Multi-aircraft multi-site delivery scheduling planning method, storage medium and equipment

Abstract

The invention discloses a multi-aircraft multi-site delivery scheduling planning method, a storage medium and equipment, and belongs to the technical field of aircraft delivery scheduling planning. According to the method, the problem that an optimal scheduling scheme for accessing a plurality of sites by a large-scale aircraft cannot be obtained by an existing method is solved. According to the method, the scheduling planning problem that a plurality of aircrafts successively arrive at a plurality of sites is considered, the processes of unloading, loading, refueling, maintenance and the like are considered, the access sequence of the aircrafts to the sites is optimized, and a multi-aircraft multi-site delivery scheduling model is established; a discrete data iterative sorting algorithm based on a neural network is provided for the problem of complex sorting of a large number of aircrafts, the sorting problem of high-dimensional input is solved through layer-by-layer decomposition iteration, the target of shortest total consumed time of material delivery is achieved, and a better delivery scheduling scheme is obtained. The method can be applied to multi-aircraft multi-site delivery scheduling planning.

Description

Technical Field

[0001] The present invention belongs to the technical field of aircraft delivery scheduling and planning, and particularly relates to a multi-aircraft multi-site delivery scheduling and planning method, a storage medium and a device. Background Art

[0002] Unmanned aircraft is a type of small and medium-sized unmanned aerial vehicle with a large lift-to-drag ratio gliding ability and a large inner cabin transportation ability, which can replace manned aircraft to complete freight tasks. With the rapid development of the low-altitude economy and the Internet of Everything, higher requirements are put forward for the timeliness and economy of air freight.

[0003] However, the existing remote aircraft collaborative planning methods mainly consider time or space collaboration, and have not considered the optimal scheduling problem of a large number of aircraft successively visiting multiple targets. There is no relevant research on related optimization models and optimization problems. Moreover, the conventional sorting algorithm sorts all data at once, with insufficient calculation speed and accuracy, and lacks consideration for the scheduling and sorting problem of a large number of aircraft visiting multiple sites, and cannot obtain the best scheduling plan. Summary of the Invention

[0004] The purpose of the present invention is to solve the problem that the existing methods cannot obtain the best scheduling plan for a large number of aircraft to visit multiple sites, and a multi-aircraft multi-site delivery scheduling and planning method, a storage medium and a device are proposed.

[0005] The technical solution adopted by the present invention to solve the above technical problems is: a multi-aircraft multi-site delivery scheduling and planning method, and the method specifically includes the following steps:

[0006] Step 1: Denote the number of aircraft as N, the number of ground sites as M, and the waiting time consumed by each aircraft at each ground site is the same;

[0007] Step 2: Define Constraint (1), Constraint (2) and Constraint (3):

[0008] Constraint (1): An aircraft can reach a ground site at most once;

[0009] Constraint (2): At any moment, only one aircraft is admitted to a ground site;

[0010] Constraint (3): An aircraft arrives at each ground site in sequence according to the order of its delivery to the ground sites;

[0011] Establish a scheduling model according to the defined Constraint (1), Constraint (2) and Constraint (3);

[0012] Step 3: Establish an objective function for the scheduling plan according to the scheduling model;

[0013] Step 4: Solve the objective function established in Step 3 to obtain the scheduling plan result.

[0014] Further, the waiting time of the aircraft at the ground station includes the loading and unloading time, refueling time, and maintenance time.

[0015] Further, the scheduling model is as follows:

[0016]

[0017] C(R 1 ,j) = C(R 1 ,j - 1) + P 1,j , j = 2, 3..., M

[0018]

[0019] where C(R 1 , 1) represents the departure time of the first aircraft arriving at the first ground station from the first ground station, taking the arrival time of the first aircraft at the first ground station as the 0 moment; represents the waiting time of the first aircraft arriving at the first ground station at the first ground station; C(R i-1 , 1) represents the departure time of the (i - 1)-th aircraft arriving at the first ground station from the first ground station; represents the waiting time of the i-th aircraft arriving at the first ground station at the first ground station; C(R i , 1) represents the departure time of the i-th aircraft arriving at the first ground station from the first ground station; C(R 1 , j) represents the departure time of the first aircraft from the j-th ground station it arrives at; C(R 1 , j - 1) represents the departure time of the first aircraft from the (j - 1)-th ground station it arrives at; P 1,j represents the waiting time of the first aircraft at the j-th ground station it arrives at; C(R i , j) represents the departure time of the i-th aircraft arriving at the j-th ground station from the j-th ground station; C(R i-1 , j) represents the departure time of the (i - 1)-th aircraft arriving at the j-th ground station from the j-th ground station; C(R i , j - 1) represents the departure time of the i-th aircraft arriving at the j-th ground station from the previous ground station it arrives at; represents the waiting time of the i-th aircraft arriving at the j-th ground station at the j-th ground station.

[0020] Furthermore, the objective function of the scheduling plan is as follows:

[0021]

[0022] Among them, R space represents the set of all scheduling plans; R * represents the optimal scheduling plan; C(R * , M) represents the moment when the last aircraft leaves the last ground station after the delivery is completed under the optimal scheduling plan; C(R, M) represents the moment when the last aircraft leaves the last ground station after the delivery is completed under the scheduling plan R.

[0023] Furthermore, in Step 4, the neural network sorting algorithm is used to solve the objective function established in Step 3.

[0024] Furthermore, the specific process of Step 4 is as follows:

[0025] Step 4-1: Initialize the iteration number p = 1;

[0026] Step 4-2: Input the delivery order of each aircraft to each ground station into the neural network. Denote the delivery order of the nth aircraft to each ground station as R n , n = 1, 2, … N;

[0027] For any ground station, use the neural network to output the access order of each aircraft at this ground station;

[0028] Step 4-3: Determine the ordered array and conflict array in the neural network output result, and add the ordered array to the ordered array set;

[0029] Step 4-4: Judge whether the dimension of the conflict array is less than the set threshold or the iteration number is greater than the set threshold;

[0030] If satisfied, merge the ordered array set and the conflict array, and adjust the merged result manually to obtain the final access order of each aircraft to each ground station, that is, obtain the final delivery scheduling plan result;

[0031] If not satisfied, continue to execute Step 4-5;

[0032] Step 4-5: Let the iteration number p = p + 1, remove the delivery scheduling tasks corresponding to the ordered array set from the total delivery scheduling tasks, use the remaining delivery scheduling tasks as the input of the neural network, and return to execute Step 4-3.

[0033] Even further, the determination of the ordered array and conflict array in the neural network output result is specifically:

[0034] Step 431: Initialize the time step number k = 1;

[0035] Step 432: In the neural network output result, determine whether there is a spacecraft access conflict within the k-th time step, that is, determine whether there is a situation where different spacecraft access the same ground station within the k-th time step;

[0036] If not, the access order corresponding to the k-th time step in the neural network output result is an ordered array, and continue to execute Step 433;

[0037] If there is, the access order corresponding to the k-th time step in the neural network output result is a conflict array, and it is determined that the access orders in the subsequent time steps in the neural network output result are all conflict arrays, and end;

[0038] Step 433: Determine whether the time step number k reaches the maximum;

[0039] If the time step number k reaches the maximum, then end;

[0040] If the time step number k does not reach the maximum, let k = k + 1, and return to execute Step 432.

[0041] A computer storage medium, in which at least one instruction is stored, and the at least one instruction is loaded and executed by a processor to implement the above-mentioned multi-aircraft multi-site delivery scheduling planning method.

[0042] A multi-aircraft multi-site delivery scheduling planning device, the device includes a processor and a memory, and at least one instruction is stored in the memory, and the at least one instruction is loaded and executed by the processor to implement the above-mentioned multi-aircraft multi-site delivery scheduling planning method.

[0043] The beneficial effects of the present invention are:

[0044] The present invention considers the scheduling planning problem of multiple aircraft arriving at multiple sites successively, considers processes such as unloading, loading, refueling, and maintenance, optimizes the access order of aircraft to each site, establishes a multi-aircraft multi-site delivery scheduling model, and proposes a discrete data iterative sorting algorithm based on a neural network for the problem of complex sorting of a large number of aircraft. The sorting problem of high-dimensional input is solved through layer-by-layer decomposition and iteration, the goal of the shortest total time for material delivery is achieved, and a better delivery scheduling plan is obtained. Description of the Drawings

[0045] Figure 1 It is a multi-aircraft multi-site delivery scheduling model diagram;

[0046] Figure 2 It is a flowchart of the neural network discrete data iterative sorting algorithm. Detailed implementation manners

[0047] Detailed implementation manner 1: Combine Figure 1 to describe this implementation manner. A multi-aircraft multi-site delivery scheduling and planning method described in this implementation manner specifically includes the following steps:

[0048] Step 1: Denote the number of aircraft as N and the number of ground sites as M, and the waiting time consumed by each aircraft at each ground site is the same (the waiting time consumed by the aircraft at each ground site is independent of the arrival order and the waiting time consumed by the aircraft at each ground site does not stop midway. For the same ground site, the waiting time consumed by each aircraft at this ground site is the same. For the same aircraft, the waiting time consumed by this aircraft at each ground site is the same);

[0049] Step 2: Define constraint (1), constraint (2), and constraint (3):

[0050] Constraint (1): An aircraft can arrive at a ground site at most once;

[0051] Constraint (2): At any moment, only one aircraft is accepted at a ground site;

[0052] Constraint (3): An aircraft arrives at each ground site in sequence according to the order of its delivery to the ground sites;

[0053] Establish a scheduling model according to the defined constraint (1), constraint (2), and constraint (3);

[0054] Step 3: Establish an objective function for the scheduling plan according to the scheduling model;

[0055] Step 4: Solve the objective function established in Step 3 to obtain the scheduling plan result

[0056]

[0057] The main features of the present invention are as follows:

[0058] (1) Multi-aircraft multi-site delivery scheduling model

[0059] Consider the scenario where a large number of aircraft deliver supplies to multiple ground targets respectively. Considering time factors such as loading and unloading time, refueling time, and maintenance time, establish a mathematical model to optimize the arrival order of each aircraft at each target, and form an optimization model with the shortest total time consumed for transporting goods.

[0060] (2) Neural network discrete data iterative sorting algorithm

[0061] The scheduling plan involves dozens of aircraft and multiple targets. The entire problem space is very large, and there will be a very large number of existing scheduling plans, making it difficult to solve directly. Although neural networks have strong data processing capabilities, there is still a possibility of incorrect sorting when faced with such a large-scale problem. Therefore, the present invention proposes a neural network iterative sorting method to complete the sorting task through multiple rounds of iteration.

[0062] Specific Embodiment 2: The difference between this embodiment and Specific Embodiment 1 is that the waiting time consumed by the aircraft at the ground station includes loading and unloading time, refueling time, and maintenance time.

[0063] Other steps and parameters are the same as those in Specific Embodiment 1.

[0064] Specific Embodiment 3: The difference between this embodiment and Specific Embodiment 1 or 2 is that the scheduling model is:

[0065]

[0066] C(R 1 ,j) = C(R 1 ,j - 1) + P 1,j , j = 2, 3..., M

[0067]

[0068] Among them, C(R 1 , 1) represents the moment when the first aircraft arriving at the first ground station leaves the first ground station, and the moment when the first aircraft arrives at the first ground station is taken as the 0 moment; represents the waiting time consumed by the first aircraft arriving at the first ground station at the first ground station; C(R i-1 , 1) represents the moment when the (i - 1)-th aircraft arriving at the first ground station leaves the first ground station; represents the waiting time consumed by the i-th aircraft arriving at the first ground station at the first ground station; C(R i , 1) represents the moment when the i-th aircraft arriving at the first ground station leaves the first ground station; C(R 1 , j) represents the moment when the first aircraft leaves the j-th ground station it arrives at; C(R 1 , j - 1) represents the moment when the first aircraft leaves the (j - 1)-th ground station it arrives at; P 1,j represents the waiting time consumed by the first aircraft at the j-th ground station it arrives at; C(R i , j) represents the moment when the i-th aircraft arriving at the j-th ground station leaves the j-th ground station; C(R i-1, j) represents the (i - 1)-th aircraft arriving at the j-th ground station and the departure time from the j-th ground station; C(R i , j - 1) represents the i-th aircraft arriving at the j-th ground station and the departure time from the previous ground station it arrived at; represents the waiting time consumed by the i-th aircraft arriving at the j-th ground station at the j-th ground station.

[0069] Other steps and parameters are the same as those in the first or second specific implementation manner.

[0070] Specific implementation manner four: The difference between this implementation manner and one of the first to third specific implementation manners is that the objective function of the scheduling plan is:

[0071]

[0072] Among them, R space represents the set of all scheduling plans; R * represents the optimal scheduling plan; C(R * , M) represents the time when the last aircraft departs from the last ground station after the delivery of the last aircraft is completed under the optimal scheduling plan; C(R, M) represents the time when the last aircraft departs from the last ground station after the delivery of the last aircraft is completed under the scheduling plan R.

[0073] Other steps and parameters are the same as those in one of the first to third specific implementation manners.

[0074] Specific implementation manner five: The difference between this implementation manner and one of the first to fourth specific implementation manners is that in step four, the neural network sorting algorithm is used to solve the objective function established in step three.

[0075] Other steps and parameters are the same as those in one of the first to fourth specific implementation manners.

[0076] Specific implementation manner six: Combine Figure 2 to illustrate this implementation manner. The difference between this implementation manner and one of the first to fifth specific implementation manners is that the specific process of step four is as follows:

[0077] Step four one: Initialize the iteration number p = 1;

[0078] Step four two: Input the delivery order of each aircraft to each ground station into the neural network (the neural network adopted in the present invention can be a DNN network, but is not limited to the DNN network), and record the delivery order of the n-th aircraft to each ground station as R n , n = 1, 2,... N;

[0079] For each aircraft, the order in which the aircraft accesses the ground stations is fixed and unchanged. For any ground station, a neural network is used to output the order in which each aircraft accesses the ground station;

[0080] Step 43: Determine the ordered array and the conflict array in the neural network output result, and add the ordered array to the ordered array set;

[0081] Step 44: Determine whether the dimension of the conflict array is less than the set threshold or the number of iterations is greater than the set threshold;

[0082] If satisfied, merge the ordered array set and the conflict array, and manually adjust the merged result to obtain the final access order of each aircraft to each ground station, that is, obtain the final delivery scheduling plan result;

[0083] If not satisfied, continue to execute Step 45;

[0084] Step 45: Let the number of iterations p = p + 1, remove the delivery scheduling tasks corresponding to the ordered array set from the total delivery scheduling tasks, and use the remaining delivery scheduling tasks (based on the ordered arrays that have been determined so far, we can initially obtain the access order of some spacecraft to some ground stations, and then obtain which aircraft still need to access which ground stations, that is, the remaining tasks) as the input of the neural network, and return to execute Step 43.

[0085] Other steps and parameters are the same as those in any one of the specific embodiments 1 to 5.

[0086] The dimension of the sorting R = [R 1 , R 2 …, R N or the number of aircraft is the dimension of the optimization variable. Obviously, if the scheduling plan involves dozens of aircraft and multiple targets, there will be (N!) full permutations, the entire problem space will be very large, and there may be a large number of possible scheduling plans, making it difficult to directly solve. For example, swarm intelligence algorithms such as particle swarm optimization, genetic algorithms, and fish swarm algorithms are often used to solve multi-parameter optimization problems, but they will be very inefficient when facing high-dimensional optimization variables.

[0087] Use the neural network sorting algorithm to obtain the sorting R = [R 1 , R 2 …, R N, from a mathematical perspective, sorting is a process of mapping data elements from an unordered state to an ordered state. Ideally, a neural network should sort a large amount of input data at once, which requires huge training resources and may result in the phenomenon that two different input data are mapped to the same output, thus affecting the correctness of neural network sorting. To this end, the present invention proposes an iterative sorting method based on a neural network. Instead of directly training a complex model to sort all input data at once, the method of the present invention uses a simpler model to complete the sorting task through multiple rounds of iteration.

[0088] Specific Embodiment 7: The difference between this embodiment and any one of Embodiments 1 to 6 is that the determination of the ordered array and the conflict array in the output result of the neural network is specifically as follows:

[0089] Step 431: Initialize the time step number k = 1 (the length of each time step is equal to the waiting time of an aircraft at a ground station);

[0090] Step 432: In the output result of the neural network, determine whether there is a spacecraft access conflict within the k-th time step, that is, determine whether there is a situation where different spacecraft access the same ground station within the k-th time step;

[0091] If not, the access order corresponding to the k-th time step in the output result of the neural network is the ordered array, and continue to execute Step 433;

[0092] If so, the access order corresponding to the k-th time step in the output result of the neural network is the conflict array, and it is determined that the access orders in the subsequent time steps in the output result of the neural network are all conflict arrays, and end;

[0093] Step 433: Determine whether the time step number k reaches the maximum;

[0094] If the time step number k reaches the maximum, end;

[0095] If the time step number k does not reach the maximum, let k = k + 1, and return to execute Step 432.

[0096] Other steps and parameters are the same as any one of Embodiments 1 to 6.

[0097] The present invention optimizes the access order of all aircraft to the same ground station based on the fact that the access order of each aircraft to each ground station remains unchanged, and obtains the final delivery scheduling plan result. It should be noted that the dimension of the conflict array in the present invention refers to the total number of time steps corresponding to the conflict array in the output result of the neural network.

[0098] Embodiment 8: A computer storage medium according to this embodiment, in which at least one instruction is stored, and the at least one instruction is loaded and executed by a processor to implement the multi-aircraft multi-site delivery scheduling planning method described above.

[0099] It should be understood that the instruction includes a computer program product, software, or computerized method corresponding to any method described in the present invention; the instruction can be used to program a computer system or other electronic devices. The computer storage medium may include a readable medium on which the instruction is stored, and may include, but is not limited to, a magnetic storage medium and an optical storage medium; the magneto-optical storage medium includes a read-only memory, a random access memory, an erasable programmable memory such as, and as well as a flash memory layer, or other types of media suitable for storing electronic instructions.

[0100] Embodiment 9: A multi-aircraft multi-site delivery scheduling planning device according to this embodiment, the device includes a processor and a memory. It should be understood that it includes any device including a processor and a memory described in the present invention, and the device may further include other units and modules for display, interaction, processing, control, etc. through signals or instructions, as well as other functions;

[0101] At least one instruction is stored in the memory, and the at least one instruction is loaded and executed by the processor to implement the multi-aircraft multi-site delivery scheduling planning method described above.

[0102] The above calculation examples of the present invention are only to illustrate in detail the calculation model and calculation process of the present invention, rather than to limit the embodiments of the present invention. For those of ordinary skill in the art, other different forms of changes or variations can be made based on the above description. It is impossible to list all the embodiments here. Any obvious changes or variations derived from the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A multi-aircraft multi-site delivery scheduling planning method, characterized in that: The method specifically comprises the following steps: Step 1: The number of aircraft is recorded as N, the number of ground stations is recorded as M, and the waiting time spent by each aircraft at each ground station is the same; Step 2: Define constraints (1), (2), and (3): Constraint (1) An aircraft can arrive at a ground station at most once; Constraint (2): a ground station can only accommodate one aircraft at any time; Constraint (3) An aircraft arrives at each ground station in the order in which it delivers to the ground station; Establish a scheduling model based on the defined constraints (1), (2) and (3); Step 3: Establish the objective function of scheduling planning according to the scheduling model; Step 4: Solve the objective function established in step 3 to obtain the scheduling planning result.

2. The multi-aircraft multi-site delivery scheduling planning method according to claim 1 is characterized in that: The waiting time spent by the aircraft at the ground station includes loading and unloading time, refueling time and maintenance time.

3. The multi-aircraft multi-site delivery scheduling planning method according to claim 2 is characterized in that: The scheduling model is: C(R1,j)=C(R1,j-1)+P 1,j ,j=2,3...,M Wherein, C(R1,1) represents the time when the first aircraft arriving at the first ground station leaves the first ground station, and the time when the first aircraft arrives at the first ground station is taken as time 0; represents the waiting time spent by the first aircraft arriving at the first ground station at the first ground station; C(R i-1 ,1) represents the time when the i-1th aircraft arriving at the first ground station leaves the first ground station; represents the waiting time spent by the i-th aircraft arriving at the first ground station at the first ground station; C(R i ,1) represents the time when the i-th aircraft that arrives at the first ground station leaves the first ground station; C(R1,j) represents the time when the first aircraft leaves the j-th ground station it arrives at; C(R1,j-1) represents the time when the first aircraft leaves the j-1-th ground station it arrives at; P 1,j represents the waiting time spent by the first aircraft at the jth ground station it arrives at; C(R i ,j) represents the time when the i-th aircraft arrives at the j-th ground station and leaves the j-th ground station; C(R i-1 ,j) represents the time when the i-1th aircraft arriving at the jth ground station leaves the jth ground station; C(R i ,j-1) represents the time when the i-th aircraft arriving at the j-th ground station leaves the last ground station it arrived at; represents the waiting time spent by the i-th aircraft arriving at the j-th ground station at the j-th ground station.

4. The multi-aircraft multi-site delivery scheduling planning method according to claim 3 is characterized in that: The objective function of the scheduling plan is: Among them, R space Represents the set of all scheduling schemes; R * represents the optimal scheduling solution; C(R * ,M) represents the time when the last aircraft leaves the last ground station after completing the delivery under the optimal scheduling plan; C(R,M) represents the time when the last aircraft leaves the last ground station after completing the delivery under the scheduling plan R.

5. The multi-aircraft multi-site delivery scheduling planning method according to claim 4 is characterized in that: In step 4, a neural network sorting algorithm is used to solve the objective function established in step 3.

6. A multi-aircraft multi-site delivery scheduling planning method according to claim 5, characterized in that: The specific process of step 4 is as follows: Step 41: Initialize the number of iterations p = 1; Step 42: Input the order in which each aircraft delivers to each ground station into the neural network, and record the order in which the nth aircraft delivers to each ground station as R n , n=1,2,…N; For any ground station, the neural network is used to output the visit order of each aircraft on the ground station; Step 43: Determine the ordered array and conflict array in the output result of the neural network, and add the ordered array to the ordered array set; Step 44: determine whether the conflict array dimension is less than a set threshold or the number of iterations is greater than a set threshold; If satisfied, the ordered array set and the conflict array are merged, and the merged result is manually adjusted to obtain the final access sequence of each aircraft to each ground station, that is, the final delivery scheduling planning result; If not satisfied, proceed to step 4 and 5; Step 45: Let the number of iterations p = p + 1, remove the delivery scheduling tasks corresponding to the ordered array set from the total delivery scheduling tasks, use the remaining delivery scheduling tasks as the input of the neural network, and return to execute step 43.

7. The multi-aircraft multi-site delivery scheduling planning method according to claim 6 is characterized in that: The method of determining the ordered array and the conflict array in the output result of the neural network is specifically: Step 431: Initialize the time step number k=1; Step 432: In the output result of the neural network, determine whether there is a spacecraft access conflict in the kth time step, that is, determine whether there are different spacecraft accessing the same ground station in the kth time step; If it does not exist, the access order corresponding to the kth time step in the output result of the neural network is an ordered array, and step 433 is continued; If it exists, the access sequence corresponding to the kth time step in the output result of the neural network is a conflict array, and the access sequence in the subsequent time steps in the output result of the neural network is determined to be a conflict array, and the process ends; Step 433: Determine whether the number of time steps k reaches the maximum; If the number of time steps k reaches the maximum, it ends; If the time step number k has not reached the maximum, set k=k+1 and return to execute step 432.

8. A computer storage medium, characterized in that: The storage medium stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement a multi-aircraft multi-site delivery scheduling planning method as described in any one of claims 1 to 7.

9. A multi-aircraft multi-site delivery scheduling planning device, characterized in that: The device includes a processor and a memory, wherein the memory stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement a multi-aircraft multi-site delivery scheduling planning method as described in any one of claims 1 to 7.

Images