Heterogeneous UAV Scheduling Method, Device, Electronic Equipment and Storage Medium Considering Airport Charging and Battery Replacement Compatibility and Cooperative Operations

Abstract

The present invention discloses a heterogeneous UAV scheduling method, device, electronic device, and storage medium considering airport charging and swapping compatibility and collaborative operation, belonging to the technical field of UAVs. The method includes characterizing the task scenario to obtain the states of UAVs at each passing flight point, and then constructing a UAV scheduling model with the constraints of meeting the flight consistency and energy consumption of UAVs, the service requirements of all task points, and intelligent airport charging and swapping, and taking the minimization of the weighted sum of waiting times at all task points as the objective, and solving to obtain the optimal UAV scheduling plan. The present invention takes into account the type requirements of task points for collaborative operation UAVs, and also considers the type limitations of intelligent airports for chargeable and swappable UAVs. By incorporating the synchronous arrival of UAVs at task points and the charging and swapping decisions of intelligent airports into traditional UAV task allocation and path planning problems, it can further reduce energy consumption waste and improve the applicability of UAVs in various scenarios.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and particularly to a heterogeneous unmanned aerial vehicle scheduling method, device, electronic device, and storage medium that consider airport charging and swapping compatibility and collaborative operations. Background Art

[0002] In recent years, there has been a significant increase in unmanned aerial vehicles, which are becoming an important driving force for services such as public safety, Internet of Things applications, remote healthcare, and situation awareness. Although unmanned aerial vehicles have many advantages such as high mobility, low cost, and easy deployment, their flight time is usually short, which greatly limits their application in large geographical areas.

[0003] To overcome the above disadvantages of unmanned aerial vehicles, in-depth research needs to be carried out in terms of hardware and methods. Among them, in terms of hardware, intelligent airports can provide charging and swapping services for unmanned aerial vehicles. An unmanned aerial vehicle intelligent airport is a ground device composed of functions such as unmanned aerial vehicle storage, intelligent waiting and observation, automatic charging or replacement of unmanned aerial vehicle batteries, wireless communication, automatic battery maintenance, UPS power-off protection, fault self-checking, and takeoff condition detection. In terms of methods, it is necessary to optimize the scheduling scheme of the unmanned aerial vehicle system including intelligent airports.

[0004] Currently, there are some problems when heterogeneous multi-unmanned aerial vehicles perform tasks: (1) When heterogeneous multi-unmanned aerial vehicles provide services for multiple task points, different task points may require multiple types of unmanned aerial vehicles to cooperate. Therefore, it is necessary to reasonably arrange the task allocation and flight trajectories of unmanned aerial vehicles, and try to achieve the synchronous arrival of unmanned aerial vehicles at task points to avoid the waste of hovering energy consumption during waiting caused by some unmanned aerial vehicles arriving too early. (2) When using intelligent airports to charge and swap unmanned aerial vehicles, different types of intelligent airports may have restrictions on the types of unmanned aerial vehicles that can be charged and swapped. In the scenario where heterogeneous unmanned aerial vehicles jointly execute multiple tasks, it is necessary to simultaneously optimize the charging and swapping decisions of intelligent airports so that all unmanned aerial vehicles can meet the energy consumption constraints. Summary of the Invention

[0005] To solve the problems mentioned in the background art, the present invention provides a heterogeneous unmanned aerial vehicle scheduling method, device, electronic device, and storage medium that consider airport charging and swapping compatibility and collaborative operations.

[0006] The present invention is realized through the following technical solutions.

[0007] In a first aspect, the present invention provides a heterogeneous unmanned aerial vehicle scheduling method that considers airport charging and swapping compatibility and collaborative operations, including:

[0008] Characterize the task scenario where heterogeneous multi-UAVs serve multiple task points distributed in the task area, multiple intelligent airports provide multiple charging and battery swapping services for UAVs, different intelligent airports have restrictions on the types of UAVs that can be charged and battery swapped, and different task points have requirements for UAV types;

[0009] Determine the decision variables according to the task scenario, solve the UAV state transition equation, and obtain the states of the UAVs at each passing flight point;

[0010] Based on the states of the UAVs at each passing flight point, construct a UAV scheduling model with the constraints of meeting UAV flight consistency and energy consumption, service requirements for all task points, and charging and battery swapping at intelligent airports, and with the goal of minimizing the weighted sum of waiting times at all task points;

[0011] Use the genetic algorithm to solve the UAV scheduling model and obtain the optimal UAV scheduling plan.

[0012] Furthermore, the task scenario characterization is specifically as follows:

[0013] After numbering all UAVs, form the UAV number set:

[0014] ,

[0015] wherein, N represents n a set composed of the numbers of

[0016] Task points and intelligent airports are collectively referred to as UAV flight points, and after numbering all flight points, form the flight point number set:

[0017] ,

[0018] wherein, represents n + m a set composed of the numbers of n flight points, where the first n flight points are m intelligent airports, the last m flight points are j task points, and the UAV numbered j takes off from the intelligent airport numbered

[0019] Calculate the distances between flight points:

[0020] ,

[0021] wherein, represents the distance between the flight point numbered and the flight point numbered The distance between flight points, where and respectively represent the two-dimensional coordinates of the flight point numbered and the flight point numbered ;

[0022] After numbering all the UAV types, a set of UAV type numbers is formed:

[0023] ,

[0024] In the formula, P represents p a set composed of the numbers of

[0025] The type number of the UAV numbered j is , where ;

[0026] The flight speed of the UAV numbered j is ;

[0027] The flight energy consumption per unit time of the UAV numbered j is ;

[0028] The hovering energy consumption per unit time of the UAV numbered j is ;

[0029] The energy upper limit of the UAV numbered j is ;

[0030] The energy lower limit of the UAV numbered j is ;

[0031] The service time of the task point numbered i is ;

[0032] The importance weight of the task point numbered i is ;

[0033] The service UAV vector of the task point numbered i is , where , represents how many UAV types numbered i are needed to serve the task point numbered k ;

[0034] The UAV numberedj The number set of the rechargeable and replaceable UAV types in the intelligent airport is , where ;

[0035] The number is j The time required for the intelligent airport numbered k to charge and replace the UAV type numbered is . When , . When .

[0036] Furthermore, the scheduling decision variable of the UAV numbered j is:

[0037] ,

[0038] In the formula, represents whether the flight path of the UAV numbered j contains the edge between the flight point numbered and the flight point numbered . When = 1, it means that the flight path of the UAV numbered j contains the edge between the flight point numbered and the flight point numbered . When = 0, it means it does not contain.

[0039] Furthermore, the UAV state transition equation is specifically:

[0040] The set of the numbers of all the passing flight points of the UAV numbered j is:

[0041] ,

[0042] In the formula, represents the set of the numbers of all the passing flight points of the UAV numbered j , represents whether the flight path of the UAV numbered j contains the edge between the flight point numbered i and the flight point numbered ;

[0043] The transfer equation of the passing flight point number of the UAV numbered j is:

[0044] ,

[0045] In the formula, Indicates the number of the j th flight point passed by the drone numbered r +1, where Indicates the number of the j th flight point passed by the drone numbered r , and for any set S, represents the number of elements in S, where represents the number of elements in the set ;

[0046] For the drone numbered j , the time transfer equation for arriving at the flight point is:

[0047] ,

[0048] In the formula, represents the time for the drone numbered j to reach the r +1th flight point passed by, represents the time for the drone numbered j to leave the r th flight point passed by, where , represents the distance between the flight point numbered and the flight point numbered , represents the flight speed of the drone numbered j ;

[0049] For the drone numbered j , the time transfer equation for leaving the flight point is:

[0050] 1) When , ,

[0051] 2) In other cases, ,

[0052] In the formula, represents the time for the drone numbered j to leave the r +1th flight point passed by, represents the time for the drone numbered j to reach the r +1th flight point passed by, represents the time required for the intelligent airport numbered to charge and replace the battery for the drone type numbered , represents the number of the The service time of the task point indicates the time when the UAV numbered l arrives at the th flight point on the route; It indicates the time when the UAV numbered l arrives at the th flight point on the route,

[0053] The remaining energy transfer equation for the UAV numbered j arriving at the flight point on the route is:

[0054] 1) When , ,

[0055] 2) In other cases, ,

[0056] In the formula, indicates the remaining energy when the UAV numbered j arrives at the r +1th flight point on the route. Among them, , indicates the remaining energy when the UAV numbered j arrives at the r th flight point on the route, indicates the time when the UAV numbered j leaves the r th flight point on the route, indicates the time when the UAV numbered j arrives at the r th flight point on the route, ;

[0057] The set of numbers and passing orders of the UAVs that are charged and replaced at the intelligent airport numbered j is:

[0058] ,

[0059] In the formula, indicates the set of numbers and passing orders of the UAVs that are charged and replaced at the intelligent airport numbered j , indicates the number of the a th flight point passed by the UAV numbered b . Among them, a , b are index variables, indicates the set of numbers of all flight points passed by the UAV numbered a , indicates the set of the number of elements;

[0060] The number of each type of UAV serving the task point numbered i :

[0061] ,

[0062] where represents the number of UAVs of type numbered i serving the task point numbered k .

[0063] Furthermore, the constraints are specifically as follows:

[0064] ,

[0065] where indicates whether the flight path of the UAV numbered j contains the edge between the flight point numbered and the flight point numbered ;

[0066] The takeoff / return point constraint of the UAV numbered j :

[0067] ,

[0068] where represents the number of the first flight point passed by the UAV numbered j , represents the number of the j th flight point passed by the UAV numbered ;

[0069] Calculate the number of flights of the UAV numbered j between any two flight point sets:

[0070] ,

[0071] where represents the number of times the UAV numbered j flies from flight point set A to flight point set B, represents whether the flight path of the UAV numbered j contains the edge between the flight point numbered and the flight point numbered ;

[0072] The flight path validity constraint of the UAV numbered j :

[0073] ,

[0074] In the formula, represents the number of times the drone numbered j flies from the set of flight points to the set of flight points , where represents any set of flight points, represents n+m the set composed of the numbers of

[0075] The energy consumption constraint of the drone numbered j :

[0076] ,

[0077] In the formula, represents the remaining energy of the drone numbered j when it reaches the r th passing flight point, represents the lower limit of the energy of the drone numbered j ;

[0078] The service drone type constraint for the task point numbered i :

[0079] ,

[0080] In the formula, represents the number of drones numbered i that serve the task point numbered k , represents how many drones of the type numbered i are needed to serve the task point numbered k ;

[0081] The service drone type constraint for charging and battery swapping of the intelligent airport numbered j :

[0082] ,

[0083] In the formula, represents the type number of the drone numbered , represents the number of the th passing flight point of the drone numbered r , represents the set of the numbers of all passing flight points of the drone numbered , represents the number of elements in the set , represents the number of the drone numbered jNumber set of rechargeable and replaceable UAV types in the intelligent airport;

[0084] The UAV numbered j in the intelligent airport can only perform recharge and replacement for one UAV at most with the constraint:

[0085] ,

[0086] In the formula, represents the time when the UAV numbered a arrives at the b th flight point on the route, represents the time when the UAV numbered arrives at the th flight point on the route, where a , , b and are index variables, represents the type number of the UAV numbered , represents the time required for the intelligent airport numbered j to perform recharge and replacement for the UAV type numbered ;

[0087] Calculate the waiting time of the task point numbered i :

[0088] ,

[0089] In the formula, represents the time when the UAV numbered j arrives at the r th flight point on the route.

[0090] Furthermore, the objective function is:

[0091] ,

[0092] In the formula, G represents the weighted sum of the waiting times of all task points, n represents the number of UAVs, m represents the number of task points, represents the importance weight of the task point numbered i , represents the waiting time of the task point numbered i .

[0093] Furthermore, a genetic algorithm is used to solve the UAV scheduling model, including:

[0094] Encoding and decoding the UAV scheduling decision using 0-1 coding;

[0095] Randomly generate an initial population with a preset size of K ;

[0096] For any encoding, if the scheduling scheme corresponding to the encoding is infeasible, set its fitness to - D , where D is a sufficiently large positive number. If the scheduling scheme corresponding to the encoding is feasible, proceed;

[0097] Calculate the objective function value G under the scheduling scheme corresponding to the encoding, and use as the fitness of the encoding;

[0098] Use the exponential sorting selection method to select individuals in the current population for replication;

[0099] Randomly cross-pair the individuals generated by the selection-replication operation to obtain a crossover probability;

[0100] Perform a mutation operation with a mutation probability;

[0101] Iterate in this way until iterating to L generations, output the encoding with the maximum fitness at this time, and decode it to obtain the optimal scheduling scheme.

[0102] In a second aspect, the present invention also provides a heterogeneous UAV scheduling device considering airport charging and swapping compatibility and collaborative operation, including:

[0103] A task scenario characterization module for characterizing a task scenario where heterogeneous multi-UAVs serve multiple task points distributed in a task area, multiple intelligent airports provide multiple charging and swapping services for UAVs, there are restrictions on the types of UAVs that can be charged and swapped at different intelligent airports, and there are requirements for UAV types at different task points;

[0104] A state transition equation solving module for determining decision variables according to the task scenario, solving the UAV state transition equation, and obtaining the states of the UAVs at each passing flight point;

[0105] A model construction module for constructing a UAV scheduling model based on the states of the UAVs at each passing flight point, with the constraints of meeting UAV flight consistency and energy consumption, service requirements for all task points, and intelligent airport charging and swapping, and with the goal of minimizing the weighted sum of waiting times at all task points;

[0106] A model solving module for using a genetic algorithm to solve the UAV scheduling model and obtain the optimal UAV scheduling scheme.

[0107] In a third aspect, the present invention further provides an electronic device, where the memory is used to store program code, and the processor is used to execute the heterogeneous UAV scheduling method provided in the first aspect according to the instructions in the program code, which takes into account airport charging and swapping compatibility and collaborative operations.

[0108] In a fourth aspect, the present invention further provides a computer storage medium, where the computer-readable storage medium is used to store program code, and the program code is used to execute the heterogeneous UAV scheduling method provided in the first aspect, which takes into account airport charging and swapping compatibility and collaborative operations.

[0109] The present invention provides a heterogeneous UAV scheduling method, device, electronic device, and storage medium that take into account airport charging and swapping compatibility and collaborative operations. This method is aimed at a task scenario where heterogeneous multi-UAVs are used to provide services to multiple task points distributed in a task area, multiple intelligent airports provide multiple charging and swapping operations for UAVs, there are restrictions on the types of UAVs that can be charged and swapped at different intelligent airports, and there are requirements for UAV types at different task points. Decision variables are determined according to the task scenario, the UAV state transition equation is solved to obtain the states of the UAVs at each passing flight point, and based on the states of the UAVs at each passing flight point, with the constraints of meeting UAV flight consistency, energy consumption, service requirements for all task points, and intelligent airport charging and swapping, a UAV scheduling model is constructed with the goal of minimizing the weighted sum of waiting times at all task points, and a genetic algorithm is used to solve the UAV scheduling model to obtain the optimal UAV scheduling plan.

[0110] The beneficial effects of the present invention are as follows: In the traditional optimization problem of heterogeneous multi-UAVs providing services to multiple task points, the requirements for the types of collaborative UAVs at task points are considered, and at the same time, the restrictions on the types of UAVs that can be charged and swapped at intelligent airports are also considered. By incorporating the synchronous arrival of UAVs at task points and the charging and swapping decisions of intelligent airports into the traditional UAV task allocation and path planning problems, energy consumption waste can be further reduced, and the applicability of UAVs in various scenarios can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0111] Figure 1 is a schematic flowchart of the heterogeneous UAV scheduling method introduced in Embodiment 1 of the present invention.

[0112] Figure 2 is a schematic diagram of the implementation scenario of the heterogeneous UAV scheduling method introduced in Embodiment 1 of the present invention.

[0113] Figure 3 is a schematic diagram of the task scenario of the heterogeneous UAV scheduling method introduced in Embodiment 1 of the present invention.

[0114] Figure 4 is a schematic diagram of the optimal scheduling plan introduced in Embodiment 1 of the present invention.

[0115] Figure 5 This is a schematic structural diagram of the heterogeneous UAV scheduling device introduced in the second embodiment of the present invention. Specific implementation manners

[0116] The following further explains the structures involved in the present invention or the technical terms used herein. These explanations are only examples to illustrate how the present invention is implemented and shall not constitute any limitation to the present invention.

[0117] The present invention relates to a heterogeneous UAV scheduling method, device, electronic device and storage medium considering airport charging and swapping compatibility and collaborative operation. Among them, heterogeneous UAVs refer to those where not all the UAVs used belong to the same type but have different types. Collaborative operation means using multiple different types of UAVs to provide services to multiple task points. Among them, each task point may simultaneously require multiple types of UAVs to serve it, and the types of UAVs required by different task points may be different. Embodiment 1

[0118] This embodiment introduces a heterogeneous UAV scheduling method considering charging and swapping compatibility and collaborative operation. Refer to Figure 1 and Figure 2 .

[0119] S110. Characterize the task scenario where heterogeneous multi-UAVs provide services to multiple task points distributed in the task area, multiple intelligent airports provide multiple charging and swapping operations for UAVs, there are restrictions on the types of UAVs that can be charged and swapped at different intelligent airports, and there are requirements for UAV types at different task points.

[0120] In this embodiment, first, determine the task scenario, including heterogeneous multi-UAVs providing services to task points distributed in the task area, multiple intelligent airports providing multiple charging and swapping operations for UAVs, there are restrictions on the types of UAVs that can be charged and swapped at different intelligent airports, and there are requirements for UAV types at different task points.

[0121] Among them, UAV types include fixed-wing UAVs, multi-rotor UAVs, vertical takeoff and landing UAVs, professional sensor UAVs, etc.

[0122] An intelligent airport providing charging and swapping for UAVs means charging the battery of the UAV or replacing the battery of the UAV. Whether it is charging or swapping, within a certain period (such as one day), it can generally be carried out multiple times.

[0123] The task scenario is like traffic accident rescue:

[0124] Multi-rotor UAV: It arrives at the accident scene first and transmits real-time on-site images to help the emergency command center quickly understand the situation. Fixed-wing UAV: It patrols the traffic conditions over a larger area, provides road conditions information around the accident, and assists in planning the best driving routes for emergency vehicles. Professional sensor UAV: It is equipped with thermal imaging instruments to search for injured people at the accident scene, especially at night or in low visibility conditions. It can be considered that the accident scene and its surrounding areas are the mission areas, the accident scene is one mission point, and the surrounding areas of the accident are another mission point.

[0125] Flood disaster emergency management:

[0126] Multi-rotor UAV: It flies low over the affected area, takes real-time pictures of the disaster situation, and transmits detailed images back. Vertical take-off and landing UAV: It drops relief supplies such as medicines and food in remote or inaccessible areas and conducts environmental monitoring at the same time. Professional sensor UAV: It is equipped with water quality detection instruments to monitor pollutants in the flood and evaluate the environmental impact. It can be considered that the affected area is the mission area, and the affected area, remote or inaccessible areas, and flood areas are mission points respectively.

[0127] Large public event safety management:

[0128] Fixed-wing UAV: It conducts extensive patrols over the event area to monitor crowd flow and overall safety conditions. It can be considered that the event area is the mission area.

[0129] The above UAV types and mission scenarios are only examples and are not limited to the above.

[0130] Secondly, the mission scenarios are characterized as follows:

[0131] After numbering all UAVs, a UAV number set is formed:

[0132] ,

[0133] where N represents n a set composed of the numbers of n UAVs. In this embodiment, Figure 3 = 3, and the UAVs are represented by triangles in

[0134] The mission points and intelligent airports are collectively referred to as the flight points of the UAVs. After numbering all flight points, a flight point number set is formed:

[0135] ,

[0136] where represents n + mThe set composed of the numbers of flight points, where the first n flight points are n intelligent airports, and the subsequent m flight points are m mission points. And the drone numbered j takes off from the intelligent airport numbered j and returns to this intelligent airport after completing all tasks. In this embodiment, m = 17, the mission points are represented by circles in Figure 3 , and the numbers of the mission points are marked in the circles. The intelligent airports are represented by squares in Figure 3 , and the numbers of the intelligent airports are marked in the squares.

[0137] Calculate the distances between flight points:

[0138] ,

[0139] In the formula, represents the distance between the flight point numbered and the flight point numbered . Among them, and respectively represent the two-dimensional coordinates of the flight point numbered and the flight point numbered .

[0140] After numbering all drone types, a drone type number set is formed:

[0141] ,

[0142] In the formula, P represents p the set composed of the numbers of p drone types. In this embodiment,

[0143] The type number of the drone numbered j is , where . In this embodiment, w 1 = 1, w 2 = w 3 = 2.

[0144] The flight speed of the drone numbered j is . In this embodiment, v 1 = 90 km / h, v2 = v 3 = 70 km / h.

[0145] The energy consumption per unit time of the UAV numbered j is . In this embodiment, all are 100 W.

[0146] The hovering energy consumption per unit time of the UAV numbered j is . In this embodiment, all are 100 W.

[0147] The energy upper limit of the UAV numbered j is . In this embodiment, , .

[0148] The energy lower limit of the UAV numbered j is . In this embodiment, , .

[0149] The service time of the task point numbered i is . In this embodiment, all are 10 min.

[0150] The importance weight of the task point numbered i is . In this embodiment, , and for the others are all 1.

[0151] The service UAV vector of the task point numbered i is , where , represents how many UAVs numbered i are required to serve the task point numbered k . In this embodiment, , , .

[0152] The set of numbers of the UAV types that can be charged and replaced at the intelligent airport numbered j is , where . In this embodiment, , .

[0153] The intelligent airport numbered j takes k time to charge and replace the drones of the type numbered . Among them, when , . When , . In this embodiment, , .

[0154] S120. Determine the decision variables according to the mission scenario, solve the UAV state transition equation, and obtain the states of the UAVs at each passing flight point.

[0155] The constraints and objective function of the UAV scheduling model are related to the UAV state transition equation. Therefore, it is necessary to solve the UAV state transition equation first. When the decision variables are given, it is equivalent to giving the flight path of each UAV. Therefore, according to its flight order, the states of each UAV at each passing flight point can be obtained in turn, that is, the state transition equation is obtained.

[0156] After obtaining the state transition equation, it is known that under the given decision variables, the state of each UAV at each flight point during the entire flight process.

[0157] It is necessary to know the state of each UAV at each flight point during the entire flight process through the state transition equation, and these states are related to the constraints and objective function of the UAV scheduling model.

[0158] Among them, the states of the UAVs at each passing flight point include the arrival and departure times of the UAVs at each passing flight point and the remaining energy of the UAVs when they arrive at each passing flight point.

[0159] Specifically, solving the UAV state transition equation is as follows:

[0160] The scheduling decision variables of the UAV numbered j are:

[0161] ,

[0162] In the formula, represents whether the flight path of the UAV numbered j contains the edge between the flight point numbered and the flight point numbered . When = 1, it means that the flight path of the UAV numbered j contains the edge between the flight point numbered and the flight point numbered . When When it is = 0, it means not included.

[0163] The set of the numbers of all the passing flight points of the drone numbered j is:

[0164] ,

[0165] In the formula, represents the set of the numbers of all the passing flight points of the drone numbered j , represents whether the flight path of the drone numbered j includes the edge between the flight point numbered i and the flight point numbered .

[0166] The transfer equation of the numbers of the passing flight points of the drone numbered j is:

[0167] ,

[0168] In the formula, represents the number of the j ( r + 1)-th passing flight point of the drone numbered , j represents the number of the r -th passing flight point of the drone numbered , where j represents the set of the numbers of all the passing flight points of the drone numbered , for any set S, represents the number of elements of S, where represents the number of elements of the set

[0169] The transfer equation of the arrival time of the passing flight points of the drone numbered j is:

[0170] ,

[0171] In the formula, represents the time when the drone numbered j arrives at the r ( + 1)-th passing flight point, j represents the time when the drone numbered r leaves the -th passing flight point, where represents the distance between the flight point numbered and the flight point numbered . Indicates the flight speed of the UAV numbered j . Indicates the set of numbers of all the flight points passed by the UAV numbered j .

[0172] For the UAV numbered j , the time transfer equation for leaving the flight point passed is:

[0173] 1) When , ,

[0174] 2) In other cases, ,

[0175] In the formula, Indicates the time when the UAV numbered j leaves the r +1-th flight point passed. Indicates the time when the UAV numbered j arrives at the r +1-th flight point passed. Indicates the time required for the intelligent airport numbered to charge and replace the battery of the UAV numbered . Indicates the service time of the task point numbered , and indicates the time when the UAV numbered arrives at the l -th flight point passed; Indicates the time when the UAV numbered arrives at the l -th flight point passed, .

[0176] For the UAV numbered j , the remaining energy transfer equation when arriving at the flight point passed is:

[0177] 1) When , ,

[0178] 2) In other cases, ,

[0179] In the formula, Indicates the remaining energy when the UAV numbered j arrives at the r +1-th flight point passed. Among them, , Indicates the remaining energy when the UAV numbered j arrives at the r -th flight point passed. Indicates the UAV numberedj The time when the UAV numbered r leaves the th flight point on the way, j represents the time when the UAV numbered r arrives at the .

[0180] The set of numbers and passing orders of the UAVs that are charged or replaced at the intelligent airport numbered j is:

[0181] ,

[0182] In the formula, represents the set of numbers and passing orders of the UAVs that are charged or replaced at the intelligent airport numbered j , represents the number of the a th flight point passed by the UAV numbered b , where a , b are index variables, represents the set of numbers of all flight points passed by the UAV numbered a , represents the number of elements in the set .

[0183] The number of each type of UAV serving the task point numbered i :

[0184] ,

[0185] In the formula, represents the number of UAVs of type number i serving the task point numbered k .

[0186] This embodiment studies the flight scheduling of UAVs, and the decision variable is naturally the flight path of each UAV. Specifically, it is whether the flight path of the UAV j contains the edge between the flight point numbered and the flight point numbered , that is, . In all task scenarios, the decision variable is represented by the symbol . However, in different task scenarios, the number and type of UAVs will be different, and the positions (i.e., two-dimensional coordinates) and types of flight points will also be different. For the sake of understanding, for example, assume that a certain scenario only contains UAV 1, flight point 1, and flight point 2. Then in this scenario is actually and These two variables.

[0187] S130. According to the states of the UAV at each passing flight point, with the constraints of meeting the flight consistency and energy consumption of the UAV, the service requirements of all task points, and the charging and swapping of the intelligent airport, and with the goal of minimizing the weighted sum of the waiting times of all task points, a UAV scheduling model is constructed.

[0188] Through the above steps, the UAV state transition equation is obtained. By means of the state transition equation, the state of each UAV at each flight point during the entire flight process is known, and these states are related to the constraints and objective function of the UAV scheduling model.

[0189] Flight consistency means that the flight path of the UAV corresponding to the decision variable is actually feasible. For the sake of easy understanding, for example, it will not occur that a certain UAV first flies from task point 1 to task point 2, and then suddenly disappears out of thin air at task point 2 and appears out of thin air at task point 3, or first flies from task point 1 to task point 2, and then suddenly obtains the "duplication" ability at task point 2 and flies to task point 3 and task point 4 simultaneously, etc.

[0190] The constraints include the value constraints of the scheduling decision variables of the UAV, the takeoff / return point constraints of the UAV, the flight path validity constraints of the UAV, the energy consumption constraints of the UAV, the service UAV type constraints of the task points, the UAV type constraints for charging and swapping at the intelligent airport, and the constraint that the intelligent airport can charge and swap at most one UAV at the same time. Specifically:

[0191] For the UAV numbered j the value constraints of the scheduling decision variables:

[0192] ,

[0193] In the formula, represents whether the flight path of the UAV numbered j contains the edge between the flight point numbered and the flight point numbered .

[0194] For the UAV numbered j the takeoff / return point constraints:

[0195] ,

[0196] In the formula, represents the number of the first passing flight point of the UAV numbered j , represents the number of the j th passing flight point of the UAV numbered .

[0197] Calculate the number of flights of the drone numbered j between any two sets of flight points:

[0198] ,

[0199] wherein, represents the number of times the drone numbered j flies from the set of flight points A to the set of flight points B, represents whether the flight path of the drone numbered j contains the edge between the flight point numbered and the flight point numbered .

[0200] Flight path validity constraint of the drone numbered j :

[0201] ,

[0202] wherein, represents the number of times the drone numbered j flies from the set of flight points to the set of flight points , where represents any set of flight points, represents n+m the set composed of the numbers of

[0203] Energy consumption constraint of the drone numbered j :

[0204] ,

[0205] wherein, represents the remaining energy of the drone numbered j when it reaches the r th intermediate flight point, represents the lower energy limit of the drone numbered j .

[0206] Service drone type constraint for the task point numbered i :

[0207] ,

[0208] wherein, represents the number of drones numbered i that serve the task point numbered k , represents the drone numbered iHow many drones numbered k are required to serve the task points?

[0209] For the drone type numbered j constraints on charging and swapping at the intelligent airport numbered

[0210] ,

[0211] wherein, represents the type number of the drone numbered , represents the number of the th flight point passed by the drone numbered r , represents the set of numbers of all flight points passed by the drone numbered , represents the number of elements in the set , represents the set of numbers of drone types that can be charged and swapped at the intelligent airport numbered j .

[0212] For the intelligent airport numbered j constraints on charging and swapping at most one drone at the same time:

[0213] ,

[0214] wherein, represents the time when the drone numbered a arrives at the b th flight point passed by, represents the time when the drone numbered arrives at the th flight point passed by, where a , , b and are index variables, represents the type number of the drone numbered , represents the time required for the intelligent airport numbered j to charge and swap the drone type numbered .

[0215] Calculate the waiting time of the task point numbered i :

[0216] ,

[0217] wherein, represents the time when the drone numbered j arrives at ther The time passing through the flight points.

[0218] After the constraints are determined, an objective function is established with the goal of minimizing the weighted sum of the waiting times of all task points, and the objective function is determined as:

[0219] ,

[0220] In the formula, represents the weighted sum of the waiting times of all task points, n represents the number of UAVs, m represents the number of task points, represents the importance weight of the task point numbered i , represents the waiting time of the task point numbered i .

[0221] Summarize the constraints and the objective function to obtain a heterogeneous UAV scheduling model with the goal of minimizing the weighted sum of the waiting times of all task points.

[0222] The state of each UAV at each flight point during the entire flight process is obtained through the state transition equation, and these states are related to the constraints and the objective function of the UAV scheduling model. For example, the state of "the arrival time of the UAV passing through the flight point" is related to the constraint that "the intelligent airport numbered j can only charge and replace the power of one UAV at most at the same time", and is also related to the "waiting time of the task point numbered i " and thus related to the objective function; the state of "the remaining energy of the UAV numbered j when it arrives at the passing flight point" is related to the "energy consumption constraint of the UAV numbered j ". Therefore, only by solving the state transition equation can we know whether the given decision variables violate the constraints and the corresponding objective function values.

[0223] For easier understanding, the relationship between the state transition equation, the constraints, and the objective function is further explained: The formulas of the constraints and the objective function include the states of the UAVs at each passing flight point as independent variables. For example, assume the objective function is , and b represents the state of the UAV's departure time. Then, to calculate y , we need to know b . Another example, assume a certain constraint is , where d represents the state of the UAV's arrival time. Then, to determine whether this constraint is satisfied, we need to know d。The state and constraints of the UAV and the objective function are related in this way. Furthermore, to find the state of the UAV at each flight point, it is necessary to know how the state of the UAV at one flight point transfers from its state at the previous flight point, that is, it is necessary to find the state transition equation.

[0224] It should be noted that the meanings of the same parameters involved in all the above formulas are consistent. For parameters without directly stated meanings, they can be found in the formulas in the context.

[0225] S140. Solve the UAV scheduling model using the genetic algorithm to obtain the optimal UAV scheduling plan.

[0226] In this embodiment, according to the constraints and the objective function, a UAV scheduling model can be constructed, and then the genetic algorithm is used to solve the UAV scheduling model to obtain the optimal UAV scheduling plan. Specifically:

[0227] S1401. Encode and decode the UAV scheduling decision using 0-1 encoding.

[0228] S1402. Randomly generate an initial population with a preset size of 100.

[0229] S1403. For any encoding, if the scheduling plan corresponding to this encoding is infeasible, set its fitness to - D , where D is a sufficiently large positive number. If the scheduling plan corresponding to this encoding is feasible, then perform S1404. In this embodiment, D = 200.

[0230] S1404. Calculate the objective function value G under the scheduling plan corresponding to this encoding, and use as the fitness of this encoding.

[0231] S1405. Use the exponential ranking selection method to select individuals in the current population for replication.

[0232] S1406. Randomly cross-pair the individuals generated by the selection-replication operation, and the crossover probability is . In this embodiment, = 0.9.

[0233] S1407. Perform the mutation operation with the mutation probability . In this embodiment, = 0.4.

[0234] S1408. The above S1403 to S1407 are one iteration. Iterate in this way until it iterates to 30 generations, output the encoding with the maximum fitness at this time, and decode it to obtain the optimal scheduling plan.

[0235] In this embodiment, the minimum value of the objective function can be obtained as 2983.75, and the UAV scheduling scheme as shown in Figure 4 can be obtained. Among them, the thin solid arrows represent the flight paths of the UAVs numbered 1, the dashed arrows represent the flight paths of the UAVs numbered 2, and the thick solid arrows represent the flight paths of the UAVs numbered 3. The arrow directions are the flight directions of the UAVs.

[0236] Through the heterogeneous UAV scheduling method, in the process of heterogeneous multi-UAVs providing services to multiple task points, not only the type requirements of the task points for collaborative operation UAVs are considered, but also the type restrictions of the intelligent airports on the rechargeable and replaceable UAVs are considered. By incorporating the synchronous arrival of UAVs at task points and the recharge and replacement decision of intelligent airports into the traditional UAV task allocation and path planning problems, the energy consumption waste can be further reduced, and the applicability of UAVs in various scenarios can be improved.

[0237] Embodiment 2

[0238] This embodiment introduces a heterogeneous UAV scheduling device that considers the compatibility of UAV recharge and replacement and collaborative operation. Refer to Figure 5 , this heterogeneous UAV scheduling device implements the heterogeneous UAV scheduling method described in Embodiment 1. This heterogeneous UAV scheduling device can be implemented in the form of hardware and / or software, and this heterogeneous UAV scheduling device can be configured in an electronic device. This heterogeneous UAV scheduling device includes: a task scenario characterization module 210, a state transition equation solving module 220, a model construction module 230, and a model solving module 240.

[0239] The task scenario characterization module 210 is used to characterize the task scenario where heterogeneous multi-UAVs provide services to multiple task points distributed in the task area, multiple intelligent airports provide multiple recharge and replacement operations for UAVs, there are type restrictions on rechargeable and replaceable UAVs in different intelligent airports, and there are UAV type requirements in different task points.

[0240] The state transition equation solving module 220 is used to determine decision variables according to the task scenario, solve the UAV state transition equation, and obtain the states of the UAVs at each passing flight point.

[0241] The model construction module 230 is used to construct a UAV scheduling model based on the states of the UAVs at each passing flight point, with the constraints of meeting the UAV flight consistency, energy consumption, service requirements of all task points, and recharge and replacement of intelligent airports, and with the goal of minimizing the weighted sum of the waiting times of all task points.

[0242] The model solving module 240 is used to solve the UAV scheduling model by using a genetic algorithm to obtain the optimal UAV scheduling scheme.

[0243] Through the heterogeneous UAV scheduling device, the heterogeneous UAV scheduling method described in Embodiment 1 that considers the compatibility of airport charging and battery swapping and collaborative operations can be implemented.

[0244] Embodiment 3

[0245] This embodiment also introduces an electronic device, which includes a memory and a processor. The memory is used to store program code and transmit the program code to the processor. The processor is used to execute the steps of the heterogeneous UAV scheduling method provided in any of the above embodiments according to the instructions in the program code.

[0246] Among them, the processor may include one or more processing cores, such as a 3-core processor, an 8-core processor, etc. The processor may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor may also include a main processor and a coprocessor. The main processor is a processor used to process data in the wake state, also known as the central processing unit (CPU); the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor may further include an artificial intelligence (AI) processor, which is used to process computational operations related to machine learning. The memory may include one or more storage media. The memory may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices and flash storage devices. In this embodiment, the memory is at least used to store the following program code. After the program code is loaded and executed by the processor, the relevant steps in the heterogeneous UAV scheduling method disclosed in any of the foregoing embodiments can be implemented. In addition, the resources stored in the memory may also include an operating system and data, etc., and the storage method may be temporary storage or permanent storage. The operating system may be Windows or other types of operating systems. The data may include, but is not limited to, the data involved in the above method.

[0247] Embodiment 4

[0248] This embodiment also introduces a computer-readable storage medium, on which program code is stored. When the program code is executed by a processor, the steps of the heterogeneous UAV scheduling method provided in any of the above embodiments can be implemented.

[0249] Among them, the storage medium may include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0250] In several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0251] In addition, in each embodiment of the present invention, the functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in a unit. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0252] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A heterogeneous UAV scheduling method considering airport charging and battery swapping compatibility and collaborative operation, characterized in that: include: Characterize the mission scenarios where heterogeneous multiple drones provide services for multiple mission points distributed in the mission area, multiple smart airports provide multiple charging and battery replacement for drones, different smart airports have restrictions on the types of drones that can be charged and replaced, and different mission points have requirements for drone types; Determine the decision variables according to the mission scenario, solve the UAV state transition equation, and obtain the state of the UAV at each flight point; According to the status of the drone at each flight point, the drone scheduling model is constructed with the goal of minimizing the weighted waiting time of all mission points, meeting the flight consistency and energy consumption of the drone, the service requirements of all mission points, and the charging and swapping of smart airports as constraints; Genetic algorithm is used to solve the UAV scheduling model and obtain the optimal UAV scheduling solution; Among them, each mission point requires multiple types of drones to serve it at the same time; Number j The scheduling decision variables of the UAV are: ， In the formula, Indicates the number j Does the flight path of the drone contain The flight point and number are The edge between the flight points, when =1, indicating that the number is j The flight path of the drone contains The flight point and number are The edge between the flight points, when =0, it means not included. n Indicates the number of smart airports, m Indicates the number of task points; The specific state transfer equation of the drone is: Number j The set of numbers of all the flight points passed by the drone is: ， In the formula, Indicates the number j The set of numbers of all the flight points that the drone passes through, Indicates the number j Does the flight path of the drone contain i The flight point and number are The edges between the flight points; Number j The transfer equation of the flight point number passed by the UAV is: ， In the formula, Indicates the number j The first drone r +1 number of the flight point you passed through, Indicates the number j The first drone r The flight point numbers passed by, among which, , for any set S, represents the number of elements of S, where Representing a collection The number of elements of ; Number j The time transfer equation of the drone passing through the flight point is: ， In the formula, Indicates the number j The drone arrived at r +1 time for passing through the flight point, Indicates the number j The drone left r The time of passing the flight points, among which, , Indicates the number The flight point and number are The distance between the flight points, Indicates the number j The flight speed of the drone; Number j The time transfer equation of the UAV passing through the flight point is: 1) When hour, , 2) In other cases, , In the formula, Indicates the number j The drone left r +1 time for passing through the flight point, Indicates the number j The drone arrived at r +1 time for passing through the flight point, Indicates the number The number of smart airport pairs is The time required to charge and replace the battery of the drone type, Indicates the number The service time of the task point, Indicates the number The drone arrived at The time of passing the flight point, ; Number j The remaining energy transfer equation for the UAV to reach the passing flight point is: 1) When hour, , 2) In other cases, , In the formula, Indicates the number j The drone arrived at r +1 remaining energy when passing through a flight point, of which: , Indicates the number j The drone arrived at r The remaining energy when passing the flight point, Indicates the number j The drone left r The time of passing the flight point, Indicates the number j The drone arrived at r The time of passing the flight point, ; Number j The serial numbers and route sequence of the drones that are charged and replaced at the smart airport are as follows: ， In the formula, Indicates the number j The number and route sequence of drones that are charged and replaced at smart airports. Indicates the number a The first drone b The flight point numbers passed by, among which, a , b is an indicator variable, Indicates the number a The set of numbers of all the flight points that the drone passes through, Representing a collection The number of elements of ; For the number i The number of drone types serving the mission points: ， In the formula, Indicates that the pair number is i The type number of the task point to provide service is k The number of drones; Number j The value constraints of the UAV scheduling decision variables are: ， In the formula, Indicates the number j Does the flight path of the drone contain The flight point and number are The edges between the flight points; Number j The take-off / return point constraints of the drone: ， In the formula, Indicates the number j The number of the first flight point that the drone passes through. Indicates the number j The first drone The number of the flight point passed by; The calculation number is j The number of flights of the drone between any two flight point sets: ， In the formula, Indicates the number j The number of times the drone flies from flight point set A to flight point set B, Indicates the number j Does the flight path of the drone contain The flight point and number are The edges between the flight points; Number j The flight path validity constraints of the UAV are: ， In the formula, Indicates the number j The drones are collected by flight points Fly to the flight point The number of times, among which, represents any set of flight points, express n+m A set of flight point numbers; Number j Energy consumption constraints of the UAV: ， In the formula, Indicates the number j The drone arrived at r The remaining energy when passing the flight point, Indicates the number j The lower energy limit of the UAV; Number i The service drone type constraints of the mission point are: ， In the formula, Indicates that the pair number is i The type number of the task point to provide service is k The number of drones, Indicates the number i How many task points do you need to have? k Types of drones to be serviced; Number j The types of drones that can be charged and replaced at smart airports are as follows: ， In the formula, Indicates the number The type number of the drone, Indicates the number The first drone r The number of the flight points passed by, Indicates the number The set of numbers of all the flight points that the drone passes through, Representing a collection The number of elements of Indicates the number j A numbered set of drone types that can be charged and replaced at smart airports; Number j The smart airport can only charge and replace one drone at a time: ， In the formula, Indicates the number a The drone arrived at b The time of passing the flight point, Indicates the number The drone arrived at The time of passing the flight points, among which, a , , b and is an indicator variable, Indicates the number The type number of the drone, Indicates the number j The number of smart airport pairs is The time required to charge and replace the battery of the drone type; The calculation number is i Waiting time for task point: ， In the formula, Indicates the number j The drone arrived at r The time for passing the flight point.

2. The heterogeneous UAV scheduling method considering airport charging and battery replacement compatibility and collaborative operation according to claim 1 is characterized in that: The task scenario is specifically characterized as follows: After all drones are numbered, they form a drone number set: , In the formula, N express n A set of drone numbers; Mission points and smart airports are collectively referred to as the flight points of the drone, and all flight points are numbered to form a flight point number set: , In the formula, express n + m A set of flight point numbers, of which the first n The flight points are n A smart airport, m The flight points are m Task points, and numbered j The drone is numbered j Take off from the smart airport and return to the smart airport after completing all tasks; Calculate the distance between flight points: ， In the formula, Indicates the number The flight point and number are The distance between the flight points, where and Respectively indicate the number The flight point and number are The two-dimensional coordinates of the flight point; All drone types are numbered to form a drone type number set: ， In the formula, P express p A collection of numbers of drone types; Number j The type number of the drone is ,in, ; Number j The flight speed of the drone is ; Number j The energy consumption per unit time of the UAV is ; Number j The hovering energy consumption per unit time of the UAV is ; Number j The energy limit of the drone is ; Number j The lower limit of the energy of the UAV is ; Number i The service time of the task point is ; Number i The importance weight of the task point is ; Number i Mission point service drone vector ,in, , Indicates the number i How many task points do you need to have? k Types of drones to be serviced; Number j The number set of drone types that can be charged and replaced at smart airports is ,in, ; Number j The number of smart airport pairs is k The time required for charging and replacing the battery of the drone type is , among which, when hour, ,when hour, .

3. The heterogeneous UAV scheduling method considering airport charging and battery replacement compatibility and collaborative operation according to claim 1 is characterized in that: The objective function is: ， In the formula, G represents the weighted sum of waiting time of all task points, n Indicates the number of drones, m Indicates the number of task points. Indicates the number i The importance weight of the task point, Indicates the number i The waiting time of the task point.

4. The heterogeneous UAV scheduling method considering airport charging and battery replacement compatibility and collaborative operation according to claim 1 is characterized in that: Genetic algorithm is used to solve the UAV scheduling model, including: Use 0-1 encoding to encode and decode drone scheduling decisions; Randomly generate a preset size K The initial population of For any code, if the scheduling scheme corresponding to the code is not feasible, let its fitness be - D ,in D is a sufficiently large positive number. If the scheduling scheme corresponding to the encoding is feasible, proceed to the next step; Calculate the objective function value G under the encoding corresponding scheduling scheme, and use as the fitness of the encoding; Use exponential sorting selection to select individuals in the current population for replication; The individuals generated by the selection-copy operation are randomly cross-paired to obtain the crossover probability; Perform mutation operation with mutation probability; Iterate in this way until the iteration reaches L The code with the highest fitness is output, and the optimal scheduling solution is obtained after decoding.

5. A device, characterized in that: The device is used to implement the heterogeneous drone scheduling method considering airport charging and battery replacement compatibility and collaborative operation as described in any one of claims 1 to 4, comprising: Mission scenario characterization module, used to characterize mission scenarios where heterogeneous multiple drones provide services for multiple mission points distributed in the mission area, multiple smart airports provide multiple charging and battery replacement for drones, different smart airports have restrictions on the types of drones that can be charged and replaced, and different mission points have requirements for drone types; The state transfer equation solving module is used to determine the decision variables according to the mission scenario, solve the UAV state transfer equation, and obtain the state of the UAV at each passing flight point; The model building module is used to build a UAV scheduling model based on the status of the UAV at each flight point, with the constraints of UAV flight consistency and energy consumption, service requirements of all mission points, and charging and swapping at smart airports, and with the goal of minimizing the weighted sum of waiting times at all mission points; The model solving module is used to solve the UAV scheduling model using a genetic algorithm to obtain the optimal UAV scheduling solution.

6. An electronic device, characterized in that: It includes a processor and a memory, the memory is used to store program code, and the processor is used to execute the heterogeneous UAV scheduling method considering airport charging and battery replacement compatibility and collaborative operation as described in any one of claims 1-4 according to the instructions in the program code.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program code, and the program code is used to execute the heterogeneous drone scheduling method considering airport charging and battery replacement compatibility and collaborative operation as described in any one of claims 1-4.

Images