Method, device and medium for taking and delivering goods based on cooperation of dual-role unmanned aerial vehicle and vehicle

Abstract

The present application relates to a kind of based on dual role unmanned aerial vehicle and vehicle cooperation's pick-up and delivery path planning method, equipment and medium, comprising: at least one vehicle and multiple unmanned aerial vehicles are configured in the space region corresponding to task point, unmanned aerial vehicle is configured to execute task and / or parcel transfer between vehicles, constructs delivery task graph;All pure pick-up point and pure delivery point are assigned to vehicle, constructs vehicle initial path;Pair of pick-up and delivery task point is inserted into vehicle path;Task point that satisfies the energy consumption constraint of unmanned aerial vehicle is changed by unmanned aerial vehicle service, constructs greedy initial solution;Greedy initial feasible solution is encoded as the four-tuple of each task point, corresponding task point number, unmanned aerial vehicle label, parcel transfer index and take-off point index respectively;Optimization is carried out by cyclically calling neighborhood search operator, and optimal planning path is decoded and output after meeting termination condition.Compared with prior art, the present application considers the efficiency and economy demand of pick-up and delivery path solution.

Description

Technical Field

[0001] This invention relates to the field of pickup and delivery route planning, and in particular to a pickup and delivery route planning method, device and medium based on dual-role drone and vehicle collaboration. Background Technology

[0002] The Pick-up and Delivery Problem (PDP) requires completing three types of tasks while minimizing costs: pick-up-only tasks, delivery-only tasks, and paired pick-up-delivery tasks. For paired tasks, packages collected from the pick-up point must be delivered to their corresponding delivery point, a prerequisite constraint that makes the PDP more complex than the pure delivery problem. With the rapid development of e-commerce and instant delivery, the need for efficient PDP solutions is becoming increasingly urgent.

[0003] Drones have garnered significant attention in logistics and delivery due to their speed and immunity to traffic congestion. However, drones are limited by payload and battery capacity, restricting their flight range and allowing only a limited number of packages to be carried per flight. They must return to a refueling point to recharge and replenish their stock after completing deliveries or running out of power. Complementing this is the advantage of vehicles, which offer long-range and high-payload transport, but are slower and more susceptible to traffic congestion. Researchers have therefore proposed a collaborative model using vehicles as both mobile warehouses and charging stations for drones, leveraging the complementary strengths of both.

[0004] In last-mile delivery operations, the primary cost driver is not the distance traveled, but rather the fixed costs of each deployed vehicle-driver unit, including vehicle insurance, driver salaries, and ongoing recruitment, onboarding, and training costs due to high staff turnover (early employee turnover typically exceeds 60%). Any strategy that reduces the number of vehicle-driver units required to complete a given task will directly alleviate the most significant cost pressures in last-mile logistics.

[0005] Existing mainstream drone-vehicle collaborative strategies have the following limitations: each vehicle typically carries only a single drone, and each vehicle-drone pairing unit remains fixedly bound throughout the delivery process.

[0006] With advancements in vehicle-mounted drone technology, a single vehicle can now carry more than five drones. However, current strategies still restrict drones to independent operating units belonging to their designated vehicles, severely limiting collaborative flexibility: even if a drone can serve more tasks from other, closer vehicles, it must return to its designated vehicle; once a vehicle-drone unit has completed pickup, even if other units are more cost-effective for the corresponding delivery task, that unit must still complete the delivery independently and cannot transfer the package.

[0007] Therefore, there is an urgent need to design a solution to the complex scheduling problem in last-mile delivery scenarios where fixed binding strategies lead to low drone utilization and high vehicle-drone deployment costs. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of the existing technology by providing a method, device and medium for pickup and delivery route planning based on dual-role drone and vehicle collaboration. It aims to solve the complex scheduling problems in last-mile delivery scenarios, such as diverse task types (pickup only, delivery only, and paired pickup and delivery), strict pre-order constraints on paired tasks, and low drone utilization and high vehicle-drone deployment costs caused by fixed binding strategies.

[0009] The objective of this invention can be achieved through the following technical solutions: According to a first aspect of the present invention, a method for planning pickup and delivery routes based on dual-role drone and vehicle collaboration is provided, comprising: S1. Configure at least one vehicle and multiple drones in the spatial area corresponding to the task point. Configure the vehicle as a mobile warehouse for parcel storage and a drone take-off and landing platform. Configure the drones to perform tasks and / or transfer parcels between vehicles. After completing the service, the drones land on the available vehicles. Construct a delivery task map with the warehouse and task point as nodes. S2, Based on the pickup and delivery task graph, generate a greedy initial solution: First, assign all pure pickup points and pure delivery points to vehicles to construct the initial vehicle path; Second, insert paired pickup and delivery task points into the vehicle path; Finally, change the task points that satisfy the drone energy consumption constraint to be served by the drone to construct a greedy initial solution. S3 encodes the greedy initial feasible solution into a quadruple for each task point, which corresponds to the task point number, drone tag, package transfer index and takeoff point index, respectively. S4. Based on the total cost objective function and constraints, and on the basis of the encoded greedy initial feasible solution, the neighborhood search operator is called cyclically within the tabu search framework for optimization. The neighborhood search operator includes the exchange operator, mutation operator and insertion operator. After the termination condition is met, the optimal vehicle-UAV cooperative planning path is decoded and output.

[0010] Preferably, the delivery task map is constructed using warehouses and task points as nodes. , where the node set The set of warehouse node 0 and task node constitute, The number of task nodes and the set of edges. .

[0011] Preferably, the step of changing the task points that satisfy the UAV energy consumption constraints to be served by UAVs specifically means: for task points If from the takeoff point on the vehicle path Arrive at the mission point Then to the landing point The drone's energy consumption shall not exceed the drone's maximum battery capacity. Then the task point The service will be provided by drones instead.

[0012] Preferably, the step of encoding the greedy initial feasible solution into a quadruple for each task point, corresponding to the task point number, drone tag, package transfer index, and takeoff point index, specifically includes: Based on the vehicle path allocation mechanism, the greedy initial feasible solution is encoded as a quadruple for each task point. ); in, Number the task points, where, The time corresponds to the warehouse; The drone is tagged with a numerical code as follows: the current mission point is served by the drone in flight, the vehicle executes the current mission point and launches the drone from here, the vehicle executes the current mission point and receives the drone landing here, and the vehicle executes the current mission point and receives the drone landing here. For package transfer index, where Indicates no transfer. When the value is non-zero, it indicates that the drone is undertaking a package transfer task, and its value is the pickup point number or delivery point number corresponding to the transferred package. This is the takeoff point index, representing the mission point number for the UAV's takeoff.

[0013] Preferably, the drone tag The specific encoding is set as follows: This indicates that the current task point is continuously served by the UAV during a single flight, with multiple consecutive tasks... This corresponds to multiple mission points served sequentially by the same drone after a single takeoff; This indicates that the vehicle serves the current mission point and launches a drone from that point. (Subsequent...) The mission point is executed by the launched drone until it encounters... The landing point; This indicates that the vehicle is serving the current mission point and will receive a drone landing at that point, using the takeoff point index. Record the current mission point as the mission point number received from the drone takeoff; This indicates that the current mission point is only served by vehicles and does not involve any drone operations; in, , , and Four different encoded integer values ​​were set.

[0014] Preferably, the neighborhood search operator includes a swap operator, a mutation operator, and an insertion operator, specifically: (1) Commutation operators, including: The task point swapping operator randomly selects two task points executed by the vehicle and swaps their positions in the vehicle's path. The drone tag swapping operator randomly selects a mission point performed by a drone and a mission point performed by a vehicle, changes the drone's landing point to the vehicle's mission point, and updates relevant information. and Value; and, The drone package transfer operator randomly selects a transfer package and a vehicle mission point within the same takeoff point path, moves the takeoff point of the transfer flight to the vehicle mission point, and updates the data. value; (2) Mutation operators, including: The drone mission mutation operator randomly selects a mission point that was originally performed by a drone, converts it to a mission served by a vehicle, and adjusts the takeoff and landing points accordingly. Value; if the drone flight has only one mission point, then delete the entire flight; and, The vehicle mission mutation operator randomly selects a mission point performed by a vehicle, transforms it into a mission served by a drone, and finds adjacent mission points as takeoff and landing points, updating... and Value; and, (3) Insertion operators, including: Insert warehouse operator: randomly select a task point executed by a vehicle, insert warehouse node 0 in front of it, and split the original vehicle path into two; Insert vehicle path operator: randomly select a task point performed by the UAV, insert it after the task point performed by the vehicle in a vehicle path, and mark it as a vehicle service; Insert the drone flight operator, randomly select a task point performed by a vehicle, insert it into the task point sequence of a certain drone flight, and mark it as a drone service.

[0015] Preferably, the constraints include: (1) First-to-last dispatch preorder constraint, i.e., the pickup task point of the matching task. Must be at the delivery point Previously completed services: , In the formula: For the set of task point nodes, This represents the number of task nodes. For drones At the pickup point The moment of departure, For vehicles At the pickup point The moment of departure; For drones At the delivery point Arrival time at the location; For vehicles At the delivery point Arrival time at the location; Indicates drone Is it a service task point? 1 indicates a task point to be executed, and 0 indicates that the task point is not executed by a drone; each task point is executed once by a vehicle or a drone. (2) Energy consumption constraints for a single UAV flight: , In the formula: For drones From the mission point Arrive at the mission point Flight energy consumption; Delivery and pickup task diagram The set of edges in; For drones; This represents the maximum battery capacity of the drone. To characterize drones From the mission point Fly to the mission point binary variables, Indicates drone From the mission point Fly to the mission point ; (3) Three-node flight path integrated energy constraint to ensure that the UAV, in the dynamic cross-truck cooperation process, starts from the take-off node Through task nodes to recycling node The complete flight path satisfies the power constraint, making the generated air-to-ground cooperative scheme practically feasible. , In the formula: For drones From the mission point Arrive at the mission point Flight energy consumption; For drones At the mission point Energy consumption for performing tasks; For drones At the mission point Energy consumption during hovering wait time; For drones At the mission point Energy consumption of vehicles waiting to land; Indicates drone Service task points ; (4) Consistency constraint between arrival time and travel time: , In the formula: To start from the mission point Arrive at the mission point The distance; For the flight speed of the drone ; Delivery and pickup task diagram The set of nodes in; It is an arbitrarily large positive number; (5) Unmanned Aerial Vehicle (UAV) payload tracking constraints: , In the formula: , The drones at the mission points Task Points The load; For the task point Demand; This is the maximum payload capacity for the drone; (6) Vehicle load tracking constraints: , In the formula: This is the vehicle's maximum load capacity; For vehicle assembly.

[0016] Preferably, the total cost objective function includes vehicle operating costs, drone energy consumption costs, and fixed deployment costs, and its mathematical expression is: , , , , , In the formula: The objective function is the total cost. For drones From the mission point Arrive at the mission point Flight energy consumption; For drones At the mission point Energy consumption for performing tasks; For drones At the mission point Energy consumption during hovering wait time; For drones At the mission point Energy consumption of vehicles waiting to land; Energy consumption coefficient per unit; For the drone's own weight; , , They represent drones At the mission point Task Points Task Points The load capacity, For the drones in this flight The last task to be executed; To start from the mission point Arrive at the mission point The distance; The flight speed of the drone; For the task point Demand; For the task point Service duration, For the task point The duration of hovering; Indicates whether the vehicle departed from the mission point. Drive to the mission point Judgment variables; cost coefficient This represents the unit distance cost of a vehicle, and the cost coefficient. This indicates the unit energy consumption cost and cost coefficient of drones. This represents the fixed cost of deploying each drone-vehicle pair; When drones When performing pure parcel delivery between vehicles, only flight energy consumption is considered. and waiting vehicle energy consumption When drones When performing missions and transferring cargo packages simultaneously on the same flight, its payload... Includes the weight of the packages to be delivered.

[0017] According to a second aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement any of the methods described above.

[0018] According to a third aspect of the invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements any of the methods described above.

[0019] Compared with the prior art, the present invention has the following advantages: (1) Addressing the core limitations of existing UAV-vehicle collaborative strategies, such as the fixed binding of UAVs to vehicles, the inability to dynamically land between different vehicles, and the inability to transfer packages between vehicles, this invention endows UAVs with dual capabilities: on the one hand, as an end effector, it efficiently completes services at various task points; on the other hand, it acts as a package transfer relay, dynamically transferring picked-up packages between vehicles. This fundamentally breaks the rigid constraint of the traditional fixed binding strategy, which requires the same vehicle-UAV unit to independently complete paired tasks for pickup and delivery, thus achieving flexible collaboration across units and vehicles.

[0020] (2) The present invention designs a vehicle path-based allocation mechanism, which represents the solution as a four-value code for each task point (task point code, UAV tag, UAV payload, UAV index), and tightly couples the task point access sequence, UAV take-off and landing marker, package delivery identifier and flight traceability information, thereby accurately depicting the complex spatiotemporal dynamic relationship of UAVs dynamically taking off and landing between different vehicles, delivering packages across units, and simultaneously performing tasks and transferring packages in a single flight.

[0021] (3) A three-stage greedy initialization strategy and a two-layer logic of tabu search depth optimization are adopted: First, based on hard constraints such as vehicle load, task time window, pairing task preorder constraints, and UAV maximum battery capacity, an initial solution that meets basic feasibility is quickly constructed through three stages: nearest neighbor insertion, pairing task embedding, and task subject reallocation; then, a tabu search algorithm framework is introduced, and three types of multi-level neighborhood search operators are designed, covering exchange operators for task point exchange, UAV tag exchange, and package delivery exchange, mutation operators for UAV task and vehicle task conversion, and insertion operators for inserting warehouse nodes, inserting vehicle execution, and inserting UAV execution; efficient neighborhood search is performed in the solution space. In particular, the warehouse node insertion operator can dynamically split a vehicle route into two, enabling the algorithm to adaptively adjust the total number of deployed vehicle-UAV units, achieving an optimal balance between solution quality and deployment cost.

[0022] (4) The optimization process of this invention is controlled by a comprehensive objective function, which dynamically weighs the vehicle driving cost, the total energy consumption of the UAV and the fixed deployment cost of the vehicle-UAV, thereby ensuring that the output path scheme satisfies all time window constraints, first-pick-then-delivery preorder constraints and energy feasibility constraints, and can achieve the lowest total cost with the fewest unit deployments in the last-mile delivery scenario. Attached Figure Description

[0023] Figure 1 This is a flowchart of the method of the present invention.

[0024] Figure 2 A flowchart for collaborative path planning of dynamic dual-role drones and vehicles.

[0025] Figure 3 This is a schematic diagram illustrating a DRDC (Device Response Control) strategy. The scenario includes 10 task points, 3 vehicles (k1, k2, k3), and 3 drones (u1, u2, u3). The diagram shows the complete flow path of package 1: k1 → u1 picks up the package → lands at k2 to transfer the package → u2 continues delivery → lands at k3 → k3 delivers the package to its destination. This embodies the dual innovations of multi-hop package delivery and drone landing across vehicles.

[0026] Figure 4 This is an example of coding a vehicle route-based allocation mechanism. Each of the three vehicle routes is represented by a four-row matrix of TP / DT / DL / DI. The location distribution of DT=1 (take-off point), DT=-1 (landing point), DT=0 (Drone execution), and DT=2 (vehicle execution alone) is highlighted, as well as the cargo delivery operation corresponding to the non-zero column of DL.

[0027] Figure 5 To illustrate the operator swapping operations, we show the encoding changes after three operations: swapping the original solution with the task point, swapping the drone tag, and swapping the package delivery. The change column is highlighted.

[0028] Figure 6 Examples of mutation operator operations are shown, illustrating the encoding changes after a drone mission is mutated into a vehicle mission and a vehicle mission is mutated into a drone mission.

[0029] Figure 7 Examples of insertion operator operations are shown, illustrating the changes in the encoding matrix after inserting a warehouse node, inserting a vehicle, and inserting a drone. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0031] Example like Figure 1 As shown, this embodiment provides a method for pickup and delivery route planning based on dual-role drone and vehicle collaboration. The method includes the following steps: S1. Configure at least one vehicle (which can be an unmanned vehicle, truck, etc.) and multiple drones in the spatial area corresponding to the task point. Configure the vehicle as a mobile warehouse for parcel storage and a drone take-off and landing platform. Configure the drones to perform tasks and / or transfer parcels between vehicles. After completing the service, the drones land on an available vehicle. Construct a delivery task map with the warehouse and task point as nodes.

[0032] In one embodiment of the present invention, a delivery task graph is constructed using warehouses and task points as nodes. , where the node set The set of warehouse node 0 and task node constitute, The number of task nodes and the set of edges. .

[0033] S2. Based on the pickup and delivery task graph, generate a greedy initial solution: First, assign all pure pickup points and pure delivery points to vehicles to construct the initial vehicle path; second, insert paired pickup and delivery task points into the vehicle path; finally, change the task points that satisfy the drone energy consumption constraint to be served by the drone to construct a greedy initial solution.

[0034] In this embodiment, the task points that satisfy the UAV energy consumption constraints are instead served by the UAV. Specifically, for the task points... If from the takeoff point on the vehicle path Arrive at the mission point Then to the landing point The drone's energy consumption shall not exceed the drone's maximum battery capacity. Then the task point The service will be provided by drones instead.

[0035] Figure 3 This invention illustrates the path planning strategy, where the scenario includes 10 task points and 3 trucks ( , , ) and 3 drones ( , , The diagram shows the complete flow path of package 1: → Pick up package → Landing Transfer package → Continue passing → Landing → Delivery to 2 This embodies the dual innovations of multi-hop package delivery and drone landing across vehicles.

[0036] S3 encodes the greedy initial feasible solution into a quadruple for each task point, corresponding to the task point number, drone tag, package transfer index, and takeoff point index, specifically including: Based on the vehicle path allocation mechanism, the greedy initial feasible solution is encoded as a quadruple for each task point. ); in, Number the task points, where, The time corresponds to the warehouse; The drone is tagged with a numerical code as follows: the current mission point is served by the drone in flight, the vehicle executes the current mission point and launches the drone from here, the vehicle executes the current mission point and receives the drone landing here, and the vehicle executes the current mission point and receives the drone landing here. For package transfer index, where Indicates no transfer. When the value is non-zero, it indicates that the drone is undertaking a package transfer task, and its value is the pickup point number or delivery point number corresponding to the transferred package. This is the takeoff point index, representing the mission point number for the UAV's takeoff.

[0037] Specifically, drone tags The encoding process is set as follows: This indicates that the current task point is continuously served by the UAV during a single flight, with multiple consecutive tasks... This corresponds to multiple mission points served sequentially by the same drone after a single takeoff; This indicates that the vehicle serves the current mission point and launches a drone from that point. (Subsequent...) The mission point is executed by the launched drone until it encounters... The landing point; This indicates that the vehicle is serving the current mission point and will receive a drone landing at that point, using the takeoff point index. Record the current mission point as the mission point number received from the drone takeoff; This indicates that the current mission point is only served by vehicles and does not involve any drone operations; in, , , and In this embodiment, four different encoded integer values ​​are set. Set to 0, Set to 1, Set to 1, Set it to 2.

[0038] Figure 4 The example of a vehicle route allocation mechanism is shown. The three truck routes are each represented by a four-row matrix of TP / DT / DL / DI. The location distribution of DT=1 (take-off point), DT=-1 (landing point), DT=0 (drone execution), and DT=2 (truck execution alone) is highlighted, as well as the cargo delivery operation corresponding to the non-zero column of DL.

[0039] S4. Based on the total cost objective function and constraints, and on the basis of the encoded greedy initial feasible solution, the neighborhood search operator is called cyclically within the tabu search framework to find the optimal path. After the termination condition is met, the optimal vehicle-UAV cooperative planning path is decoded and output.

[0040] In this embodiment, the neighborhood search operator includes a swap operator, a mutation operator, and an insertion operator, specifically: (1) Commutation operators, including: The task point swapping operator randomly selects two task points executed by the vehicle and swaps their positions in the vehicle's path. The drone tag swapping operator randomly selects a mission point performed by a drone and a mission point performed by a vehicle, changes the drone's landing point to the vehicle's mission point, and updates relevant information. and Value; and, The drone package transfer operator randomly selects a transfer package and a vehicle mission point within the same takeoff point path, moves the takeoff point of the transfer flight to the vehicle mission point, and updates the data. value.

[0041] (2) Mutation operators, including: The drone mission mutation operator randomly selects a mission point that was originally performed by a drone, converts it to a mission served by a vehicle, and adjusts the takeoff and landing points accordingly. Value; if the drone flight has only one mission point, then delete the entire flight; and, The vehicle mission mutation operator randomly selects a mission point performed by a vehicle, transforms it into a mission served by a drone, and finds adjacent mission points as takeoff and landing points, updating... and Value; and, (3) Insertion operators, including: Insert warehouse operator: randomly select a task point executed by a vehicle, insert warehouse node 0 in front of it, and split the original vehicle path into two; Insert vehicle path operator: randomly select a task point performed by the UAV, insert it after the task point performed by the vehicle in a vehicle path, and mark it as a vehicle service; Insert the drone flight operator, randomly select a task point performed by a vehicle, insert it into the task point sequence of a certain drone flight, and mark it as a drone service.

[0042] Figures 5-7 Specific operational examples of the exchange operator, mutation operator, and insertion operator are given respectively.

[0043] In this embodiment, the total cost objective function includes vehicle operating costs, drone energy consumption costs, and fixed deployment costs, and its mathematical expression is: , , , , , In the formula: The objective function is the total cost. For drones From the mission point Arrive at the mission point Flight energy consumption; For drones At the mission point Energy consumption for performing tasks; For drones At the mission point Energy consumption during hovering wait time; For drones At the mission point Energy consumption of vehicles waiting to land; Energy consumption coefficient per unit; For the drone's own weight; , , They represent drones At the mission point Task Points Task Points The load capacity, For the drones in this flight The last task to be executed; To start from the mission point Arrive at the mission point The distance; The flight speed of the drone; For the task point Demand; For the task point Service duration, For the task point The duration of hovering; Indicates whether the vehicle departed from the mission point. Drive to the mission point Judgment variables; cost coefficient This represents the unit distance cost of a vehicle, and the cost coefficient. This indicates the unit energy consumption cost and cost coefficient of drones. This represents the fixed cost of deploying each drone-vehicle pair; When drones When performing pure parcel delivery between vehicles (drones only deliver packages between vehicles and do not perform any other tasks), only flight energy consumption is considered. and waiting vehicle energy consumption When drones When performing missions and transferring cargo packages simultaneously on the same flight, its payload... Includes the weight of the packages to be delivered.

[0044] In this embodiment, the constraints include: the pickup point is executed before the corresponding delivery point, the total energy consumption of a single drone flight does not exceed E, vehicle traffic is balanced, each task point is executed exactly once by one vehicle / drone, time window constraints, and load constraints.

[0045] The specific settings are as follows: (1) First-to-last dispatch preorder constraint, i.e., the pickup task point of the matching task. Must be at the delivery point Previously completed services: , In the formula: For the set of task point nodes, Number of task nodes ; For drones At the pickup point The moment of departure, For vehicles At the pickup point The moment of departure; For drones At the delivery point Arrival time at the location; For vehicles At the delivery point Arrival time at the location; Indicates drone Is it a service task point? 1 indicates a task point to be executed, and 0 indicates that the task point is not executed by a drone. Each task point is executed once by either a truck or a drone.

[0046] (2) Energy consumption constraints for a single UAV flight: , In the formula: For drones From the mission point Arrive at the mission point Flight energy consumption; Delivery and pickup task diagram The set of edges in; For drones; This represents the maximum battery capacity of the drone. To characterize drones From the mission point Fly to the mission point binary variables, Indicates drone From the mission point Fly to the mission point ; (3) Three-node flight path integrated energy constraint to ensure that the UAV, in the dynamic cross-truck cooperation process, starts from the take-off node Through task nodes to recycling node The complete flight path satisfies the power constraint, making the generated air-to-ground cooperative scheme practically feasible. , In the formula: For drones From the mission point Arrive at the mission point Flight energy consumption; For drones At the mission point Energy consumption for performing tasks; For drones At the mission point Energy consumption during hovering wait time; For drones At the mission point Energy consumption of vehicles waiting to land; Indicates drone Service task points ; (4) Consistency constraint between arrival time and travel time: , In the formula: To start from the mission point Arrive at the mission point The distance; The flight speed of the drone is set to [specific value] in this embodiment. ; Delivery and pickup task diagram The set of nodes in; It is an arbitrarily large positive number; (5) Unmanned Aerial Vehicle (UAV) payload tracking constraints: , In the formula: , For drones At the mission point Task Points The load; For the task point Demand; This is the maximum payload capacity for the drone; (6) Vehicle load tracking constraints: , In the formula: This is the vehicle's maximum load capacity; For vehicle assembly.

[0047] After optimization, the encoded sequence is transformed into an executable vehicle-drone cooperative path, determining the driving route of each vehicle, the take-off / landing point and flight path of each drone, and handling the connection of package transfer.

[0048] The specific implementation of the path planning process in this embodiment is as follows: Figure 2 As shown.

[0049] To more intuitively verify the performance of the method of this invention, the pickup and delivery path planning method proposed in this invention was compared with the current best air-ground cooperative path planning method (ILNS, S. Meng, Y. Chen, and D. Li, “Pickup and delivery problem with multi-visit drones considering soft time windows”) by different target point distributions, drone numbers, and flight speed settings. Transportation Research Part C: Emerging Technologies The results of the comparison with the journal (vol. 180, p.105359, 2025) are shown in Table 1. , , These are the cost reduction rates compared to the mPDP, ILNS, and DRDC_o algorithms, respectively.

[0050] Table 1 The results show that the present invention significantly improves the efficiency and economy of drone-vehicle collaboration in the pickup and delivery problem through dynamic landing and package transfer mechanisms, and has high practical value and promising prospects for promotion.

[0051] The electronic device of this invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0052] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0053] The processing unit executes the various methods and processes described above, such as methods S1 to S4. For example, in some embodiments, methods S1 to S4 may be implemented as computer software programs tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of methods S1 to S4 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S4 by any other suitable means (e.g., by means of firmware).

[0054] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload programmable logic devices (CPLDs), and so on.

[0055] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0056] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0057] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for planning pickup and delivery routes based on dual-role drone and vehicle collaboration, characterized in that, include: S1. Configure at least one vehicle and multiple drones in the spatial area corresponding to the task point. Configure the vehicle as a mobile warehouse for parcel storage and a drone take-off and landing platform. Configure the drones to perform tasks and / or transfer parcels between vehicles. After completing the service, the drones land on the available vehicles. Construct a delivery task map with the warehouse and task point as nodes. S2, Based on the pickup and delivery task graph, generate a greedy initial solution: First, assign all pure pickup points and pure delivery points to vehicles to construct the initial vehicle path; Second, insert paired pickup and delivery task points into the vehicle path; Finally, change the task points that satisfy the drone energy consumption constraint to be served by the drone to construct a greedy initial solution. S3 encodes the greedy initial feasible solution into a quadruple for each task point, which corresponds to the task point number, drone tag, package transfer index and takeoff point index, respectively. S4. Based on the total cost objective function and constraints, and on the basis of the encoded greedy initial feasible solution, the neighborhood search operator is called cyclically within the tabu search framework for optimization. The neighborhood search operator includes the exchange operator, mutation operator and insertion operator. After the termination condition is met, the optimal vehicle-UAV cooperative planning path is decoded and output.

2. The method for planning pickup and delivery routes based on dual-role UAV and vehicle collaboration according to claim 1, characterized in that, The delivery task graph is constructed using warehouses and task points as nodes. , where the node set The set of warehouse node 0 and task node constitute, The number of task nodes and the set of edges. .

3. The method for planning pickup and delivery routes based on dual-role drone and vehicle collaboration according to claim 1, characterized in that, The step of changing the service of task points that meet the energy consumption constraints of UAVs to UAVs specifically means: for task points If from the takeoff point on the vehicle path Arrive at the mission point Then to the landing point The drone's energy consumption shall not exceed the drone's maximum battery capacity. Then the task point The service will be provided by drones instead.

4. The method for planning pickup and delivery routes based on dual-role drone and vehicle collaboration according to claim 1, characterized in that, The greedy initial feasible solution is encoded into a quadruple for each task point, corresponding to the task point number, drone tag, package transfer index, and takeoff point index, respectively. Specifically, this includes: Based on the vehicle path allocation mechanism, the greedy initial feasible solution is encoded as a quadruple for each task point. ); in, Number the task points, where, The time corresponds to the warehouse; The drone is tagged with a numerical code as follows: the current mission point is served by the drone in flight, the vehicle executes the current mission point and launches the drone from here, the vehicle executes the current mission point and receives the drone landing here, and the vehicle executes the current mission point and receives the drone landing here. For package transfer index, where Indicates no transfer. When the value is non-zero, it indicates that the drone is undertaking a package transfer task, and its value is the pickup point number or delivery point number corresponding to the transferred package. This is the takeoff point index, representing the mission point number for the UAV's takeoff.

5. The method for planning pickup and delivery routes based on dual-role drone and vehicle collaboration according to claim 4, characterized in that, The drone tag The specific encoding is set as follows: This indicates that the current task point is continuously served by the UAV during a single flight, with multiple consecutive tasks... This corresponds to multiple mission points served sequentially by the same drone after a single takeoff; This indicates that the vehicle serves the current mission point and launches a drone from that point. (Subsequent...) The mission point is executed by the launched drone until it encounters... The landing point; This indicates that the vehicle is serving the current mission point and will receive a drone landing at that point, using the takeoff point index. Record the current mission point as the mission point number received from the drone takeoff; This indicates that the current mission point is only served by vehicles and does not involve any drone operations; in, , , and Four different encoded integer values ​​were set.

6. The method for planning pickup and delivery routes based on dual-role drone and vehicle collaboration according to claim 1, characterized in that, The neighborhood search operator includes a swap operator, a mutation operator, and an insertion operator, specifically: (1) Commutation operators, including: The task point swapping operator randomly selects two task points executed by the vehicle and swaps their positions in the vehicle's path. The drone tag swapping operator randomly selects a mission point performed by a drone and a mission point performed by a vehicle, changes the drone's landing point to the vehicle's mission point, and updates relevant information. and Value; and, The drone package transfer operator randomly selects a transfer package and a vehicle mission point within the same takeoff point path, moves the takeoff point of the transfer flight to the vehicle mission point, and updates the data. value; (2) Mutation operators, including: The drone mission mutation operator randomly selects a mission point that was originally performed by a drone, converts it to a mission served by a vehicle, and adjusts the takeoff and landing points accordingly. Value; if the drone flight has only one mission point, then delete the entire flight; and, The vehicle mission mutation operator randomly selects a mission point performed by a vehicle, transforms it into a mission served by a drone, and finds adjacent mission points as takeoff and landing points, updating... and Value; and, (3) Insertion operators, including: Insert warehouse operator: randomly select a task point executed by a vehicle, insert warehouse node 0 in front of it, and split the original vehicle path into two; Insert vehicle path operator: randomly select a task point performed by the UAV, insert it after the task point performed by the vehicle in a vehicle path, and mark it as a vehicle service; Insert the drone flight operator, randomly select a task point performed by a vehicle, insert it into the task point sequence of a certain drone flight, and mark it as a drone service.

7. The method for planning pickup and delivery routes based on dual-role drone and vehicle collaboration according to claim 1, characterized in that, The constraints include: (1) First-to-last dispatch preorder constraint, i.e., the pickup task point of the matching task. Must be at the delivery point Previously completed services: ， In the formula: For the set of task point nodes, This represents the number of task nodes. For drones At the pickup point The moment of departure, For vehicles At the pickup point The moment of departure; For drones At the delivery point Arrival time at the location; For vehicles At the delivery point Arrival time at the location; Indicates drone Is it a service task point? 1 indicates a task point to be executed, and 0 indicates that the task point is not executed by a drone; each task point is executed once by a vehicle or a drone. (2) Energy consumption constraints for a single UAV flight: ， In the formula: For drones From the mission point Arrive at the mission point Flight energy consumption; Delivery and pickup task diagram The set of edges in; For drones; This represents the maximum battery capacity of the drone. To characterize drones From the mission point Fly to the mission point binary variables, Indicates drone From the mission point Fly to the mission point ; (3) Three-node flight path integrated energy constraint, used to ensure that the UAV, in the dynamic cross-vehicle cooperative process, from the take-off node Through task nodes to landing node The complete flight path satisfies the power constraint: ， In the formula: For drones From the mission point Arrive at the mission point Flight energy consumption; For drones At the mission point Energy consumption for performing tasks; For drones At the mission point Energy consumption during hovering wait time; For drones At the mission point Energy consumption of vehicles waiting to land; Indicates drone Service task points ; (4) Consistency constraint between arrival time and travel time: ， In the formula: To start from the mission point Arrive at the mission point The distance; The flight speed of the drone; Delivery and pickup task diagram The set of nodes in; It is an arbitrarily large positive number; (5) Unmanned Aerial Vehicle (UAV) payload tracking constraints: ， In the formula: , The drones at the mission points Task Points The load; For the task point Demand; This is the maximum payload capacity for the drone; (6) Vehicle load tracking constraints: ， In the formula: This is the vehicle's maximum load capacity; For vehicle assembly.

8. The method for planning pickup and delivery routes based on dual-role drone and vehicle collaboration according to claim 1, characterized in that, The total cost objective function includes vehicle operating costs, drone energy consumption costs, and fixed deployment costs, and its mathematical expression is: ， ， ， ， ， In the formula: The objective function is the total cost. For drones From the mission point Arrive at the mission point Flight energy consumption; For drones At the mission point Energy consumption for performing tasks; For drones At the mission point Energy consumption during hovering wait time; For drones At the mission point Energy consumption of vehicles waiting to land; Energy consumption coefficient per unit; For the drone's own weight; , , They represent drones At the mission point Task Points Task Points The load capacity, For the drones in this flight The last task to be executed; To start from the mission point Arrive at the mission point The distance; The flight speed of the drone; For the task point Demand; For the task point Service duration, For the task point The duration of hovering; Indicates whether the vehicle departed from the mission point. Drive to the mission point Judgment variables; cost coefficient This represents the unit distance cost of a vehicle, and the cost coefficient. This indicates the unit energy consumption cost and cost coefficient of drones. This represents the fixed cost of deploying each drone-vehicle pair; When drones When performing pure parcel delivery between vehicles, only flight energy consumption is considered. and waiting vehicle energy consumption When drones When performing missions and transferring cargo packages simultaneously on the same flight, its payload... Includes the weight of the packages to be delivered.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 8.

Images