CN116384606A

CN116384606A - Scheduling optimization method and system based on cooperative distribution of vehicle unmanned aerial vehicle

Info

Publication number: CN116384606A
Application number: CN202310398180.2A
Authority: CN
Inventors: 赵金秋; 谢秉磊; 刘永武; 孙宇; 张笑桐
Original assignee: Shenzhen Graduate School Harbin Institute of Technology
Current assignee: Shenzhen Graduate School Harbin Institute of Technology
Priority date: 2023-04-04
Filing date: 2023-04-04
Publication date: 2023-07-04

Abstract

The invention discloses a scheduling optimization method and a scheduling optimization system based on cooperative distribution of a vehicle unmanned aerial vehicle, wherein a mixed integer programming model is firstly constructed, and then a complex vehicle unmanned aerial vehicle cooperative problem is divided into two problems by a logic-based Benders decomposition method: the method comprises the steps of solving a vehicle path planning problem and an unmanned aerial vehicle take-off and landing point selection problem alternately, generating a Benders cut dynamically to reduce an algorithm search space, solving a target value of a vehicle unmanned aerial vehicle cooperative problem iteratively, obtaining accurate path planning and improving the efficiency of cooperative distribution.

Description

Scheduling optimization method and system based on cooperative distribution of vehicle unmanned aerial vehicle

Technical Field

The invention relates to the technical field of logistics distribution and scheduling, in particular to a scheduling optimization method and system based on collaborative distribution of a vehicle unmanned plane.

Background

As unmanned aerial vehicle technology continues to mature, unmanned aerial vehicles have been widely used in a variety of scenarios (e.g., communication delays, surveillance, and urban logistics). However, the unmanned aerial vehicle has lower endurance mileage and bearing capacity, and when logistics distribution is carried out, the unmanned aerial vehicle is required to be distributed with different transport means in a coordinated manner, for example, the unmanned aerial vehicle is distributed with vehicles such as trucks and vans in a coordinated manner, and terminal distribution is realized by the unmanned aerial vehicle.

Although the conventional path planning algorithm can respectively plan the path of the unmanned aerial vehicle under the condition of independent delivery and the path of the vehicle under the condition of independent delivery, the limitations and constraints of cooperative delivery are not considered, so that the path planning is inaccurate and the efficiency of cooperative delivery is lower.

Disclosure of Invention

The invention mainly aims to provide a scheduling optimization method, a scheduling optimization system, an intelligent terminal and a computer readable storage medium based on collaborative distribution of a vehicle unmanned aerial vehicle, and aims to solve the problems of inaccurate path planning and lower collaborative distribution efficiency in collaborative distribution in the prior art.

In order to achieve the above object, a first aspect of the present invention provides a scheduling optimization method based on collaborative distribution of a vehicle unmanned aerial vehicle, including:

constructing a mixed integer programming model based on constraint conditions of cooperative distribution of the unmanned aerial vehicle;

decomposing the mixed integer programming model into a vehicle path programming problem and an unmanned aerial vehicle take-off and landing point selection problem according to a logic-based Benders decomposition method;

solving the vehicle path planning problem to obtain an optimal solution of the vehicle path planning problem;

solving the unmanned aerial vehicle take-off and landing point selection problem based on an optimal solution of the vehicle path planning problem, and generating a Benders cut;

And adding the nodes cut into the vehicle path planning problem, and returning to solve the vehicle path planning problem to perform iterative solution until a preset condition is met, so as to obtain a dispatching optimization result.

Optionally, the vehicle distribution path is set as a coupling variable between the vehicle path planning problem and the unmanned aerial vehicle take-off and landing point selection problem to update the vehicle path planning problem and the unmanned aerial vehicle take-off and landing point selection problem, the vehicle path planning problem is modeled as a column generation model, the vehicle path planning problem is solved by adopting a column generation algorithm, and the optimal solution of the vehicle path planning problem is a route set.

Optionally, the mixed integer programming model is further decomposed into an original vehicle path programming problem and an original unmanned aerial vehicle take-off and landing point selection problem, where the coupling variables are not the vehicle delivery paths, and the generating the nodes includes:

when the unmanned aerial vehicle take-off and landing point selection problem is not feasible, verifying the feasibility of the original unmanned aerial vehicle take-off and landing point selection problem by adopting a Benders solver to obtain a verification result;

and generating the Benders cut according to the verification result.

Optionally, the method further comprises:

when the number of iterative solutions is increased by a preset threshold value and the unmanned aerial vehicle take-off and landing point selection problem has an optimal solution, solving the original unmanned aerial vehicle take-off and landing point selection problem to generate the Benders cut.

Optionally, the solving the vehicle path planning problem by adopting a column generation algorithm includes:

generating an initial feasible solution according to a genetic algorithm;

constructing a constraint linear main problem by taking the initial feasible solution as an initial column;

solving a constraint linear main problem to obtain a dual vector;

constructing a first-class shortest path problem with resource constraint according to a pricing problem based on the dual vector;

solving the first-class shortest path problem with resource constraint to generate a new column;

and adding the new column into the constraint linear main problem, and returning to solve the constraint linear main problem to obtain a dual vector for iterative solution until the new column cannot be obtained and outputting the solution of the constraint linear main problem.

Optionally, the solving the elementary shortest path problem with resource constraint includes:

when the heuristic algorithm can find a column with negative test number, adopting the heuristic algorithm which is the same as the initial feasible solution to screen the column with negative test number so as to solve the linear-limiting main problem; otherwise, solving the linear-limiting main problem by adopting a labeling algorithm.

Optionally, the dual vector is a smooth dual vector, the dual vector of the previous iteration and the dual vector of the current iteration are weighted, the weighted dual vector is obtained, and the dual vector of the current iteration is updated.

The second aspect of the present invention provides a scheduling optimization system based on collaborative delivery of a vehicle unmanned aerial vehicle, wherein the system comprises:

the mixed integer programming model construction module is used for constructing a mixed integer programming model based on constraint conditions of cooperative distribution of the unmanned aerial vehicle;

the decomposition module is used for decomposing the mixed integer programming model into a vehicle path programming problem and an unmanned aerial vehicle take-off and landing point selection problem according to a logic-based Benders decomposition method;

the solving module is used for solving the vehicle path planning problem and obtaining an optimal solution of the vehicle path planning problem;

the nodes cutting module is used for solving the unmanned aerial vehicle take-off and landing point selection problem based on the optimal solution of the vehicle path planning problem to generate nodes cutting;

and the iteration module is used for adding the nodes cut into the vehicle path planning problem, and returning to solve the vehicle path planning problem to perform iteration solution until a preset condition is met, so as to obtain a dispatching optimization result.

The third aspect of the present invention provides an intelligent terminal, where the intelligent terminal includes a memory, a processor, and a scheduling optimization program based on coordinated delivery of vehicle unmanned aerial vehicles, where the scheduling optimization program based on coordinated delivery of vehicle unmanned aerial vehicles is stored in the memory and is executable on the processor, and when executed by the processor, implements any one of the steps of the scheduling optimization method based on coordinated delivery of vehicle unmanned aerial vehicles.

A fourth aspect of the present invention provides a computer readable storage medium, where a scheduling optimization program based on coordinated delivery of a vehicle unmanned aerial vehicle is stored in the computer readable storage medium, and the scheduling optimization program based on coordinated delivery of a vehicle unmanned aerial vehicle implements any one of the steps of the scheduling optimization method based on coordinated delivery of a vehicle unmanned aerial vehicle when executed by a processor.

From the above, the invention firstly builds a mixed integer programming model, and then separates the complex unmanned aerial vehicle coordination problem into two problems by a logic-based Benders decomposition method: the method comprises the steps of solving a vehicle path planning problem and an unmanned aerial vehicle take-off and landing point selection problem alternately, generating a Benders cut dynamically to reduce an algorithm search space, solving a target value of a vehicle unmanned aerial vehicle cooperative problem iteratively, obtaining accurate path planning and improving the efficiency of cooperative distribution.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic flow chart of a scheduling optimization method based on collaborative distribution of a vehicle unmanned aerial vehicle, which is provided by the embodiment of the invention;

FIG. 2 is a schematic diagram of a collaborative distribution scenario in the embodiment of FIG. 1;

FIG. 3 is a schematic flow chart of the generation of Benders cuts;

FIG. 4 is a flow chart diagram of solving a vehicle path planning problem using a column generation algorithm;

FIG. 5 is a graph of the results data of the embodiment of FIG. 1 applied to a small scale collaborative distribution problem;

FIG. 6 is a graph of the resulting data of the embodiment of FIG. 1 applied to a mid-scale collaborative distribution problem;

FIG. 7 is a graph of the resulting data of the embodiment of FIG. 1 applied to a large scale collaborative distribution problem;

fig. 8 is a schematic structural diagram of a scheduling optimization system based on collaborative distribution of a vehicle unmanned aerial vehicle according to an embodiment of the present invention;

fig. 9 is a schematic block diagram of an internal structure of an intelligent terminal according to an embodiment of the present invention.

Detailed Description

In the following description, for purposes of explanation and not limitation, specific details are set forth such as the particular system architecture, techniques, etc., in order to provide a thorough understanding of the embodiments of the present invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

It should be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

It is also to be understood that the terminology used in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.

As used in this specification and the appended claims, the term "if" may be interpreted in context as "when …" or "upon" or "in response to a determination" or "in response to detection. Similarly, the phrase "if a condition or event described is determined" or "if a condition or event described is detected" may be interpreted in the context of meaning "upon determination" or "in response to determination" or "upon detection of a condition or event described" or "in response to detection of a condition or event described".

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

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways other than those described herein, and persons skilled in the art will readily appreciate that the present invention is not limited to the specific embodiments disclosed below.

Unmanned aerial vehicles have received a great deal of attention because of their low energy consumption, low cost, high accessibility, etc., and unmanned aerial vehicles have been widely used in various scenarios (e.g., communication delay, surveillance, and city logistics). However, the endurance mileage and the bearing capacity of the unmanned aerial vehicle are low, so that the unmanned aerial vehicle and different transport means are required to be distributed in a coordinated manner, such as the coordinated distribution of the unmanned aerial vehicle and a truck, the coordinated distribution of the unmanned aerial vehicle and the unmanned aerial vehicle, and the like.

When the path planning of the dynamic vehicle unmanned aerial vehicle collaborative scheduling problem (DTDCSP) is carried out, not only the classical vehicle path problem, but also the unmanned aerial vehicle path and the combination problem thereof need to be considered. Meanwhile, attention is paid to the coupling condition of the truck and the unmanned aerial vehicle route, namely, the unmanned aerial vehicle is required to be received after the truck finishes service at a certain delivery point.

The current path planning algorithm can only plan the paths of the unmanned aerial vehicle under the condition of independent delivery and the paths of the vehicle under the condition of independent delivery respectively, and is inaccurate and low in efficiency when the path planning algorithm is applied to scheduling optimization of collaborative delivery.

Although there is a related study of unmanned aerial vehicle coordination, the main focus is on a wider application environment and computational efficiency, ignoring limitations of coordination strategies, such as: the problems of long waiting time of trucks, fixed unmanned aerial vehicle take-off and landing trucks and the like cannot be solved, and the cooperative efficient scheduling of the unmanned aerial vehicles of the vehicles can not be realized.

The invention provides a scheduling optimization method based on collaborative delivery of a vehicle unmanned aerial vehicle, which is characterized in that a vehicle unmanned aerial vehicle collaborative delivery problem is decomposed into a vehicle path planning problem and an unmanned aerial vehicle take-off and landing point selection problem according to a logic-based Benders decomposition method, the vehicle path planning problem and the unmanned aerial vehicle take-off and landing point selection problem are solved alternately, the vehicle path planning problem is optimized according to a generated Benders cut, then a solving process is carried out, and a scheduling optimization result of collaborative delivery is obtained through repeated iterative solution.

Exemplary method

The embodiment of the invention provides a scheduling optimization method based on collaborative distribution of a vehicle unmanned aerial vehicle, which is deployed on electronic equipment such as a computer, a server and the like, and aims at collaborative distribution of a truck unmanned aerial vehicle, wherein an application scene is an urban emergency response occasion. The types of the vehicles are not limited, and various vehicles such as unmanned vehicles, trucks, vans and the like can be used. Specifically, as shown in fig. 1, the method includes the following steps:

step S100: and constructing a mixed integer programming model based on constraint conditions of cooperative distribution of the unmanned aerial vehicle.

Specifically, as shown in fig. 2, the implementation scenario of the collaborative distribution includes a set of rescue delivery points C (hereinafter referred to as demand points for convenience of description), a set of trucks K, a set of unmanned aerial vehicles U, and an integrated warehouse for trucks and unmanned aerial vehicles. The emergency response scenario may be represented as a graph g= (N, a), where N is a set of nodes, including all demand points, a is a set of arcs connecting each node, a= { (i, j) |i+notej }. The time for the truck and the drone to pass one edge of a (j) is denoted d _ij And

for each truck in k.epsilon.K, it leaves the warehouse with n _k And (5) erecting an unmanned aerial vehicle.

In order to facilitate modeling without loss of generality, the following constraints are set: 1. consider only one truck and one drone type; 2. the unmanned aerial vehicle can complete charging and material supplement in the truck running process, and extra time is not required to be consumed; the service time of trucks and unmanned aerial vehicles at the demand points is known; 3. the unmanned aerial vehicle flies according to the straight line distance, and the flying distance is calculated according to the Euclidean distance; the truck is required to travel along a road network, and the travel distance is calculated according to the Manhattan distance; 4. each demand point is only accessed once by a truck or drone; 5. the take-off and landing of the unmanned aerial vehicle can occur at any truck access demand point (without the take-off and landing point), but the same demand point only takes off or lands the unmanned aerial vehicle once at most; 6. the cargo capacity of the truck is enough to meet the total requirement of distribution points on any route, and the unmanned aerial vehicle can only access one requirement point every time of flight; 7. the number of unmanned aerial vehicles that can be parked on the truck is limited, and the number of unmanned aerial vehicles on the truck at any one time cannot exceed the limit.

On the premise of meeting the constraint conditions, the response time delay is minimized to be an optimization target, meanwhile, the real constraints such as the truck paths, the number of unmanned aerial vehicles and the endurance mileage are considered, and a mixed integer programming model of the truck unmanned aerial vehicle collaborative original problem (OP for short) is built.

For convenience of description and understanding, the following will first describe the variables used in the present model:

k is the set of all trucks; u is a set of all unmanned aerial vehicles; c is a set of all delivery points; d, d _i,j The travel time of the truck from delivery point i to j;

the flight time from delivery point i to j for the unmanned aerial vehicle; t (T) _i The latest service time for the ith delivery point; />

For the service of the ith delivery pointA compartment; q _i The number of unmanned aerial vehicles for the ith delivery point; m is an arbitrarily large positive number; l is the endurance mileage of the unmanned aerial vehicle; n (N) ₃ The maximum number of unmanned aerial vehicles allowed to be carried on the truck; alpha is an adjustable coefficient in the objective function; c is the number of delivery points; />

Is a boolean variable indicating whether the ith delivery point is serviced by truck k; />

As boolean variables, indicating whether the ith delivery point is served by the drone u; />

Is a boolean variable indicating whether truck k moves from delivery point i to j; / >

A Boolean variable which represents whether the unmanned plane u flies from the delivery point i to j; t is t _i The time for the truck or unmanned aerial vehicle to reach the ith delivery point; l (L) _i For the time the truck or drone left the ith delivery point.

The specific expression of the mixed integer programming model is as follows:

the constraint conditions are as follows:

the constraint conditions (2) - (3) ensure balance of the truck flow, and the constraint conditions (4) - (5) ensure normal operation of the unmanned aerial vehicle flow. In particular, constraint (2) indicates that the number of trucks leaving the yard should not exceed the total number of available trucks. Constraint (3) indicates that for each node, the number of trucks that leave must be equal to the number of vehicles that reach that node. Constraint conditions (4) - (6) ensure that each rescue requirement node is only accessed by one unmanned aerial vehicle. Constraints (7) - (8) ensure that each demand is only accessed once by a truck or drone. In addition, constraints (9) - (10) establish a relationship between travel and arrival time, and constraint (11) ensures that each truck/drone can leave only after the delivery point is serviced. The constraint (12) specifies that the departure time of the drone cannot be earlier than the arrival time. Constraints (13) - (16) impose a limit on the number of drones carried per truck. Specifically, constraint (13) specifies that the number of drones carried by each truck at each point is greater than zero only when the truck serves that point. The constraint (14) ensures that the number of unmanned aerial vehicles carried by each truck does not exceed its unmanned aerial vehicle capacity. Constraints (15) - (16) provide a formula for calculating the number of drones per truck. Finally, the constraint condition (17) limits the endurance mileage of each unmanned aerial vehicle.

Step S200: and decomposing the mixed integer programming model into a vehicle path programming problem and an unmanned aerial vehicle take-off and landing point selection problem according to a logic-based Benders decomposition method.

In particular, benders decomposition is used to address complex variable problems. The Benders decomposition decomposes the original problem into a main problem and a series of sub-problems. Where the main problem is relaxation of the original problem, new variables are typically introduced to approximate the original objective function. After a solution of the main problem is obtained, solving the sub-problem on the basis of the solution, and generating Benders cuts to be added into the main problem. And performing multiple iterations until the stopping condition is met and obtaining an optimal solution. However, classical Benders decomposition methods typically require that the sub-problem be linear programming, by which the formula of the Benders cut is derived by the dual of the linear programming. Although the method can conveniently and efficiently generate the Benders cut, the application range of the Benders decomposition frame is limited due to the linear programming requirement of the subproblems, and the method is not suitable for the application scene of the embodiment.

The logic-based Benders decomposition method employed in this example is a variant of the classical Benders decomposition method. In principle, logic-based Benders decomposition methods allow the sub-problem to be any optimization problem, not a specific linear or nonlinear programming problem. When some decision variables are fixed, it can deal with a wide variety of large scale decoupling or simplification issues. In general, the logic-based Benders decomposition approach extends the underlying Benders decomposition policy to the case where the sub-problem is an arbitrary optimization problem. The Benders cut is obtained by solving the inference pair of the sub-problem, which can be simplified to be the pair of the linear programming when the sub-problem is linear.

When logic-based Benders decomposition is carried out, firstly, demand points in the problem of cooperative scheduling of the truck unmanned aerial vehicle are divided into two types of truck access points and unmanned aerial vehicle access points. When the type of the demand point is determined, the path planning of the truck is similar to the classical vehicle path planning problem, and the path planning of the unmanned aerial vehicle can be simplified to the selection of the take-off and landing point. Based on this teaching, the truck unmanned cooperative problem can be decomposed into a vehicle path planning problem (i.e., a main problem) accessing part of clients and an unmanned take-off and landing point selection problem (i.e., a sub-problem).

According to the fifth constraint condition set in step S100: the landing of the unmanned aerial vehicle may occur at any truck access point (including no landing point), but the same point is only one landing or take-off of the unmanned aerial vehicle at most. At most, c demand points exist in the truck unmanned aerial vehicle cooperative system

The individual demand points are accessible by the drone. In solving the vehicle path planning problem, only the truck path needs to be considered, and at least +.>

The individual demand points serve to minimize response time delays. The decision variable of the vehicle path planning problem is +.>

And->

The vehicle path planning problem can be expressed as:

The constraint conditions are as follows:

the above constraints (19) - (20) are flow balance constraints; constraint conditions (21) - (22) are time constraints to ensure timely delivery; constraint (23) is used to force the minimum necessary service for the truck

A plurality of demand points; constraints (24) are used to ensure that the variable values are consistent.

And then, on the basis of determining which demand points are accessed by the truck, distributing the demand points which are not accessed by the truck to the unmanned aerial vehicle when solving the unmanned aerial vehicle take-off and landing point selection problem. And selecting the landing position of the unmanned aerial vehicle, and re-optimizing the path of the truck.

Step S300: and solving the vehicle path planning problem to obtain an optimal solution of the vehicle path planning problem.

Step S400: and solving the unmanned aerial vehicle take-off and landing point selection problem based on the optimal solution of the vehicle path planning problem, and generating Benders cuts.

Specifically, if the vehicle path planning problem and the unmanned aerial vehicle take-off and landing point selection problem are solved separately, because the vehicle path planning problem does not consider constraints related to the cooperation of the truck unmanned aerial vehicle and the endurance capability of the unmanned aerial vehicle, the result of selecting the delivery points may be inappropriate, so that the waiting time of the truck and the unmanned aerial vehicle is too long, a poor solution is generated, and even some delivery points cannot be accessed. Therefore, the invention uses a mixed integer programming solver in a logic-based Benders decomposition method to firstly solve the optimal solution of the vehicle path programming problem, substitutes the solved optimal solution into a mixed integer programming model, reduces the variable of the mixed integer programming model, then solves the unmanned aerial vehicle take-off and landing point selection problem to obtain the solution of the unmanned aerial vehicle take-off and landing point selection problem, constructs a dual vector according to the optimal solution of the vehicle path programming problem and the solution of the unmanned aerial vehicle take-off and landing point selection problem, and generates a Benders cut according to the dual vector.

The cuts generated by the mixed integer programming solver include: feasibility cuts (feasility cuts) and optimums cuts (Optimality cuts).

Since the variables x and y in the mixed integer programming model are connected by constraints, the unmanned take-off and landing point selection problem may not be feasible, i.e., there is no solution, given a certain x. In this case, even

The vehicle path planning problem is feasible, and the optimal solution corresponding to the vehicle unmanned plane collaborative distribution problem is realized ^* (x) must satisfy->

In other words, the feasible area of the vehicle path planning problem is too large at this time, and further restrictions should be made. To correct this problem, a feasibility cut needs to be added to the vehicle path planning problem. About->

The feasibility measure of (2) is +.>

Inequality of form in which

Is a function of the following properties:

after adding a feasibility cut, the vehicle path planning problem can be re-solved and the above steps repeated until a viable solution is obtained. In the worst case, a discard (no-good cut) can be used, i.e.

This allows the currently useless solution to be cut off, but only the only solution is expelled. Ideally, it is desirable to eliminate as much as possible at a time Is not feasible.

In this embodiment, no feasible solution to the unmanned aerial vehicle take-off and landing point selection problem means that: given the truck path y, the demand points that are not accessed by the truck cannot be all accessed by the unmanned aerial vehicle, subject to the limitations of unmanned aerial vehicle flight endurance and number. Suppose C _k (Y _k ) Representing the set of demand points accessed by the truck in the current scheme, the set of demand points not accessed by the truck (i.e. the set of demand points which have to be accessed by the unmanned aerial vehicle) is C _u (Y _k )＝C-C _k (Y _k ) The discard in this embodiment can thus be expressed as:

propositions can be deduced from the formula of the discard fraction: if set up

Then the two discard cuts have the following relationship: c (C) _u ' generated cut dominates C _u The resulting cuts.

Namely: when (when)

When there is a need for

When (when)

When the optimal solution of the vehicle path planning problem and the unmanned aerial vehicle take-off and landing point selection problem have feasible solutions, the physical meaning of the integer planning model shows that the unmanned aerial vehicle take-off and landing point selection problem cannot be solved, and the optimal solution is inevitably present>

At this time, there may be +.>

In (1), wherein->

And->

The optimal target values for the vehicle path planning problem and the unmanned aerial vehicle take-off and landing point selection problem (the same as the optimal target value for the vehicle unmanned aerial vehicle collaborative distribution problem), respectively, mean that the constraints in the current vehicle path planning problem are not tight enough. At this point an optimal cut needs to be added to the vehicle path planning problem. About- >

Is->

Inequality of form, wherein B _y y→R satisfies +.>

If and only if->

The time inequality takes an equal sign. The optimal cut can be expressed as:

namely: if the demand point presses the current C _k ,C _u And if the scheme is classified, the optimal target value z of the cooperative distribution problem of the unmanned aerial vehicle is necessarily larger than or equal to the optimal target value of the current unmanned aerial vehicle take-off and landing point selection problem.

Step S500: adding the Benders cut into the vehicle path planning problem, and returning to solve the vehicle path planning problem to perform iterative solution until the preset condition is met, so as to obtain a dispatching optimization result.

In particular, the intensity of the optimal cut is determined by its other feasible solutionsThe degree of tightness is determined. The optimality conditions for the logic-based Benders algorithm are: is provided with

And->

Is added with the optimal solution of the vehicle path planning problem after some Benders cuts, and the optimal solution of the unmanned plane take-off and landing point selection problem is marked as +.>

At this time, if->

Then->

And the method is also an optimal solution to the problem of cooperative delivery of the unmanned aerial vehicle.

After adding the optimal cut to the vehicle path planning problem, re-solving the vehicle path planning problem if

The optimal solution of the vehicle path planning problem is still obtained, namely, at the moment, the target value of the vehicle unmanned plane cooperative distribution problem is accurately estimated by z, and a dispatching optimization result is obtained; otherwise, solving the unmanned aerial vehicle take-off and landing point selection problem, and generating a new Benders cut. By alternately solving the vehicle path planning problem and the unmanned aerial vehicle take-off and landing point selection problem, generating a Benders cut and adding the Benders cut to the vehicle path planning problem so as to avoid the same solution or worse solution in the subsequent solving process, the target value of the vehicle unmanned aerial vehicle collaborative distribution problem can be obtained after enough iterations, and the scheduling optimization result is obtained.

In summary, by using the logic-based Benders decomposition method, the complex unmanned aerial vehicle coordination problem is divided into two problems, namely, a vehicle path planning problem and an unmanned aerial vehicle take-off and landing point selection problem, and meanwhile, algorithm search space is reduced based on a dynamic generation mechanism and a dominance rule for designing Benders cuts by distribution logic, a target value of the unmanned aerial vehicle coordination problem is solved, accurate path planning is obtained, and the efficiency of coordination distribution is improved.

If it is

When the method is a finite set and a feasible solution exists for the collaborative problem of the unmanned aerial vehicle, the method in the embodiment can be terminated certainly because the enumeration is performed on a finite number of sub-problems, but in other cases, iteration can not be terminated certainly, and the convergence speed of the algorithm depends largely on the intensity of the Benders cut. The model becomes fuzzy and complex due to the complexity of the dynamic truck unmanned aerial vehicle collaborative scheduling problem, which brings great difficulty to the optimizer to solve even small-scale problems in a reasonable time. Because of the large solution space, heuristic algorithms are also challenging to find the optimal solution in a reasonable time.

On the other hand, the solving speed of the vehicle path planning problem and the unmanned aerial vehicle take-off and landing point selection problem also directly affects the solving efficiency of the logic-based Benders method. In the present embodiment, the vehicle path planning problem is a vehicle path planning problem of visiting a part of customers, and the solution complexity of the vehicle path planning problem is equivalent to that of a general vehicle path planning problem

Where n represents the total number of demand points and m represents the number of demand points that need truck access. Meanwhile, as the vehicle path planning problem needs to be solved again after the Benders are added each time, the vehicle path planning problem is generally solved by adopting a branch cutting method, and most of the time of the method is consumed on the same tree. Even though the added Benders cut can somewhat reduce the size of the vehicle path planning problem, evaluating some duplicate nodes in the search tree can still waste a significant amount of time.

Therefore, in order to improve the efficiency of the solution, in another embodiment of the present invention, the vehicle path planning problem is modeled as a column generation model, and only columns in the column generation model need to be checked for deletion and addition after each addition of a Benders cut. Not only can repeated searching and evaluation of the sub-paths of the truck be avoided, but also the problem of vehicle path planning can be avoided to obtain integer solutions with poor quality, and the sub-problems can be favorably generated into high-quality Benders cuts. This method is called the logical beacons solution method based on the column generation model (Logic Benders Decomposition Approach Driven by Column Generation: LBBDCG).

The specific process is as follows: the method comprises the steps of setting a coupling variable as a truck delivery path, decomposing a mixed integer programming model into a vehicle path programming problem and an unmanned aerial vehicle take-off and landing point selection problem, modeling the vehicle path programming problem as a column generation model, and solving an optimal solution of the vehicle path programming problem by adopting a column generation algorithm when solving the vehicle path programming problem to obtain a route set. Accordingly, the calculated expressions of the feasibility and optimal cuts also need to be associated with the route set. Let r= { R _k And the method comprises the steps of (1) obtaining a route set by solving a vehicle path planning problem, and if the unmanned aerial vehicle take-off and landing point selection problem is not solved when the vehicle path planning problem is solved, the feasibility cut can be expressed as:

indicating that the routes in R cannot be selected simultaneously. If the optimal solution f exists in the unmanned aerial vehicle taking-off and landing point selection problem for solving the vehicle path planning problem, the optimal cut can be expressed as:

when all routes in R are selected, the target value z is f.

The main objective of the vehicle path planning problem is to search for the route of the truck, whereas the vehicle of the invention is of single type and route symmetrical, naturally suitable for use in a column generation algorithm. The column generation algorithm not only can avoid repeated searching and evaluation of the vehicle path in the iterative process of the vehicle path planning problem, but also can gradually generate sub-paths by utilizing a plurality of efficient algorithms such as a labeling algorithm, a genetic algorithm and the like.

By changing the vehicle path planning problem into the column generation model, repeated solution of the truck path can be avoided, the unmanned aerial vehicle take-off and landing point selection problem is greatly simplified, and the efficient algorithm of the existing vehicle path planning is easily migrated to the solution vehicle path planning problem.

The Benders cuts obtained by the feasibility cutting calculation formula (3) and the optimal cutting calculation formula (4) are not tight enough although the column generation algorithm is required, the potential number is far greater than that of the Benders cuts generated by the formula feasibility cutting calculation formula (1) and the optimal cutting calculation formula (2), and the iteration times are increased. Considering that it may be difficult to solve the optimal solution for the unmanned aerial vehicle take-off and landing point selection problem, it is relatively easy to check its feasibility with a solver, and therefore in another embodiment, optimization is performed when generating the beacons cut. And other variables which are not the vehicle delivery paths are selected as coupling variables, and the mixed integer programming model is decomposed into an original vehicle path programming problem and an original unmanned aerial vehicle take-off and landing point selection problem. That is, the vehicle path planning problem, the unmanned aerial vehicle take-off and landing point selection problem, the original vehicle path planning problem and the original unmanned aerial vehicle take-off and landing point selection problem of the coupling variable non-vehicle delivery path are solved simultaneously. As shown in fig. 3, the step of generating the Benders cut includes:

Step S410: when the unmanned aerial vehicle take-off and landing point selection problem is not feasible, verifying the feasibility of the original unmanned aerial vehicle take-off and landing point selection problem by adopting a Benders solver to obtain a verification result;

step S420: and generating Benders cuts according to the verification result.

Specifically, when the unmanned aerial vehicle take-off and landing point selection problem is not feasible, the feasibility of the original unmanned aerial vehicle take-off and landing point selection problem is checked, and when the checking result is feasible, the Benders cut of the feasibility cutting formula (1) is added to the vehicle path planning problem.

Optionally, when the number of iterative solutions increases by a preset threshold value and the unmanned aerial vehicle take-off and landing point selection problem has an optimal solution, the unmanned aerial vehicle take-off and landing point selection problem can be solved to generate a Benders cut. After a certain number of iterations at each interval, when the unmanned aerial vehicle take-off and landing point selection problem has an optimal solution, solving the original man-machine take-off and landing point selection problem, and adding the Benders cut of the optimal cutting formula (2) into the vehicle path planning problem.

By optimizing the generation of the Benders cut, the defect that the Benders cut generated by the feasibility cutting formula (3) and the optimal cutting formula (4) is not tight can be overcome.

The column generation algorithm is a very powerful optimization algorithm that solves the large-scale integer programming problem. The basic principle is basically the same as that of the selection of the base variable in the iterative process of the simplex method. In the solving process of the column generation algorithm, the integer programming problem is linearly relaxed to become a linear main problem (Liner Master Problem: LMP). In view of the large scale of variables, only a small part of variables are considered initially, a linear-limiting main problem (Restricted Liner Master Problem: RLMP) is constructed, a dual variable of a linear program is obtained by solving an optimal solution of the linear-limiting main problem, a new variable which can optimize an objective function is searched by constructing a Pricing problem (price problem: PP), the new variable is added to the linear-limiting main problem and solved again, and the iteration is performed until the variable which can optimize the objective function cannot be found, and the optimal solution of the linear-limiting main problem is obtained.

In this embodiment, as shown in fig. 4, a column generation algorithm is used to solve a vehicle path planning problem, including:

step S310: generating an initial feasible solution according to a genetic algorithm;

step S320: taking the initial feasible solution as an initial column, and constructing a limited linear main problem;

specifically, the initial feasible solution is a necessary premise for constructing a set-partitioning model (Set Partition Model of RLMP) that limits the linear main problem, and a series of good initial columns can reduce the number of iterations of the column generation algorithm. The embodiment produces a batch of feasible solutions of the vehicle path planning problem through a Genetic Algorithm (GA), builds a vehicle path planning problem model by Dantzig-Wolfe (DW), constructs a set-split model for all truck tasks as a linear-limiting main problem, and takes the initial feasible solution as an initial column of the linear-limiting main problem.

The expression that limits the linear main problem is: (MP 2) Σ _r∈Ω p _r z _r Representing minimizing the delivery delay and travel distance of the selected truck route.

The constraint conditions are as follows:

/>

wherein a constraint (25) is used to ensure that N is selected ₁ A delivery route; constraints (26) are used to ensure that each demand point belongs to at most one delivery route, where a _ir Indicating whether the demand point i belongs to the distribution route r; constraint (27) is used to ensure that the total number of points of demand for truck access is at least

Constraint (28) is used for the decision variable z _r ,/>

A binary constraint is imposed.

The genetic algorithm realizes heuristic search of complex search space through simplified genetic process, and sub-optimal solution can be found out under high probability finally. In the embodiment, a chromosome of a vehicle path planning problem is constructed by adopting an arrangement coding mode, individual selection, recombination and variation are carried out by adopting a genetic algorithm (SEGA) for enhancing elite retention, and evolutionary iteration of a population is completed, so that a good initial feasible solution is obtained.

Step S330: solving a constraint linear main problem to obtain a dual vector;

specifically, the set segmentation model is subjected to linear relaxation and then serves as a model of a linear main problem of a column generation algorithm. The limiting linear main problem can be solved by a simplex method to obtain an optimal target value (also the lower numerical bound of the linear main problem) and a dual variable.

Step S340: constructing a first-class shortest path problem with resource constraint according to the pricing problem based on the dual vector;

step S350: solving the first shortest path problem with resource constraint to generate a new column;

step S360: and adding the new column into the linear-limiting main problem, and returning to solve the linear-limiting main problem to obtain dual vectors for iterative solution until the new column cannot be obtained, and outputting a solution of the linear-limiting main problem.

In this embodiment, the variable represents a viable truck route, which is a truck mission that requires minimal access point delay during operation, thus constructing the pricing problem as an elementary shortest path problem with resource constraints (Elementary Shortest Path Problem with Resource Constraints: ESPPRC).

The expression for the pricing problem is:

(PP)min cost(r)＝∑ _i∈C (w _i -a _i λ _i ) Sigma (29) for reducing route costs, where sigma and lambda _i Representing dual constraint variables represented by constraints (25), (26). w represents an intermediate variable of access demand point delay.

The constraint conditions are as follows:

t _i ,w _i ≥0, (34)

constraints (30) are used to ensure the feasibility of the route; constraint (31) is used to ensure that the number of accesses per demand point does not exceed one; constraints (32) - (34) are used to calculate the time delay w of the demand point _i The method comprises the steps of carrying out a first treatment on the surface of the Constraint (35) is used for the decision variable a _i ,

A binary constraint is imposed.

And then solving the ESPPRC model by adopting a labeling algorithm to generate a new column. And adding the generated new column into the linear constraint main problem, and returning to the step S330 for iterative solution. When a new column cannot be obtained, the solving process ends, and a solution of the linear main problem is output.

The supercriterions in the labeling algorithm are important weapons that improve performance. If the adopted super criterion is very effective, the number of elements in the U and P sets in the iterative process can be obviously reduced, so that the solving speed is obviously increased. Generally, the supercriterion is to identify useless paths by comparing the target values of two paths P and Q with the same tail node with the feasible expansion sets E and F. The basic supercriterion for the pricing problem of this embodiment can be described as:

if part of the path Q ₁ ，Q ₂ Satisfies the following formula (36) - (-is)39 Q) ₁ Predominance Q ₂ Part of the path Q ₂ Removed from U without changing the optimal solution.

cost(Q ₁ )≤cost(Q ₂ ),(37)；/>

L(Q ₁ )≤L(Q ₂ ), (39). Wherein v is _Q Representing the tail node of partial path Q, cost (Q) is the current reduced cost of Q, ζ (Q) is the scalable set of Q, L (Q) is the tail node v in partial path Q _Q Is set up in the future.

The proposition declares, any can be associated with Q ₂ The partial paths connected into the complete path can also be connected with Q ₁ Connected into a complete path. In addition, Q ₁ At least one complete path of minimum cost is always generated, the cost of which is not greater than that of the slave Q ₂ Cost of any path generated. Based on this observation, we enforce the dominance rules.

If the condition is satisfied: (Q) ₁ )≤cost(Q ₂ ),(40)；

Then part of path Q ₁ Predominance Q ₂ 。

Wherein the method comprises the steps of

Representing slave node->

To v ₂ Is a running time of the vehicle. Equation (36) and equation (39) require that the tail nodes of the two partial paths be identical. However, we replace them with equation (42), which only requires that part of the path extend to the same node and satisfy the previous dominance rule.

Due to the limitation of the main problemThe instability of the dual variable from one iteration to the next causes the column generation algorithm to have convergence problems in later iterations. In one implementation, a smoothing technique is employed to accelerate convergence. Using a smoothed dual vector pi=αpi ₀ +(1-)π ₁ To build pricing problems, where pi ₁ To limit the dual solution of the main problem in the current iteration ₀ To limit the dipolar solution of the main problem at the last iteration, pi is the smoothed dipolar solution, and alpha e (0, 1) is the smoothing factor. When using the smoothing technique described above, the generated columns may appear in the following 3 cases: a) The columns are based on the dual vector pi ₀ And pi ₁ The check number of (2) is negative, and the column with the negative check number is directly added to the current limit main problem seed; b) The columns are based on the dual vector pi ₁ Is negative but based on the dual vector pi ₀ Is non-negative, at which time the value of a is gradually reduced; c) Absence of basis of dual vector pi ₁ The number of tests in (a) is negative, at which time the alpha value is gradually reduced while solving for the updated pi ₁ . Therefore, convergence is accelerated by weighting the dual vector of the last iteration and the dual vector of the current iteration, obtaining weighted dual vectors, and updating the dual vector of the current iteration.

Although the labeling method based on dynamic programming concept can accurately solve the esperc model, it is time-consuming. In fact, in the column generation algorithm, except for the fact that the exact solution is needed in the last iteration to verify the optimality of the current solution, the rest of iterations need not find the column with the most negative number of tests, but only find the column with the more negative number of tests. A fast and efficient heuristic (e.g., greedy, tabu search, genetic algorithm, etc.) may be employed in the early iterations to find columns with negative numbers of tests. Only when the heuristic finds no columns with negative checknumbers, the labeling algorithm is performed for accurate solution. Therefore, in the initial stage of the column generation iteration, the same heuristic algorithm as the initial solution generation is adopted to search for columns with negative check numbers, and the number of times of calling the labeling algorithm is reduced. That is, when the heuristic can find a column with a negative number of tests, the same heuristic is used to screen the column with a negative number of tests to solve the constraint linear master problem as the initial feasible solution is generated; otherwise, solving the linear-limiting main problem by adopting a labeling algorithm.

And solving the limiting linear main problem and the pricing problem through repeated iteration, and gradually adding a column with the route reduction cost being a negative number into the limiting linear main problem until the route reduction cost of all routes is an integer, thereby indicating that the limiting linear main problem is optimal. At this time, if the optimal solution of the linear main problem is limited to an integer, the optimal solution of the main problem is obtained; if the solution is a decimal, it is a Lower Bound (LB) of the main problem. If the integer solution obtained by solving the linear-constrained main problem of the integer programming version is an Upper Bound (UB) of the main problem. Thus, a branch pricing algorithm (Branch and Price Algorithm) may be employed to branch the valued fractional variables, and then, based on performing the branching operation (branching) at each leaf node of the binary tree, continue to perform the column generation algorithm, updating the upper and lower bounds until the algorithm ends to find the optimal solution to the main problem.

The present invention, in the framework of logic-based Benders decomposition, may require subsequent iterations to continue to solve the master problem, thus employing a bounding method (also known as Route-enumeration method) that can add more columns into the constraint master problem in advance to find the optimal integer solution of the master problem instead of the branch pricing algorithm. The delimitation algorithm is to continue to add columns which meet the limit requirement and are not added to the limit linear main problem after all columns with the route reduction cost being negative are generated, so that the optimal solution of the limit linear main problem is ensured to be an integer.

In summary, in this embodiment, the techniques of performing the Benders decomposition algorithm and the column generation algorithm based on logic are utilized to improve the solution efficiency by organically integrating the Benders decomposition algorithm and the column generation algorithm. Firstly, a Benders decomposition method is used for dividing a complex vehicle unmanned aerial vehicle collaborative distribution problem into two problems, wherein one problem is aimed at a vehicle path and a global lower bound, and the other problem is aimed at an unmanned aerial vehicle line and an objective function, so that complex mathematical calculation is avoided, and the solving efficiency is improved. While reducing the algorithm search space based on the dynamic generation mechanism and dominant rules of the logic design Benders cut. And secondly, the characteristic that the model scale is gradually increased by using the column generation algorithm can effectively avoid the algorithm to be trapped in a local optimal solution, and an efficient genetic algorithm is designed aiming at the pricing problem, so that the efficiency of the algorithm is further improved. And meanwhile, the global searching capability of the algorithm is ensured by using a route enumeration technology and a label algorithm. The method has the advantages of high efficiency of the Benders decomposition algorithm and flexibility of the column generation algorithm, and has remarkable advantages in solving the problem of collaborative distribution.

The algorithm of the invention also has the characteristic of strong applicability. The unmanned aerial vehicle can be suitable for different scales and different unmanned aerial vehicle configurations, and can be adjusted and optimized according to specific problem characteristics so as to adapt to different requirements. For example: route enumeration and labeling algorithms in the algorithm framework are time consuming culprit and a number of experiments indicate that more than about 90% of the time of the column generation algorithm is consumed in the accurate solution of the pricing problem. Although genetic algorithms have been introduced to reduce the number of calls to the labeling algorithm, route enumeration has to be performed using the labeling algorithm in order to ensure the optimality of the final solution. In real-world applications, the problem size is often larger, the decision time is limited, a feasible better scheme needs to be obtained quickly, the scheme optimality demonstration is not concerned, and the problem is the original purpose of numerous heuristic algorithm designs. The flexibility of the column generation algorithm is benefited, the algorithm framework can be simply changed, a high-efficiency mathematical heuristic algorithm aiming at the problem can be obtained, the path enumeration step is eliminated, the calling times and the searching depth of the label algorithm are limited, and the algorithm is terminated by using the total iteration times, the non-improvement iteration times and the running time.

The scheduling optimization method based on the cooperative distribution of the vehicle unmanned aerial vehicle is respectively applied to small-scale, medium-scale and large-scale cooperative distribution scenes, and the stability, reliability and applicability of the algorithm are comprehensively verified to obtain result data shown in fig. 5-7.

Exemplary System

As shown in fig. 8, corresponding to the above-mentioned scheduling optimization method based on the collaborative delivery of the vehicle unmanned aerial vehicle, the embodiment of the invention further provides a scheduling optimization system based on the collaborative delivery of the vehicle unmanned aerial vehicle, where the scheduling optimization system based on the collaborative delivery of the vehicle unmanned aerial vehicle includes:

the mixed integer programming model construction module 600 is configured to construct a mixed integer programming model based on constraint conditions of collaborative delivery of the vehicle unmanned aerial vehicle;

a decomposition module 610, configured to decompose the mixed integer programming model into a vehicle path programming problem and an unmanned aerial vehicle take-off and landing point selection problem according to a logic-based Benders decomposition method;

a solving module 620, configured to solve the vehicle path planning problem, and obtain an optimal solution of the vehicle path planning problem;

the Benders cut module 630 is configured to solve the unmanned aerial vehicle take-off and landing point selection problem based on an optimal solution of the vehicle path planning problem, and generate a Benders cut;

And the iteration module 640 is used for adding the nodes to the vehicle path planning problem, and returning to solve the vehicle path planning problem to perform iteration solution until a preset condition is met, so as to obtain a dispatching optimization result.

Specifically, in this embodiment, the specific function of the scheduling optimization system based on the coordinated delivery of the vehicle unmanned aerial vehicle may also refer to the corresponding description in the scheduling optimization method based on the coordinated delivery of the vehicle unmanned aerial vehicle, which is not described herein again.

Based on the above embodiment, the present invention further provides an intelligent terminal, and a functional block diagram thereof may be shown in fig. 9. The intelligent terminal comprises a processor, a memory, a network interface and a display screen which are connected through a system bus. The processor of the intelligent terminal is used for providing computing and control capabilities. The memory of the intelligent terminal comprises a nonvolatile storage medium and an internal memory. The nonvolatile storage medium stores an operating system and a scheduling optimization program based on collaborative delivery of the vehicle unmanned aerial vehicle. The internal memory provides an environment for the operation of an operating system and a scheduling optimization program based on the collaborative delivery of the vehicle unmanned aerial vehicle in a nonvolatile storage medium. The network interface of the intelligent terminal is used for communicating with an external terminal through network connection. The scheduling optimization program based on the cooperative distribution of the vehicle unmanned aerial vehicle is executed by the processor to realize the step of any scheduling optimization method based on the cooperative distribution of the vehicle unmanned aerial vehicle. The display screen of the intelligent terminal can be a liquid crystal display screen or an electronic ink display screen.

It will be appreciated by those skilled in the art that the schematic block diagram shown in fig. 9 is merely a block diagram of a portion of the structure associated with the present invention and is not limiting of the smart terminal to which the present invention is applied, and that a particular smart terminal may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

In one embodiment, an intelligent terminal is provided, where the intelligent terminal includes a memory, a processor, and a scheduling optimization program based on coordinated delivery of a vehicle unmanned aerial vehicle, where the scheduling optimization program based on coordinated delivery of a vehicle unmanned aerial vehicle is stored in the memory and is capable of running on the processor, and when executed by the processor, the steps of any one of the scheduling optimization methods based on coordinated delivery of a vehicle unmanned aerial vehicle provided by the embodiments of the present invention are implemented.

The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium is stored with a scheduling optimization program based on the cooperative distribution of the vehicle unmanned aerial vehicle, and the scheduling optimization program based on the cooperative distribution of the vehicle unmanned aerial vehicle realizes the steps of any scheduling optimization method based on the cooperative distribution of the vehicle unmanned aerial vehicle provided by the embodiment of the invention when being executed by a processor.

It should be understood that the sequence number of each step in the above embodiment does not mean the sequence of execution, and the execution sequence of each process should be determined by its function and internal logic, and should not be construed as limiting the implementation process of the embodiment of the present invention.

It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, the specific names of the functional units and modules are only for distinguishing from each other, and are not used for limiting the protection scope of the present invention. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.

In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.

Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the examples described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other manners. For example, the apparatus/terminal device embodiments described above are merely illustrative, e.g., the division of the modules or units described above is merely a logical function division, and may be implemented in other manners, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed.

The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that; the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions are not intended to depart from the spirit and scope of the various embodiments of the invention, which are also within the spirit and scope of the invention.

Claims

1. The scheduling optimization method based on the cooperative distribution of the unmanned aerial vehicle is characterized by comprising the following steps of:

2. The scheduling optimization method based on collaborative delivery of a vehicle unmanned aerial vehicle according to claim 1, wherein a coupling variable between the vehicle path planning problem and the unmanned aerial vehicle take-off and landing point selection problem is a vehicle delivery path, the vehicle path planning problem is modeled as a column generation model, a column generation algorithm is adopted to solve the vehicle path planning problem, and an optimal solution of the vehicle path planning problem is a route set.

3. The scheduling optimization method based on collaborative delivery of a vehicle unmanned aerial vehicle according to claim 2, further decomposing a mixed integer programming model into an original vehicle path programming problem and an original unmanned aerial vehicle take-off and landing point selection problem, the coupling variables of which are not the vehicle delivery paths, the generating a Benders cut comprising:

And generating the Benders cut according to the verification result.

4. The scheduling optimization method based on collaborative distribution of a vehicle unmanned aerial vehicle according to claim 3, further comprising:

5. The scheduling optimization method based on collaborative delivery of a vehicle unmanned aerial vehicle according to claim 2, wherein solving the vehicle path planning problem using a column generation algorithm comprises:

generating an initial feasible solution according to a genetic algorithm;

solving a constraint linear main problem to obtain a dual vector;

6. The scheduling optimization method based on collaborative delivery of a vehicle unmanned aerial vehicle according to claim 5, wherein solving the elementary shortest path problem with resource constraints comprises:

7. The scheduling optimization method based on collaborative delivery of a vehicle unmanned aerial vehicle according to claim 5, wherein the dual vector is a smooth dual vector, the dual vector of the last iteration and the dual vector of the current iteration are weighted, the weighted dual vector is obtained, and the dual vector of the current iteration is updated.

8. Scheduling optimization system based on unmanned aerial vehicle cooperation delivery, characterized in that, the system includes:

9. The intelligent terminal, characterized in that the intelligent terminal comprises a memory, a processor and a scheduling optimization program which is stored in the memory and can run on the processor and is based on the cooperative distribution of the vehicle unmanned aerial vehicle, and the scheduling optimization program based on the cooperative distribution of the vehicle unmanned aerial vehicle realizes the steps of the scheduling optimization method based on the cooperative distribution of the vehicle unmanned aerial vehicle according to any one of claims 1 to 7 when being executed by the processor.

10. A computer readable storage medium, wherein a scheduling optimization program based on the coordinated delivery of the vehicle unmanned aerial vehicle is stored on the computer readable storage medium, and the scheduling optimization program based on the coordinated delivery of the vehicle unmanned aerial vehicle realizes the steps of the scheduling optimization method based on the coordinated delivery of the vehicle unmanned aerial vehicle according to any one of claims 1 to 7 when being executed by a processor.