J of each crow's position according to the multi-glider aircraft airborne delivery scheduling model (1), and sort the fitness values ​​from low to high; 1d) Repeat steps 1b) to 1c) for β times to obtain the position of each crow sorted from low to high fitness value; 1e) Select the positions of the bottom 15% of crows in 1d) and delete them; 1f) Take the positions of the remaining crows as the positions of all randomly initialized crows.

5. The method for optimizing the dispatching of multiple gliders from a carrier aircraft according to claim 4, characterized in that: In the above 1d), β=10.

6. The method for optimizing the dispatching of multiple gliders from a carrier aircraft according to claim 5, characterized in that: In the steps 3 and 5, the current positions of all crows and the locations where all crows store food are updated. The specific process is as follows: Let the position of crow i in the lth iteration be Suppose crow j visits the location where crow j stores food in iteration l Crow i follows crow j and steals the food stored by crow j; j≠i; If crow i is not found by crow j, crow i updates its position in the l+1th iteration according to the following result. The expression is: Among them, r i is a random number between [0,1]; L is the location where crow j stores food. distance; is the position of crow i in the l+1th iteration; is the position of crow i in the lth iteration; If crow i has been discovered by crow j, crow j will adopt a deceptive strategy and fly to a location where crow j stores food. Irrelevant random positions The expression is: in, is the position of crow j in the l+1th iteration, is the position of crow j in the lth iteration, R j is a random number between [-1,1]; The position of crow i after being deceived is: in, is the position of crow i in the l+1th iteration, is the position of crow i in the lth iteration, R i is a random number between [-1,1]; Crow i also stores food, and crow i will compare the location of crow i's food storage and crow i follows crow j to find the new position According to the allocation index J, a better location is selected as the new food storage location of crow i The expression is:

7. The method for optimizing the dispatching of multiple gliders by a carrier aircraft according to claim 6, characterized in that: In step 36, a disturbance is added to the current positions of all crows updated in step 35 to obtain the current positions of all crows updated again. The specific process is as follows: Generate random numbers and apply them to the top 15% of crows: in, is the position of crow j in the lth iteration, y j is a random number between [0,1] corresponding to crow j.

Images