Multi-jurisdictional drone service system integration planning method and system

Abstract

This invention discloses an integrated planning method and system for a multi-jurisdictional unmanned aerial vehicle (UAV) service system. The method constructs an overall planning model with the goal of minimizing total cost, and sequentially establishes a front-end UAV allocation model and a back-end battery swapping station allocation model. With the objectives of minimizing average response time and minimizing the total expected queue length, respectively, the total number of UAVs, charging stations, and backup batteries, as well as the jurisdiction-level configuration scheme, are obtained. This invention achieves integrated collaborative planning of front-end transportation and back-end battery swapping, reducing the total system cost and average response time, balancing the operating load of each jurisdiction, effectively avoiding service interruptions due to a lack of available batteries during peak periods, and significantly improving system reliability and engineering practicality.

Description

Technical Field

[0001] This invention relates to the field of drone logistics and delivery technology, and in particular to a method and system for integrated planning of multi-jurisdictional drone service systems. Background Technology

[0002] In recent years, with the rapid development of the low-altitude economy and drone logistics applications, drone delivery systems have gradually evolved from a single aircraft path optimization problem to an infrastructure collaborative planning problem for city-level service networks. In actual operation, drones not only need to complete transportation tasks, but also need to rely on backend facilities such as take-off and landing points, battery swapping stations, charging stations, and spare batteries to maintain continuous operation capabilities. In terms of power replenishment methods, existing drone systems have evolved from returning to base for charging and in-flight charging to rapid battery swapping at stations. Compared with returning to base for charging, battery swapping can transform front-end downtime into a back-end battery inventory turnover and charging station power replenishment process, which is more suitable for high-frequency, continuous, and cross-regional urban logistics operation scenarios. At the same time, the demand for city-level drone transportation has obvious regional differences and randomness; the order intensity, service distance, cross-regional support relationships, and response requirements vary between different jurisdictions.

[0003] Existing technologies can be broadly categorized into four types: The first type is single-drone endurance and path planning technology, which mainly uses battery capacity, flight distance, payload weight, and energy consumption models as constraints to study the reachability and delivery route feasibility of a single drone. The second type is charging facility layout technology, which typically expands the service radius by setting up intermediate charging nodes or return charging stations, but most of these treat recharging time as a fixed parameter and do not further characterize the queuing and battery turnover processes within the station. The third type is battery swapping facility and inventory scheduling technology, which focuses on the number of charging slots, the number of spare batteries, battery rotation, and replenishment mechanisms within a battery swapping station, which can improve the operating efficiency of a single station, but the research objects are mostly concentrated on a single station or a single area. The fourth type is multi-regional or multi-vehicle collaborative delivery technology, which models cross-regional scheduling and the collaboration between drones and other vehicles, but usually assumes that recharging resources are sufficient or simplifies recharging processing to a fixed time parameter, and has not yet jointly considered the probability of a battery swapping station having a fully charged battery available with transportation scheduling.

[0004] Existing technologies suffer from the following drawbacks: Insufficient consideration of service differences; current drone scheduling research often assumes homogeneous demand across jurisdictions, failing to fully reflect differences in order intensity, service distance, and response requirements. Lack of front-end and back-end system coordination optimization; existing research typically handles drone transportation scheduling and battery swapping station resource allocation separately, making it difficult to ensure a match between front-end capacity and back-end replenishment capabilities. Insufficiently refined modeling of cross-regional coordination mechanisms; while some studies consider cross-regional support, descriptions of scheduling priorities, cross-regional service time differences, call loss, and queuing effects are rather coarse. Lack of the crucial service constraint of immediate access to fully charged batteries; existing methods often fail to explicitly constrain whether drones can directly obtain fully charged batteries upon arrival at the station with a sufficiently high probability, potentially leading to downtime or even service interruptions during peak periods. Overall, in multi-jurisdictional city scenarios, existing technologies lack a unified system integration method that integrates front-end transportation services, cross-regional scheduling, back-end battery swapping supply, and total resource planning. Summary of the Invention

[0005] Therefore, the technical problem to be solved by the present invention is to overcome the problems in the existing multi-jurisdictional drone service system, such as the fragmentation of transportation scheduling and battery swapping resource allocation, the lack of systematic consideration of jurisdictional differences and cross-regional collaborative mechanisms, and the failure to introduce the probability constraint of available fully charged batteries in the planning.

[0006] To address the aforementioned technical problems, this invention provides a method for integrated planning of a multi-jurisdictional unmanned aerial vehicle (UAV) service system, characterized by comprising: A planning model is established with the goal of minimizing the total cost of a multi-jurisdictional drone service system. The total cost of the multi-jurisdictional drone service system includes the fixed cost of drones, the fixed cost of charging stations, the fixed cost of backup batteries, and the operational cost. The planning model is solved under the preset operational constraints of the multi-jurisdictional drone service system to obtain the total number of drones, the total number of charging stations, and the total number of backup batteries. Based on the total number of drones, a front-end drone allocation model is established with the goal of minimizing the average response time of the multi-jurisdictional drone service system. The number of drones configured in each jurisdiction is obtained by solving the front-end drone allocation model. Based on the total number of charging slots and the total number of spare batteries, a back-end battery swapping station allocation model is established with the goal of minimizing the total expected queue length of each charging station. The number of charging racks and the number of spare batteries in each jurisdiction are obtained by solving the back-end battery swapping station allocation model. Based on the number of drones configured in each jurisdiction, as well as the number of charging racks and spare batteries in each jurisdiction, an integrated planning scheme for a multi-jurisdictional drone service system is obtained. The integrated planning scheme includes the total number of drones, charging racks, and spare batteries, as well as the configuration results of drones, charging racks, and spare batteries in each jurisdiction.

[0007] In one embodiment of the present invention, the method for establishing a planning model with the objective of minimizing the total cost of a multi-jurisdictional unmanned aerial vehicle (UAV) service system is as follows: Construct the objective function ,in, For the total number of drones, Total number of charging slots Total number of backup batteries, , , , These represent the costs of a single drone, a single charging station, a single backup battery, and a single battery swap. The average driving distance within the service area and , This refers to the number of drones per unit area within the jurisdiction. For the occurrence rate of transportation demand; Set constraints, including: Maximum operating intensity constraints of drones ,in, For the occurrence rate of transportation demand, For the service rate of a single drone and , The average loading and unloading time, The average flight speed of the drone. This is the preset upper limit for workload; The probability that a drone will immediately receive a fully charged battery upon arrival at a battery swapping station. The constraint of not being lower than a preset threshold is expressed as: , in, To determine the occurrence rate of battery swapping demand, This is the average charging time. Total number of charging slots Total number of backup batteries, The probability of a charging station being idle is calculated using the following formula: , in, This is a preset probability threshold; Matching constraints between the average battery swapping rate of drones and the rate of transportation demand generation and the average charging rate of charging stations .

[0008] In one embodiment of the present invention, the method for establishing a front-end drone allocation model based on the total number of drones with the goal of minimizing the average response time of the multi-jurisdictional drone service system is as follows: the number of drones configured in each jurisdiction is used as a decision variable, the average response time of the multi-jurisdictional drone service system is used as the objective function, and constraints are set such that each jurisdiction is allocated at least one drone and the workload of drones in each jurisdiction does not exceed a preset upper limit. The front-end drone allocation model is established based on the following assumptions: A1: The transportation demand in each jurisdiction follows a Poisson distribution; A2: The command center processes transportation requests on a first-come, first-served basis, prioritizing the dispatch of the nearest and idle drones. When all drones in the jurisdiction are busy, the cross-regional dispatch mechanism is activated. A3: Calls are lost when all drones are busy, and the probability of loss is the same for all jurisdictions; A4: Drones from the jurisdiction to the jurisdiction The average service time includes cross-regional dispatch time, average travel time, average loading and unloading time, and average travel time to the return station.

[0009] In one embodiment of the present invention, the operational intensity of UAVs in each jurisdiction is calculated using an iterative formula, wherein the iterative formula is: , in, For the jurisdiction The number of drones, For the jurisdiction The rate of occurrence of transportation demand, As a correction factor, For the jurisdiction Relative to the jurisdiction scheduling priority, For the first Workload at priority sites For drones from the jurisdiction to the jurisdiction The average service time; the scheduling priority is determined by the priority matrix. It is confirmed that, among them, The size represents the jurisdiction With jurisdiction The degree of distance.

[0010] In one embodiment of the present invention, the method for obtaining the number of drones configured in each jurisdiction by solving the front-end drone allocation model is as follows: One drone is assigned to each jurisdiction. The remaining drones are then assigned one by one to the jurisdiction that can reduce the average response time of the multi-jurisdiction drone service system the most under the current conditions, until all drones have been assigned. The formula for calculating the average response time of a multi-jurisdictional drone service system is as follows: , in, For drones Heading to the jurisdiction The workload of the service For drones to the jurisdiction The average travel time.

[0011] In one embodiment of the present invention, the method for establishing a back-end battery swapping station allocation model with the goal of minimizing the sum of expected queue lengths of each charging station based on the total number of charging slots and the total number of backup batteries is as follows: the number of charging slots and the number of backup batteries in each jurisdiction are used as decision variables, the sum of expected queue lengths of each charging station is used as the objective function, and constraints are set such that the probability of each charging station having available batteries is not lower than a preset threshold, the total number of charging slots in each jurisdiction is equal to the total number of charging slots, and the total number of backup batteries in each jurisdiction is equal to the total number of backup batteries. The back-end battery swapping station allocation model is established based on the following assumptions: A4: The drone can only go to the charging station for battery loading and unloading after completing the preset number of battery swapping tasks. A5: Treat the charging rack as a server, and the depleted battery as a customer. The charging time of each charging rack follows an exponential distribution.

[0012] In one embodiment of the present invention, the expected queue length of each charging station is calculated using a finite-source queuing model, within the jurisdiction. The expected number of battery batches currently charging and waiting to be charged at the charging station is [value missing]. ,in, The probability of a charging station's state is calculated using the following formula: , in, For the jurisdiction The rate of battery swapping demand This is the average charging time. For the jurisdiction The number of charging slots, For the jurisdiction The number of spare batteries, For the jurisdiction The number of drones; Let be the probability that a charging station is idle; the probability that each charging station has usable batteries is . And must meet ,in, This is a preset threshold for the probability of available batteries.

[0013] In one embodiment of the present invention, in the front-end drone allocation model, if all drones in one jurisdiction are busy, the idle drones in other jurisdictions are sequentially called to provide cross-jurisdictional services in order of increasing distance from the first jurisdiction to the second.

[0014] In one embodiment of the present invention, the back-end battery swapping station allocation model uses a greedy allocation algorithm to solve for the number of charging racks and spare batteries in each jurisdiction. Specifically, each jurisdiction is allocated one charging slot and one spare battery. The remaining charging slots and spare batteries are then allocated one by one to the jurisdiction that can reduce the total expected queue length of each charging station the most in the current state, until all charging slots and spare batteries are allocated.

[0015] Based on the same inventive concept, this invention also provides a multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning system, comprising: The overall planning module is used to establish a planning model with the goal of minimizing the total cost of the multi-jurisdictional drone service system, and solve the planning model under the condition of satisfying the preset operating constraints of the multi-jurisdictional drone service system to obtain the total number of drones, the total number of charging tanks, and the total number of backup batteries. The front-end allocation and back-end allocation modules are used to establish a front-end drone allocation model based on the total number of drones, with the goal of minimizing the average response time of the multi-jurisdiction drone service system, and to solve the front-end drone allocation model to obtain the number of drones configured in each jurisdiction; and to establish a back-end battery swapping station allocation model based on the total number of charging slots and the total number of spare batteries, with the goal of minimizing the sum of the expected queue lengths of each charging station, and to solve the back-end battery swapping station allocation model to obtain the number of charging racks and the number of spare batteries in each jurisdiction. The scheme output module is used to obtain an integrated planning scheme for a multi-jurisdictional drone service system based on the number of drones configured in each jurisdiction, the number of charging racks and the number of backup batteries in each jurisdiction. The integrated planning scheme includes the total number of drones, charging racks and backup batteries, as well as the configuration results of drones, charging racks and backup batteries in each jurisdiction.

[0016] The technical solution of the present invention has the following advantages compared with the prior art: This invention integrates drone transportation scheduling and battery swapping station resource allocation into a unified optimization framework, aiming to minimize the total system cost. It overcomes the shortcomings of existing technologies where front-end transportation capacity and back-end energy replenishment capabilities are disconnected, achieving optimal cost while ensuring service levels. By introducing the Poisson distribution assumption of transportation demand, the first-come-first-served principle, cross-regional scheduling mechanisms, and differentiated service time models, it accurately calculates scheduling probabilities and workloads based on priority matrices and iterative formulas. This accurately reflects the differences in order intensity, service distance, and response requirements across different jurisdictions, achieving scientific scheduling of cross-regional support and balancing the operational load of each jurisdiction. By setting a constraint that the probability of a drone immediately obtaining a fully charged battery upon arrival at a battery swapping station is no less than a preset threshold, and using a finite-source queuing model to accurately calculate this probability, it effectively avoids service interruptions due to a lack of available batteries during peak periods, significantly improving system reliability and user experience. Following a hierarchical process, it outputs the total number of drones, charging stations, and spare batteries, along with their specific configuration schemes in each jurisdiction, directly serving the infrastructure investment, equipment procurement, and operational scheduling of urban drone logistics systems, demonstrating strong engineering practicality. Attached Figure Description

[0017] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0018] Figure 1 This is a flowchart illustrating the multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning method provided in this embodiment of the invention. Detailed Implementation

[0019] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.

[0020] Example 1: like Figure 1 As shown, the present invention provides a method for integrated planning of a multi-jurisdictional unmanned aerial vehicle (UAV) service system, comprising the following steps: A planning model is established with the goal of minimizing the total cost of a multi-jurisdictional drone service system. The total cost of the multi-jurisdictional drone service system includes the fixed cost of drones, the fixed cost of charging stations, the fixed cost of backup batteries, and the operational cost. The planning model is solved under the preset operational constraints of the multi-jurisdictional drone service system to obtain the total number of drones, the total number of charging stations, and the total number of backup batteries. Based on the total number of drones, a front-end drone allocation model is established with the goal of minimizing the average response time of the multi-jurisdictional drone service system. The number of drones configured in each jurisdiction is obtained by solving the front-end drone allocation model. Based on the total number of charging slots and the total number of spare batteries, a back-end battery swapping station allocation model is established with the goal of minimizing the total expected queue length of each charging station. The number of charging racks and the number of spare batteries in each jurisdiction are obtained by solving the back-end battery swapping station allocation model. Based on the number of drones configured in each jurisdiction, as well as the number of charging racks and spare batteries in each jurisdiction, an integrated planning scheme for a multi-jurisdictional drone service system is obtained. The integrated planning scheme includes the total number of drones, charging racks, and spare batteries, as well as the configuration results of drones, charging racks, and spare batteries in each jurisdiction.

[0021] This invention establishes a planning model aimed at minimizing the total cost of a multi-jurisdictional drone service system. Under constraints such as maximum drone workload, probability of fully charged batteries, and front-end / back-end rate matching, the total number of drones, charging stations, and backup batteries is calculated. Based on this total, a front-end drone allocation model is established with the goal of minimizing average response time. By introducing assumptions such as Poisson distribution of transport demand, first-come-first-served principle, cross-regional scheduling mechanism, and differentiated service time, the number of drones configured in each jurisdiction is accurately calculated using iterative formulas and priority matrices. Based on the total number of charging stations and backup batteries, a back-end battery swapping station allocation model is established with the goal of minimizing the sum of expected queue lengths at each charging station. Under a finite-source queuing model, the number of charging racks and backup batteries in each jurisdiction is calculated, thus forming an integrated planning scheme that includes both total quantity and jurisdiction-level configuration. By integrating front-end transportation scheduling and back-end battery swapping and replenishment into a unified optimization framework, this approach overcomes the shortcomings of existing methods, such as the separation of front-end and back-end, insufficient consideration of regional differences, and a crude cross-regional scheduling mechanism. It introduces a probability constraint on available fully charged batteries, effectively avoiding service interruptions caused by the lack of available batteries during peak periods. At the same time, through a hierarchical planning process, it outputs infrastructure investment and operation scheduling solutions that can be directly implemented, reducing the total system cost and average response time, balancing the operating load of each region, and improving the overall system reliability and engineering practicality.

[0022] In one specific embodiment of the present invention, a planning model is first established with the objective of minimizing the total cost of a multi-jurisdictional drone service system. The planning model takes a holistic view of the multi-jurisdictional drone service system, comprehensively considering the fixed resource costs and operational costs of front-end transportation equipment and back-end charging facilities, aiming to minimize the total system cost while meeting service performance requirements. Since urban planning and land use have a decisive impact on jurisdictional division and battery swapping station locations, in this embodiment of the invention, it is assumed that the jurisdictional division and the location of battery swapping stations within each jurisdiction are predetermined. Therefore, the key resources that the multi-jurisdictional drone service system needs to decide on are the total number of drones, the total number of charging slots, and the total number of spare batteries.

[0023] Specifically, the objective function is constructed as follows: , in, For the total number of drones, Total number of charging slots Total number of backup batteries; , , , These represent the fixed costs of a single drone, a single charging slot, a single backup battery, and the operational cost of a single battery swap, respectively. The average driving distance within the service area is determined by the number of drones per unit area of ​​the service area. The decision is made, and the calculation formula is as follows: ; Let be the transportation demand occurrence rate. The objective function consists of four terms: the first term is the fixed cost of the drone fleet, the second is the fixed cost of the charging station facilities, the third is the fixed cost of backup batteries (including batteries carried by the drones themselves), and the fourth is the operating cost determined by the frequency of battery swapping, service distance, and unit price. By minimizing this objective function, economic optimization can be achieved while satisfying system operating constraints, that is, achieving the required transportation service capacity with minimal resource input.

[0024] In this embodiment of the invention, to ensure the safe and efficient operation of the unmanned aerial vehicle (UAV) transportation system, the following three types of core constraints are set.

[0025] The first type of constraint is the maximum workload constraint for the drone. Workload Defined as the ratio of transportation demand occurrence rate to the system's maximum service capacity, i.e. ,in For the occurrence rate of transportation demand, Service rate for a single drone. Compared with average loading and unloading time Average driving distance and the average flight speed of drones Related, the specific expression is The constraint requirement for the maximum operating strength of the drone is as follows: , This is a preset upper limit for workload. The purpose of the maximum workload constraint for drones is to prevent drones from operating in an oversaturated state for extended periods, ensuring flight safety and the long-term stability of the system. If the workload exceeds the upper limit, it means that the drone fleet will be unable to respond to all transport requests in a timely manner, leading to a sharp increase in call loss rate and a serious decline in service quality.

[0026] The second constraint is that the probability of the drone immediately obtaining a fully charged battery upon arriving at the battery swapping station is not less than a preset threshold. Let the probability of the drone immediately obtaining a fully charged battery upon arriving at the battery swapping station be denoted as... The physical meaning is the possibility that drones can be directly replaced with fully charged batteries without having to wait in line after returning to the station.

[0027] The finite-source queuing model is a classic model in queuing theory, used to describe queuing systems where the total number of customers is finite and the arrival rate of each customer is affected by the number of customers in the system. In classical queuing theory, the finite-source model is usually denoted as... Among them, the customer base is limited (e.g.) (A finite-source queuing model is used to analyze system performance under a finite user group.) Each customer generates demand at a certain rate when idle. No new demand arrives when all customers are in a "served" state.

[0028] In this embodiment of the invention, the spare batteries in the battery swapping station are considered as a finite customer source (the total number of batteries is fixed), and the charging racks are considered as service counters. The process of depleted batteries arriving at the charging station to request charging is a typical finite-source queuing process. Therefore, the steady-state probability formula of this model can be directly applied to describe the state distribution of the charging station. Based on the finite-source queuing model, in this embodiment of the invention, the following is derived: The parsing expression is: , in, To determine the occurrence rate of battery swapping demand, This is the average charging time. Total number of charging slots Total number of backup batteries, This indicates the number of battery batches currently charging and waiting to be charged at the charging station (each batch corresponds to one battery). The probability that a charging station is idle is the probability that no battery is charging or waiting to be charged at the charging station. The formula for calculating this probability is: , The requirement of this constraint is: ,in This is a preset probability threshold. By explicitly introducing this constraint, it is possible to effectively avoid situations where drones are grounded or even experience service interruptions during peak hours due to a lack of available fully charged batteries, significantly improving the system's continuous operation capability and user experience.

[0029] The third type of constraint is the matching constraint between the average battery swapping rate of drones, the rate of transportation demand generation, and the average charging rate of charging stations. This constraint ensures that front-end transportation demand and back-end energy replenishment capacity are coordinated, avoiding an imbalance where there is excess capacity but insufficient energy replenishment, or vice versa. Its expression is: , in, The system must be able to generate no more transportation tasks than the maximum service rate that the drone fleet can provide. The requirement is that the rate of battery swapping demand generated by the drone fleet during missions cannot exceed the maximum charging rate of the charging station, the latter being determined by the total number of charging slots. and the average charging rate of each charging slot A joint decision. If This is not true, indicating that there are not enough drones to handle all transportation needs in a timely manner; if If this is not the case, it means that there are not enough charging slots, which cannot replenish the depleted battery in time, and will eventually cause the drone to stop flying due to lack of power.

[0030] Under the premise of satisfying the above objective function and all constraints, this embodiment uses an appropriate integer programming algorithm to solve the overall programming model, thereby obtaining the total number of UAVs. Total number of charging slots Total number of backup batteries The optimal values ​​are determined. These totals will serve as input parameters for the subsequent front-end drone allocation model and back-end battery swapping station allocation model, thus forming a hierarchical planning process of "overall scale determination - front-end drone allocation - back-end battery swapping station allocation". This ensures a macro-level match between the total global resources and service demand, while providing feasible boundary conditions for subsequent refined regional resource allocation.

[0031] Furthermore, the total number of drones is obtained through an overall planning model. Next, these drones need to be rationally allocated to various jurisdictions to minimize the system's average response time while meeting service performance requirements. Therefore, in this embodiment of the invention, a front-end drone allocation model is established with the objective of minimizing the average response time of the multi-jurisdiction drone service system. The front-end drone allocation model configures the number of drones in each jurisdiction. (in , Using the total number of districts as the decision variable, the summation constraint is satisfied. The optimization objective is to improve the average response time of the multi-jurisdictional drone service system. Minimize, i.e. At the same time, two constraints are set: at least one drone must be allocated to each jurisdiction. ), and the workload of drones in each jurisdiction It must not exceed the preset upper limit. This is to prevent drones from operating under overload conditions for extended periods.

[0032] The formula for calculating the system's average response time is: , in, This indicates that, under a given drone allocation scheme, the number of drones... Heading to the jurisdiction The workload of the service This represents the corresponding average travel time.

[0033] In order to accurately characterize the interaction between different jurisdictions in the front-end transportation service process, a front-end service model was established based on queuing theory, and the following four basic assumptions were introduced.

[0034] Assumption A1: The transportation demand in each district follows a Poisson distribution. Let the district be... to the jurisdiction The rate of transportation demand occurrence Then the jurisdiction Total transportation demand rate The total transportation demand occurrence rate of the system is The Poisson distribution assumption allows the randomness of transportation demand to be described using classical queuing theory models, and the arrival process is memoryless.

[0035] Assumption A2: The command center processes transport requests using a First Come First Service (FCFS) principle, prioritizing the dispatch of the nearest and idle drones. When all drones within a certain jurisdiction are busy, a cross-jurisdictional dispatch mechanism is activated, allowing idle drones from other jurisdictions to respond to the service request. This mechanism ensures the sharing of system resources and can alleviate localized demand peaks.

[0036] Assumption A3: When all drones are busy, newly generated calls will be lost. Let the probability of loss be... Furthermore, the probability of call loss is the same across all jurisdictions; call arrival rates are unaffected by drone busy status. This assumption aims to simplify system analysis. Queuing models are used to describe the service process in each jurisdiction.

[0037] Assumption A4: To reflect the service differences between different jurisdictions, assume that the drone departs from the jurisdiction... to the jurisdiction Average service time This includes cross-regional dispatch time, average travel time, average loading and unloading time, and average travel time to the return station.

[0038] Specifically, ,in To be from the jurisdiction to the jurisdiction One-way trip time, For loading and unloading time. If (i.e., services within this jurisdiction), then When providing services across regions, The calculation is performed by dividing the distance between the centroids of different jurisdictions by the flight speed. This assumption allows the model to capture the differences in service time caused by geographical location between different jurisdictions, thus more realistically reflecting actual operations.

[0039] Under the above assumptions, the entire front-end transportation system can be abstracted as a queuing network with multiple service stations and multiple queues. Each jurisdiction is considered a service station, and the number of drones within it is the number of service stations (i.e., the number of parallel servers). Since transportation demands arrive randomly and drone service times follow an exponential distribution, the service process for each jurisdiction can be described using... An approximate description of a queuing system, wherein The number of drones in this jurisdiction. This indicates that the system capacity equals the number of servers. According to queuing theory, in the system... The drone is in the air or The probabilities of each drone being in a busy state are as follows: , , in, The workload of a single drone, and its relationship with the total system transport demand rate and service rate are as follows: , For the service rate of a single drone, This represents the total workload of the jurisdiction.

[0040] definition For the jurisdiction When a transportation request is issued, it shall be handled by the jurisdiction. The probability that a drone will respond, i.e., the scheduling probability. According to Little's formula in queuing theory, the jurisdiction... The expected number of busy drones is equal to the product of the transportation demand occurrence rate in the jurisdiction and the average service time, expressed as: , in, For the jurisdiction The number of drones, For the jurisdiction The average workload (i.e., the average busy probability of a single drone). Divide the above by The iterative relationship for workload is obtained as follows: .

[0041] In this embodiment of the invention, in order to solve the scheduling probability Further consideration is needed regarding the priority order of cross-regional scheduling, and a priority matrix should be introduced. To indicate the jurisdiction Relative to the jurisdiction Scheduling priority: Indicates jurisdiction With jurisdiction Closest distance Indicates the next closest, and so on. When the jurisdiction... When a transportation demand is generated, the system will follow The system attempts to call up available drones in the corresponding jurisdictions in ascending order of priority. Once a jurisdiction has an available drone, it will respond, and lower priority jurisdictions will no longer be considered.

[0042] Therefore, scheduling probability In reality: when At that time, before All priority jurisdictions are busy and the first The probability that there is at least one idle drone in each priority jurisdiction. When calculating the scheduling probability, the mutual influence between multiple drone base stations needs to be considered. Therefore, the "rescue vehicle independence assumption" is introduced, which assumes that the busy status of each drone base station is independent of each other; whether one base station is busy or not does not affect the scheduling probability of other base stations. A correction factor is also introduced. To correct for the correlation in the case of multiple drones. Based on this assumption, for the case where only one drone is configured in each jurisdiction, the scheduling probability can be approximated as: For the simple case where there is only one drone in each jurisdiction, the scheduling probability can be approximated as: , in, It is relative to the jurisdiction The Work intensity in priority jurisdictions It is the jurisdiction Work intensity. Correction factor. The expression is: , in, It is the probability of loss (i.e., the probability that all drones are busy), by The calculation and correction factor corrects for the bias caused by neglecting correlation in the simple product form by taking into account the distribution of the number of busy drones in the system.

[0043] When multiple drones may be deployed in each jurisdiction, the above formula needs to be generalized. Let the formula be relative to the jurisdiction. The The number of drones owned by the priority jurisdiction is ,forward The total number of drones in each priority jurisdiction is Then all the preceding The probability that all priority jurisdictions are busy should be taken as follows: This is because each jurisdiction needs all its drones to be busy for it to be considered that the jurisdiction has no idle drones. And the... The probability that there is at least one idle drone in each priority jurisdiction is ,in It is the jurisdiction The average workload. The approximate formula for the scheduling probability after generalization is: , Among the correction factors Depends on the drone configuration set of each priority jurisdiction subscript This indicates that the factor is relevant to the target jurisdiction. The above... Substituting into the work intensity iteration formula, we get: , Scheduling priority .

[0044] To solve this system of equations, an initial workload is first given. A new workload is then calculated using the formula described above, and this process is repeated iteratively until the change is less than a preset tolerance. During the iteration process, a correction factor is used. The correction factor needs to be recalculated based on the current workload and the number of drones in each jurisdiction. The detailed expression for the correction factor can be derived using the law of total probability and conditional probability.

[0045] Specifically, suppose the system has a total of Launching drones, Indicates the first Priority events where all drones in the jurisdiction are busy. Indicates the first The priority zone has at least one idle drone. The scheduling probability is... According to the law of total probability, the total number of busy drones in the system... Under the following conditions: , in, Indicates that there are a total of The busy events involving drones This is given by the queuing theory formula. Given... Under the conditions, the former The probability that all priority jurisdictions are busy is: (when (time), and the first The probability that at least one aircraft is available in each priority jurisdiction is: (in Finally, the correction factor can be obtained. The closing expression: , in This represents the average workload. In actual calculations, correction factor tables for different parameters can be pre-calculated, or numerical integration can be used as an approximation.

[0046] Using the iterative formula described above, given the number of drones in each jurisdiction... In this case, the workload of each jurisdiction can be calculated. This allows for the calculation of the system's average response time. Since the response time... It can also be directly derived from the workload.

[0047] Furthermore, it is necessary to solve the front-end allocation model, that is, given the total number of drones... Given the distribution of transportation demand in each jurisdiction, find a set of Make Minimum.

[0048] In this embodiment of the invention, a greedy allocation algorithm is used to solve the problem: one drone is allocated to each jurisdiction, i.e. For all The number of drones allocated at this time The remaining number of unassigned drones is .

[0049] when At that time, each jurisdiction was inspected in turn. Calculate the change in the system's average response time if one more drone is added to the jurisdiction. Since adding drones typically reduces response time, therefore Negative values ​​are selected. The jurisdiction that minimizes the response time is chosen. smallest jurisdiction Assign a drone to the jurisdiction to update , Then recalculate the workload and average system response time for each jurisdiction. Repeat the above process until all drones are assigned. Return to the final... This refers to the number of drones deployed in each jurisdiction.

[0050] When resources are limited, allocating resources to the most scarce jurisdiction at each step yields the greatest performance improvement. Greedy algorithms can achieve near-optimal solutions in multi-jurisdictional drone allocation problems, and their computational efficiency is far superior to precise algorithms such as branch and bound.

[0051] Meanwhile, in actual operation, it is also necessary to clarify the specific rules for cross-regional dispatching at the front end. Specifically, when all drones within a certain jurisdiction are busy, the system will sequentially call upon idle drones from other jurisdictions for cross-regional service, according to the order of distance from the current jurisdiction to the nearest, and so on. In the actual dispatching system, the command center only needs to maintain a distance sorting table. When there are no idle drones in its own jurisdiction, it queries neighboring jurisdictions for idle drones in descending order of priority. If there are, they are dispatched immediately; otherwise, it continues to query the next priority level.

[0052] By using the front-end drone allocation model and solution algorithm, the number of drones configured in each jurisdiction can be obtained. .

[0053] Furthermore, the total number of charging slots is obtained through an overall planning model. Total number of backup batteries The number of drones configured in each jurisdiction is obtained through a front-end allocation model. Then, these charging slots and backup batteries need to be rationally allocated to the battery swapping stations in each jurisdiction, and the expected queuing length of each charging station should be shortened as much as possible while meeting the requirements for energy replenishment reliability.

[0054] Therefore, in this embodiment of the invention, a back-end battery swapping station allocation model is established with the objective of minimizing the total expected queue length of each charging station. The back-end battery swapping station allocation model allocates the number of charging slots in each jurisdiction. and number of spare batteries As decision variables, among which, To meet the total quantity constraints and The optimization objective is to minimize the sum of the expected queue lengths for all charging stations, i.e. ,in Indicates jurisdiction The expected number of battery batches charging and waiting to be charged in a charging station; with the constraint that the probability of having available batteries at each charging station is not less than a preset threshold. Furthermore, the number of charging slots cannot be less than a basic minimum (usually taken as...). Furthermore, the number of charging slots cannot exceed the number of spare batteries. This ensures that each charging slot is equipped with at least one spare battery, thereby maintaining the efficient use of the charging rack.

[0055] To accurately depict the dynamic process of battery charging and queuing within the back-end battery swapping stations, a supply model was established based on a finite-source queuing model. When the battery carried by the drone runs out of power, the drone returns to the local battery swapping station, removes the depleted battery, immediately replaces it with a fully charged battery, and then continues its transport mission; the removed depleted battery then enters the charging station to await charging. Each battery swapping station is equipped with... One charging station (i.e., service counter) and A spare battery (excluding the battery currently in use on the drone). The total number of drones within the jurisdiction is... Each drone carries one battery, therefore the total number of batteries served by this battery swapping station is [number missing]. ,in Block batteries are being used in drones. The batteries are rotated within the station.

[0056] In this embodiment of the invention, it is assumed that the drone only needs to be charged after completing several battery swapping tasks. However, for the sake of simplifying the analysis, the drone's battery swapping needs can be considered as arriving at the charging station at a certain rate. (Note: The last part, "responsibility area," appears to be an unrelated fragment and has been omitted from the translation.) The occurrence rate of battery swapping demand is It depends on the volume of drone transport tasks in the jurisdiction, which can be calculated from the transport demand occurrence rate and average service time in the front-end model. The average time for each charging rack to fully charge a depleted battery is... Charging time follows an exponential distribution, therefore the service rate of a single charging rack is... The total service rate of the entire charging station is .

[0057] The number of battery batches currently charging and waiting to be charged at the charging station is recorded as follows: Since the total number of batteries in the system is fixed, and each battery is either being used on the drone, charging at the station, or waiting to be charged, this charging process constitutes a finite-source queuing system, which can be modeled as follows: The total number of customers According to the finite-source queuing theory, the charging station is in a state of... steady-state probability Given by the following formula: when hour, ;when hour, .in This represents the probability that a charging station is idle, i.e., the probability that no batteries are charging or waiting at the station. The expression is: If ,but ;like ,but .

[0058] These formulas are standard results of the finite-source queuing model, used to describe the probability distribution of each state when the system is in equilibrium.

[0059] The expected number of battery batches currently charging and waiting to be charged at a charging station is the expected queue length, calculated using the following formula: This expected value directly reflects the level of congestion at charging stations. The larger the value, the more batteries are in the queue or charging, meaning the drone may have to wait a long time to get a fully charged battery after it arrives.

[0060] To ensure continuous operation of the drone without interruption due to power shortages, this embodiment introduces a probability constraint on the availability of fully charged batteries. This constraint applies when the number of fully charged batteries in the charging station is less than the total number of spare batteries. At that time, that is This means that at least one fully charged battery is available for replacement; conversely, if... If all backup batteries are either charging or in a standby state, then no fully charged battery is available. Therefore, the probability that a charging station has a usable battery is... According to the state probability formula, the probability can be specifically expressed as: , in, , A preset available battery probability threshold is set to ensure that when the drone arrives at the battery swapping station, there is a sufficiently high probability that it can immediately obtain a fully charged battery, thereby avoiding service interruption.

[0061] After obtaining the number of drones in each jurisdiction Total number of charging slots Total number of backup batteries Subsequently, the back-end battery swapping station allocation model needs to solve for the constraint that satisfies the above conditions and minimizes the total expected queue length. and .

[0062] In this embodiment of the invention, since the problem is an integer nonlinear programming problem with many decision variables, a greedy allocation algorithm is used to solve it: First, a charging tank and a spare battery are allocated to each jurisdiction, i.e., initialization. , The number of charging slots allocated at this time is: The number of backup batteries is The remaining number of charging slots to be allocated is The number of remaining backup batteries is Then, the main loop begins. While there are still unallocated charging slots or backup batteries, it sequentially calculates the total expected queue length if a charging slot is added to a certain jurisdiction (while keeping the backup battery unchanged) or if a backup battery is added (while keeping the charging slot unchanged). The change in resource allocation. Since increasing resources typically reduces queue length, the adjustment scheme that maximizes the reduction in the total expected queue length is selected; that is, the district and resource type with the largest reduction are allocated. Only one charging slot or one spare battery is allocated at a time, and the corresponding district is updated. or Then, recalculate the expected queue length and available battery probability for each charging station to ensure that the constraints are always met. Repeat the above process until all charging slots and backup batteries are allocated. Finally, output the number of charging slots for each jurisdiction. and number of spare batteries .

[0063] The allocation process always satisfies as well as If an allocation results in a jurisdiction failing to meet the constraints, the allocation scheme is abandoned, and other jurisdictions or other resource types are tried instead. By utilizing the convexity of the objective function and the diminishing marginal returns, a near-optimal allocation scheme can be obtained in polynomial time, which has high engineering practicality.

[0064] Using the aforementioned back-end battery swapping station allocation model and solution algorithm, the number of charging racks and spare batteries in each jurisdiction are obtained. These results are then combined with the total number of drones, charging slots, and spare batteries output from the overall planning model, as well as the number of drones configured in each jurisdiction output from the front-end allocation model, to form a complete multi-jurisdictional drone service system integrated planning scheme. This scheme can be directly used to guide the procurement and deployment of drones, charging slots, and spare batteries in the actual system, as well as the fine-tuning of resources in each jurisdiction.

[0065] Example 2: Based on the same inventive concept as Embodiment 1, the present invention also provides a multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning system, comprising: The overall planning module is used to establish a planning model with the goal of minimizing the total cost of the multi-jurisdictional drone service system, and solve the planning model under the condition of satisfying the preset operating constraints of the multi-jurisdictional drone service system to obtain the total number of drones, the total number of charging tanks, and the total number of backup batteries. The front-end allocation and back-end allocation modules are used to establish a front-end drone allocation model based on the total number of drones, with the goal of minimizing the average response time of the multi-jurisdiction drone service system, and to solve the front-end drone allocation model to obtain the number of drones configured in each jurisdiction; and to establish a back-end battery swapping station allocation model based on the total number of charging slots and the total number of spare batteries, with the goal of minimizing the sum of the expected queue lengths of each charging station, and to solve the back-end battery swapping station allocation model to obtain the number of charging racks and the number of spare batteries in each jurisdiction. The scheme output module is used to obtain an integrated planning scheme for a multi-jurisdictional drone service system based on the number of drones configured in each jurisdiction, the number of charging racks and the number of backup batteries in each jurisdiction. The integrated planning scheme includes the total number of drones, charging racks and backup batteries, as well as the configuration results of drones, charging racks and backup batteries in each jurisdiction.

[0066] The overall planning module is used to establish a planning model with the objective of minimizing the total cost of a multi-jurisdictional drone service system. It solves this model under preset operational constraints to obtain the total number of drones, charging stations, and backup batteries. The specific implementation of this module is the same as the process of establishing and solving the overall planning model in Example 1, and will not be repeated here.

[0067] The front-end allocation and back-end allocation modules further include a front-end drone allocation unit and a back-end battery swapping station allocation unit. The front-end drone allocation unit, based on the total number of drones output by the overall planning module, establishes a front-end drone allocation model with the objective of minimizing the system's average response time, and solves for the number of drones configured in each jurisdiction using a greedy allocation algorithm. The specific modeling process, assumptions, workload iteration formula, and solution algorithm of this unit are consistent with the description of the front-end drone allocation model in Example 1. The back-end battery swapping station allocation unit, based on the total number of charging slots and spare batteries output by the overall planning module, and the number of drones configured in each jurisdiction output by the front-end allocation unit, establishes a back-end battery swapping station allocation model with the objective of minimizing the sum of the expected queue lengths of each charging station. It uses a finite-source queuing model to calculate the state probability and expected queue length of each charging station, and solves for the number of charging racks and spare batteries in each jurisdiction using a greedy allocation algorithm. The specific calculation formulas, constraints, and solution steps of this unit are the same as the description of the back-end battery swapping station allocation model in Example 1.

[0068] The solution output module receives the number of drones, charging racks, and spare batteries for each jurisdiction from the front-end and back-end allocation modules. It then combines this information with the total number of drones, charging racks, and spare batteries output by the overall planning module to generate a complete integrated planning solution for a multi-jurisdictional drone service system. This solution includes the total global resources and detailed configuration results for each jurisdiction, and can be directly used to guide the procurement, deployment, and operational scheduling of drones, charging racks, and spare batteries in the actual system.

[0069] Through the collaborative work of the above system modules, the entire planning process, from determining the overall resource scale to allocating resources at the jurisdiction level, can be completed automatically, significantly reducing the complexity and time cost of manual decision-making, while ensuring the scientific nature and engineering practicality of the planning results.

[0070] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0071] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0072] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0073] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0074] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A multi-jurisdictional drone service system integration planning method, characterized by, include: A planning model is established with the goal of minimizing the total cost of a multi-jurisdictional drone service system. The total cost of the multi-jurisdictional drone service system includes the fixed cost of drones, the fixed cost of charging stations, the fixed cost of backup batteries, and the operational cost. The planning model is solved under the preset operational constraints of the multi-jurisdictional drone service system to obtain the total number of drones, the total number of charging stations, and the total number of backup batteries. Based on the total number of drones, a front-end drone allocation model is established with the goal of minimizing the average response time of the multi-jurisdictional drone service system. The number of drones configured in each jurisdiction is obtained by solving the front-end drone allocation model. Based on the total number of charging slots and the total number of spare batteries, a back-end battery swapping station allocation model is established with the goal of minimizing the total expected queue length of each charging station. The number of charging racks and the number of spare batteries in each jurisdiction are obtained by solving the back-end battery swapping station allocation model. Based on the number of drones configured in each jurisdiction, as well as the number of charging racks and spare batteries in each jurisdiction, an integrated planning scheme for a multi-jurisdictional drone service system is obtained. The integrated planning scheme includes the total number of drones, charging racks, and spare batteries, as well as the configuration results of drones, charging racks, and spare batteries in each jurisdiction.

2. The multi-jurisdictional drone service system integration planning method of claim 1, wherein: The method for establishing a planning model with the objective of minimizing the total cost of a multi-jurisdictional unmanned aerial vehicle (UAV) service system is as follows: Construct the objective function ,in, For the total number of drones, Total number of charging slots Total number of backup batteries, , , , These represent the costs of a single drone, a single charging station, a single backup battery, and a single battery swap. The average driving distance within the service area and , This refers to the number of drones per unit area within the jurisdiction. For the occurrence rate of transportation demand; Set constraints, including: Maximum operating intensity constraints of drones ,in, For the occurrence rate of transportation demand, For the service rate of a single drone and , The average loading and unloading time, The average flight speed of the drone. This is the preset upper limit for workload; The probability that a drone will immediately receive a fully charged battery upon arrival at a battery swapping station. The constraint of not being lower than a preset threshold is expressed as: ， in, To determine the occurrence rate of battery swapping demand, This is the average charging time. Total number of charging slots Total number of backup batteries, The probability of a charging station being idle is calculated using the following formula: ， in, This is a preset probability threshold; Matching constraints between the average battery swapping rate of drones and the rate of transportation demand generation and the average charging rate of charging stations .

3. The multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning method according to claim 1, characterized in that: Based on the total number of drones, the method for establishing a front-end drone allocation model with the goal of minimizing the average response time of the multi-jurisdictional drone service system is as follows: the number of drones configured in each jurisdiction is used as a decision variable, the average response time of the multi-jurisdictional drone service system is used as the objective function, and constraints are set such that each jurisdiction is allocated at least one drone and the workload of drones in each jurisdiction does not exceed a preset upper limit. The front-end drone allocation model is established based on the following assumptions: A1: The transportation demand in each jurisdiction follows a Poisson distribution; A2: The command center processes transportation requests on a first-come, first-served basis, prioritizing the dispatch of the nearest and idle drones. When all drones in the jurisdiction are busy, the cross-regional dispatch mechanism is activated. A3: Calls are lost when all drones are busy, and the probability of loss is the same for all jurisdictions; A4: Drones from the jurisdiction to the jurisdiction The average service time includes cross-regional dispatch time, average travel time, average loading and unloading time, and average travel time to the return station.

4. The multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning method according to claim 3, characterized in that: The workload of drones in each jurisdiction is calculated using an iterative formula, which is: ， in, For the jurisdiction The number of drones, For the jurisdiction The rate of occurrence of transportation demand, As a correction factor, For the jurisdiction Relative to the jurisdiction scheduling priority, For the first Workload at priority sites For drones from the jurisdiction to the jurisdiction The average service time; the scheduling priority is determined by the priority matrix. It is confirmed that, among them, The size represents the jurisdiction With jurisdiction The degree of distance.

5. The multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning method according to claim 1, characterized in that: The method for solving the aforementioned front-end drone allocation model to obtain the number of drones configured in each jurisdiction is as follows: One drone is assigned to each jurisdiction. The remaining drones are then assigned one by one to the jurisdiction that can reduce the average response time of the multi-jurisdiction drone service system the most under the current conditions, until all drones have been assigned. The formula for calculating the average response time of a multi-jurisdictional drone service system is as follows: ， in, For drones Heading to the jurisdiction The workload of the service For drones to the jurisdiction The average travel time.

6. The multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning method according to claim 1, characterized in that: Based on the total number of charging slots and the total number of backup batteries, the method for establishing a back-end battery swapping station allocation model with the goal of minimizing the total expected queue length of each charging station is as follows: the number of charging slots and the number of backup batteries in each jurisdiction are used as decision variables, the sum of the expected queue lengths of each charging station is used as the objective function, and constraints are set such that the probability of each charging station having available batteries is not lower than a preset threshold, the total number of charging slots in each jurisdiction is equal to the total number of charging slots, and the total number of backup batteries in each jurisdiction is equal to the total number of backup batteries. The back-end battery swapping station allocation model is established based on the following assumptions: A4: The drone can only go to the charging station for battery loading and unloading after completing the preset number of battery swapping tasks. A5: Treat the charging rack as a server, and the depleted battery as a customer. The charging time of each charging rack follows an exponential distribution.

7. The multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning method according to claim 6, characterized in that: The expected queue length at each charging station is calculated using a finite-source queuing model within the jurisdiction. The expected number of battery batches currently charging and waiting to be charged at the charging station is [value missing]. ,in, The probability of a charging station's state is calculated using the following formula: ， in, For the jurisdiction The rate of battery swapping demand This is the average charging time. For the jurisdiction The number of charging slots, For the jurisdiction The number of spare batteries, For the jurisdiction The number of drones; Let be the probability that a charging station is idle; the probability that each charging station has usable batteries is . And must meet ,in, This is a preset threshold for the probability of available batteries.

8. The multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning method according to claim 1, characterized in that: In the aforementioned front-end drone allocation model, if all drones in one jurisdiction are busy, the idle drones in other jurisdictions are sequentially called to provide cross-jurisdictional services in order of increasing distance from the current jurisdiction to the previous jurisdiction.

9. The multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning method according to claim 1, characterized in that: In the back-end battery swapping station allocation model, a greedy allocation algorithm is used to solve for the number of charging racks and spare batteries in each jurisdiction. The specific method is as follows: allocate one charging slot and one spare battery to each jurisdiction, and then allocate the remaining charging slots and spare batteries one by one to the jurisdiction that can reduce the total expected queue length of each charging station the most in the current state, until all charging slots and spare batteries are allocated.

10. A multi-jurisdictional unmanned aerial vehicle (UAV) service system integration planning system, characterized in that, include: The overall planning module is used to establish a planning model with the goal of minimizing the total cost of the multi-jurisdictional drone service system, and solve the planning model under the condition of satisfying the preset operating constraints of the multi-jurisdictional drone service system to obtain the total number of drones, the total number of charging tanks, and the total number of backup batteries. The front-end allocation and back-end allocation modules are used to establish a front-end drone allocation model based on the total number of drones, with the goal of minimizing the average response time of the multi-jurisdiction drone service system, and to solve the front-end drone allocation model to obtain the number of drones configured in each jurisdiction; and to establish a back-end battery swapping station allocation model based on the total number of charging slots and the total number of spare batteries, with the goal of minimizing the sum of the expected queue lengths of each charging station, and to solve the back-end battery swapping station allocation model to obtain the number of charging racks and the number of spare batteries in each jurisdiction. The scheme output module is used to obtain an integrated planning scheme for a multi-jurisdictional drone service system based on the number of drones configured in each jurisdiction, the number of charging racks and the number of backup batteries in each jurisdiction. The integrated planning scheme includes the total number of drones, charging racks and backup batteries, as well as the configuration results of drones, charging racks and backup batteries in each jurisdiction.

Images