Travel and distribution scheduling method based on online car-hailing-unmanned aerial vehicle cooperation

Abstract

The invention discloses a travel and distribution scheduling method based on online car-hailing-unmanned aerial vehicle cooperation, and relates to the technical field of travel service and distribution collaboration.The method comprises the steps that firstly, parcel distribution data, online car-hailing GPS track data and passenger order data are preprocessed based on a big language model, the track data is associated with the order data, and a driver order receiving record is generated; in combination with behavior characteristics of different drivers on the online car-hailing platform, a driver response model is established through machine learning based on historical order receiving records of the drivers, real-time orders are analyzed by means of the model, and the order receiving probability of the drivers for newly-added passenger orders is output; the method comprises the following steps: constructing a travel and distribution scheduling mathematical model by taking minimization of an online car-hailing unloaded distance and an unmanned aerial vehicle unloaded distance as a target, optimizing an online car-hailing order dispatching and unmanned aerial vehicle delivery scheme based on a heuristic algorithm, and searching a scheduling scheme with a minimum target value. The resource utilization rate of the vehicle and the unmanned aerial vehicle is improved, and the passenger and cargo flow transportation efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of travel service and delivery collaboration technology, and in particular to a travel and delivery scheduling method based on ride-hailing-drone collaboration. Background Technology

[0002] With the rapid development of mobile internet and on-demand delivery, the "last mile" delivery problem in cities is becoming increasingly prominent. Existing research and industry practices in vehicle-drone collaboration mainly focus on hybrid solutions such as trucks carrying drones, small delivery vehicles and human couriers, or cooperation between public transportation and drones. These models rely on fixed warehousing networks or dedicated delivery routes. Furthermore, traditional dedicated delivery vehicles often incur significant empty-running costs when orders are scattered or at different times. While drones have a speed advantage in short-distance, small-item delivery, they are limited by factors such as range and takeoff / landing space, making it difficult to independently cover a larger delivery area and resulting in high costs in low-density scenarios. Therefore, existing research cannot fully address the problem of simultaneously optimizing ground driving and aerial flight in a dynamic, real-time order dispatch environment.

[0003] Collaborating ride-hailing networks with drones demonstrates significant resource complementarity and engineering feasibility. Ride-hailing services, with their widely distributed mobile platforms and mature real-time location-based dispatching systems, can serve as flexible relay or recovery nodes for drones, significantly extending their effective service radius and reducing their independent round-trip energy consumption. In situations where passenger travel and delivery tasks overlap in time and space, simultaneously triggering drones to perform short-range deliveries and then rendezvousing with vehicles for charging after delivery can complete delivery tasks without significantly increasing driver detours or impacting passenger experience. This reduces ground mileage and flight distance, improves platform resource utilization, and lowers operating costs. Simultaneously, this model helps provide drivers with additional income, reduces reliance on dedicated delivery fleets, and mitigates road congestion and environmental impact. However, current research has not yet proposed a systematic delivery solution combining drones and ride-hailing services. Summary of the Invention

[0004] The technical problem this invention aims to solve is to overcome the shortcomings of existing technologies and propose a travel and delivery scheduling method based on ride-hailing-drone collaboration. This method fully considers the behavioral characteristics of ride-hailing drivers during the order-accepting process, establishes a driver response model through machine learning, and predicts the order-accepting probability of each driver under given dispatch and empty-running costs. Based on this, with the objective of simultaneously minimizing the empty-running distance of ride-hailing vehicles and the empty-load distance of drones, a joint optimization problem is constructed, constrained by factors such as the number of orders accepted by drivers and the drone's endurance. A heuristic algorithm is then used to search for the optimal scheduling scheme.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] A travel and delivery scheduling method based on ride-hailing-drone collaboration proposed in this invention includes:

[0007] Step A: Based on the large language model, preprocess the parcel delivery data, ride-hailing GPS trajectory data, and passenger order data;

[0008] The pre-processed ride-hailing GPS trajectory data is associated with passenger order data to generate driver order records. At the same time, based on the driver order records, the parcel delivery data within the preset delivery area is initially extracted.

[0009] Step B: Combining the behavioral characteristics of different drivers on ride-hailing platforms, and based on their historical order records, establish a driver response model through machine learning;

[0010] Based on the driver response model, real-time orders are analyzed to predict the probability of each driver accepting an order given by the ride-hailing platform and the cost of empty driving.

[0011] Step C: Determine the initial ride-hailing-drone collaborative dispatch plan, which includes a ride-hailing order dispatch plan and a drone delivery plan; among which,

[0012] For ride-hailing order dispatching schemes, ride-hailing platforms always adopt a one-to-many dispatching strategy, selecting the multiple drivers with the highest order acceptance probability to dispatch orders.

[0013] For drone delivery solutions, based on the initial extracted parcel delivery data, ride-hailing platforms select a parcel for each passenger order that minimizes the drone's empty-load distance.

[0014] Step D: Optimize the initial ride-hailing-drone collaborative scheduling scheme;

[0015] With the goal of minimizing the empty driving distance of ride-hailing vehicles and the empty carrying distance of drones, a travel and delivery scheduling model is constructed. Based on heuristic algorithms, a ride-hailing-drone collaborative scheduling scheme with the minimum objective function value of the travel and delivery scheduling model is found.

[0016] As a further optimization of the ride-hailing-drone collaborative travel and delivery scheduling method described in this invention, in step C, the one-to-many order dispatch strategy means that a single passenger order can be assigned by the ride-hailing platform. Each driver can only accept one order at a time.

[0017] As a further optimization of the travel and delivery scheduling method based on ride-hailing-drone collaboration described in this invention, in step C, the drone's empty distance includes: the distance from the ride-hailing vehicle to the parcel delivery location and the distance from the parcel's final destination to the ride-hailing vehicle meeting point.

[0018] As a further optimization scheme of the travel and delivery scheduling method based on ride-hailing-drone collaboration described in this invention, step A includes preprocessing including filtering out abnormal data and repairing missing data; wherein, filtering out abnormal data refers to identifying and removing spatial positioning offsets and speed anomaly records in package delivery data, ride-hailing GPS trajectory data and passenger order data; repairing missing data refers to filling in missing timestamps or geographical location data in ride-hailing GPS trajectory data.

[0019] As a further optimization of the ride-hailing-drone collaborative travel and delivery scheduling method described in this invention, step A, which involves linking the preprocessed ride-hailing GPS trajectory data with passenger order data, includes: matching the driver and order IDs to generate a driver order record; this driver order record includes a passenger order feature set and a ride-hailing driver attribute set: the passenger order feature set includes the order start coordinates, destination coordinates, fare information, estimated mileage, and service period; the ride-hailing driver attribute set includes the vehicle's starting location, the driver's historical order frequency, and the dispatch response distance.

[0020] As a further optimization of the ride-hailing-drone collaborative travel and delivery scheduling method described in this invention, in step C, the ride-hailing dispatch scheme and the drone delivery scheme are determined as follows:

[0021] Ride-hailing platforms generate real-time first For each passenger order, select the one with the highest probability of acceptance. One driver is used as the ride-hailing order dispatching scheme;

[0022] For the real-time generated first Each passenger order determines the set of parcel delivery orders within the drone's flight time. Calculate the empty distance for drone delivery of the package order, and select the package order with the shortest empty distance as the drone delivery solution.

[0023] As a further optimization scheme of the ride-hailing-drone collaborative travel and delivery scheduling method described in this invention, in step D, a travel and delivery scheduling model is constructed with the goal of minimizing the empty driving distance of ride-hailing vehicles and the empty carrying distance of drones. The model is expressed as follows:

[0024]

[0025] in, , , and These represent the number of passenger orders, ride-hailing drivers, drones, and candidate meeting points, respectively. Indicates the first Is the passenger order assigned to the first...? A ride-hailing driver; , They represent the first The, the The ride-hailing driver accepted the first The probability value of each passenger's order; Indicates the first The ride-hailing driver was performing the first Does the first passenger order enable the first One drone; Indicates the first The ride-hailing driver and the first Are the drones at the rendezvous point? Complete the first One passenger order; For the first A ride-hailing driver at the dispatch time Location, For the first The pick-up point for each passenger's order; , For the first The origin and delivery points for drone-delivered parcels; For the first The first drone One candidate rendezvous point location; , These are the distance functions for ride-hailing vehicles and drones, respectively. The time required for drones to load and unload packages. For the first The completion time of each passenger's order. This refers to the maximum time that ride-hailing drivers can stay at the meeting point. This represents the longest flight time of a drone when fully charged. and These are the speed parameters for ride-hailing vehicles and drones, respectively. , These are the weighting coefficients; This represents the objective function value of the travel and delivery scheduling model.

[0026] As a further optimization scheme of the ride-hailing-drone collaborative travel and delivery scheduling method described in this invention, step D involves finding the ride-hailing-drone collaborative scheduling scheme with the minimum objective function value of the travel and delivery scheduling model based on the hill-climbing algorithm HC in heuristic algorithms; including:

[0027] Step D11: The ride-hailing platform generates the first... For each passenger order, calculate the objective function value of the travel and delivery scheduling model corresponding to the initial ride-hailing-drone collaborative scheduling scheme;

[0028] Step D12: By partially adjusting the matching relationship between passenger orders and ride-hailing drivers, a new dispatch plan is generated. The partial adjustment of the matching relationship is as follows:

[0029] The dispatch objects of two passenger orders are swapped to explore neighborhood solutions. Each local adjustment only changes a finite number of items in the current scheduling scheme, thus forming a new scheduling scheme.

[0030] Step D13: Calculate the objective function value corresponding to the new scheduling scheme and compare it with the objective function value of the current scheduling scheme; if the objective function value corresponding to the new scheduling scheme is less than the objective function value corresponding to the current scheduling scheme, the ride-hailing platform replaces the current scheduling scheme with the new scheduling scheme; otherwise, retain the current scheduling scheme and continue to search for a new scheduling scheme in the remaining neighborhood.

[0031] Step D14: Repeat steps D11 to D13 until the objective function value of the scheduling scheme no longer changes or the iteration number is reached. The scheduling scheme obtained at this time is the scheduling scheme with the minimum objective function value.

[0032] As a further optimization of the travel and delivery scheduling method based on ride-hailing-drone collaboration described in this invention, after the scheduling plan is determined, a drone parked above the ride-hailing vehicle is assigned to deliver the goods. After the delivery is completed, the drone flies back to the destination where it received the passenger order at the departure time and meets up with the ride-hailing vehicle. During idle time, the drone is parked above the ride-hailing vehicle to charge.

[0033] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects:

[0034] (1) This invention is the first to organically combine ride-hailing scheduling with drone delivery, and proposes a scheduling framework of "passenger travel + drone delivery": without affecting passenger pick-up and drop-off, the drone can complete the package delivery by using the meeting point of the vehicle in motion; this mode can carry out travel and delivery functions in the same travel task at the same time, improve the resource utilization rate of vehicles and drones, and open up a new path for ride-hailing platforms to participate in urban micro-logistics.

[0035] (2) The present invention explicitly introduces the upper limit of the single flight range of the UAV as a hard constraint in the model, and pre-eliminates flight schemes with insufficient flight range when generating candidate delivery schemes; at the same time, loading and unloading time, rendezvous waiting and allowable tolerance are included in the time feasibility test to ensure that the selected scheme is feasible in terms of power, time window and operation, thereby avoiding the safety and operation risks caused by the UAV returning midway or mission failure.

[0036] (3) The scheduling method proposed in this invention optimizes the target by calculating the weighted sum of the empty driving distance of ride-hailing vehicles and the empty carrying distance of drones. This enables the platform to minimize scheduling mileage and energy consumption while also adjusting the preference for driver order acceptance cost or drone energy consumption according to the weight. This target is conducive to reducing platform operating costs and carbon emissions. Attached Figure Description

[0037] Figure 1 This is a flowchart of a travel and delivery scheduling method based on ride-hailing and drone collaboration according to the present invention. Detailed Implementation

[0038] To further understand the present invention, preferred embodiments of the present invention are described below in conjunction with examples. However, it should be understood that these descriptions are only for further illustrating the features and advantages of the present invention, and not for limiting the scope of the claims of the present invention.

[0039] The following embodiments use Chengdu as an example, but the application of this invention is not limited to the scope described in the embodiments. Combinations of different embodiments, substitution of some technical features in different embodiments, and substitution of the same or similar prior art with some technical features in the embodiments are also within the scope of description and protection of this invention.

[0040] This embodiment provides a method and system for travel and delivery scheduling based on ride-hailing-drone collaboration, referencing... Figure 1 As shown, it includes the following steps:

[0041] Step A: Parcel, ride-hailing, and passenger data processing:

[0042] Based on a large language model, preprocessing is performed on parcel delivery data, ride-hailing GPS trajectory data, and passenger order data;

[0043] The preprocessed ride-hailing GPS trajectory data is linked with passenger order data to generate driver order records. Simultaneously, based on these driver order records, preliminary parcel delivery data within the preset delivery area is extracted. The specific steps include:

[0044] Step A1) Preprocessing includes filtering out abnormal data and repairing missing data; filtering out abnormal data refers to identifying and removing spatial positioning offsets and speed anomalies in package delivery data, ride-hailing GPS trajectory data, and passenger order data; repairing missing data refers to filling in missing timestamps or geographic location data in ride-hailing GPS trajectory data.

[0045] Step A2) Linking the preprocessed ride-hailing GPS trajectory data with passenger order data includes: matching the driver and order IDs to generate driver order records; these driver order records include passenger order feature sets and ride-hailing driver attribute sets: the passenger order feature set includes the order's origin coordinates, destination coordinates, fare information, estimated mileage, and service period; the ride-hailing driver attribute set includes the vehicle's starting location, the driver's historical order frequency, and dispatch response distance.

[0046] In this embodiment, the associated ride-hailing driver order-taking data is shown in Table 1. Table 1 contains partial associated data:

[0047] Table 1

[0048] Passenger Order ID Trip start time End time of the trip Latitude and longitude of the boarding point Latitude and longitude of the drop-off point Passenger order price ride-hailing driver ID Driver order acceptance time Latitude and longitude of the trajectory point Ride-hailing dispatch distance

[0049] Step B: Calculation of the probability of ride-hailing drivers responding to orders:

[0050] By combining the behavioral characteristics of different drivers on ride-hailing platforms and based on their historical order records, a driver response model is built through machine learning.

[0051] Based on the driver response model, real-time orders are analyzed to predict the probability of each driver accepting an order given by the ride-hailing platform and the cost of empty driving.

[0052] Step C: Collaboration between ride-hailing services and drone delivery:

[0053] An initial ride-hailing-drone collaborative dispatching scheme was determined, which includes a ride-hailing order dispatching scheme and a drone delivery scheme; among which,

[0054] For ride-hailing order dispatching schemes, ride-hailing platforms consistently employ a one-to-many dispatching strategy. A single passenger order can be assigned to multiple drivers, but a single driver can only accept one order at a time. Specifically: for orders generated in real-time... For each passenger order, select the one with the highest probability of acceptance. One driver is used as the ride-hailing order dispatching scheme;

[0055] For drone delivery solutions, based on the initial extracted parcel delivery data, ride-hailing platforms select the parcel for each passenger order that minimizes the drone's empty-load distance. Specifically: for the real-time generated parcel delivery data... Each passenger order determines the set of parcel delivery orders within the drone's flight time. Calculate the empty distance for drone delivery of the package order, and select the package order with the shortest empty distance as the drone delivery solution.

[0056] The drone's airborne distance includes: the distance from the ride-hailing vehicle to the parcel delivery point, and the distance from the parcel's final destination to the ride-hailing vehicle's meeting point.

[0057] Step D: Determining the minimum distance collaborative scheduling scheme:

[0058] In this example, ride-hailing GPS trajectory data and passenger order data for one day in Chengdu are selected for platform order dispatch; the simulated parcel delivery point data is generated using the following method:

[0059] POI data is obtained from Chengdu's geographic information platform as an initial candidate set; based on the actual commercial distribution density in different areas of the city, a corresponding number of simulated parcel delivery points are generated in the main urban areas; latitude and longitude coordinates are assigned to each delivery point to form a dataset of parcel delivery points in Chengdu for the day; at the same time, the business hours attributes of each point can be associated.

[0060] With the goal of minimizing the empty driving distance of ride-hailing vehicles and the empty carrying distance of drones, a mathematical model for travel and delivery scheduling is constructed, which is expressed as follows:

[0061]

[0062] in, , , and These represent the number of passenger orders, ride-hailing drivers, drones, and candidate meeting points, respectively. Indicates the first Is the passenger order assigned to the first...? A ride-hailing driver; , They represent the first The, the The ride-hailing driver accepted the first The probability value of each passenger's order; Indicates the first The ride-hailing driver was performing the first Does the first passenger order enable the first One drone; Indicates the first The ride-hailing driver and the first Are the drones at the rendezvous point? Complete the first One passenger order; For the first A ride-hailing driver at the dispatch time Location, For the first The pick-up point for each passenger's order; , For the first The origin and delivery points for drone-delivered parcels; For the first The first drone One candidate rendezvous point location; , These are the distance functions for ride-hailing vehicles and drones, respectively. The time required for drones to load and unload packages. For the first The completion time of each passenger's order. This refers to the maximum time that ride-hailing drivers can stay at the meeting point. This represents the longest flight time of a drone when fully charged. and These are the speed parameters for ride-hailing vehicles and drones, respectively. , These are the weighting coefficients; This represents the objective function value of the travel and delivery scheduling model;

[0063] Preferably, The value is 8min. The value is 10 minutes. The value is 50 minutes. The value is 40km / h. The value is set to 75 km / h; The value is 0.7. The value is 0.3;

[0064] Step D1: Based on the hill-climbing algorithm HC in heuristic algorithms, find the ride-hailing-drone collaborative scheduling scheme that minimizes the objective function value of the travel and delivery scheduling model; including:

[0065] Step D11: The ride-hailing platform generates the first... For each passenger order, calculate the objective function value of the travel and delivery scheduling model corresponding to the initial ride-hailing-drone collaborative scheduling scheme;

[0066] Step D12: By partially adjusting the matching relationship between passenger orders and ride-hailing drivers, a new dispatch plan is generated. The partial adjustment of the matching relationship is as follows:

[0067] The dispatch objects of two passenger orders are swapped to explore neighborhood solutions. Each local adjustment only changes a finite number of items in the current scheduling scheme, thus forming a new scheduling scheme.

[0068] Step D13: Calculate the objective function value corresponding to the new scheduling scheme and compare it with the objective function value of the current scheduling scheme; if the objective function value corresponding to the new scheduling scheme is less than the objective function value corresponding to the current scheduling scheme, the ride-hailing platform replaces the current scheduling scheme with the new scheduling scheme; otherwise, retain the current scheduling scheme and continue to search for a new scheduling scheme in the remaining neighborhood.

[0069] Step D14: Repeat steps D11 to D13 until the objective function value of the scheduling scheme no longer changes or the iteration number C is reached. The scheduling scheme obtained at this time is the scheduling scheme with the minimum objective function value. Preferably, the value of C is 500.

[0070] Once the dispatch plan is determined, drones that are parked above ride-hailing vehicles are assigned to deliver goods. After the delivery is completed, the drones fly back to the destination where they received the passenger order at the departure time and meet up with the ride-hailing vehicle. During idle time, the drones are parked above the ride-hailing vehicles to charge.

[0071] In this example, considering package delivery time, we selected ride-hailing order data from Chengdu from 9:00 AM to 8:00 PM for one day, and compiled the objective function values ​​of the optimal scheduling scheme for each hour. The empty driving distance of ride-hailing vehicles (meters, m) and the empty carrying distance of drones (meters, m) are shown in Table 2:

[0072] Table 2 shows the optimal dispatching schemes for each hour. Ride-hailing empty driving distance, drone empty driving distance

[0073]

[0074] This invention reduces the empty driving distance of ride-hailing vehicles and the empty carrying distance of drones by coordinating the time and space of vehicles and drones, thereby improving the utilization efficiency of delivery and travel resources, optimizing platform revenue, and enhancing user experience.

[0075] The description and application of the present invention herein are illustrative and not intended to limit the scope of the invention to the embodiments described above. The effects or advantages described in the specification may not be apparent in actual experimental cases due to uncertainties in specific conditions or other factors, and such descriptions are not intended to limit the scope of the invention. Variations and modifications to the embodiments disclosed herein are possible, and various substitutions and equivalents of the components in the embodiments are well known to those skilled in the art. It should be understood by those skilled in the art that the invention can be implemented in other forms, structures, arrangements, proportions, and with other components, materials, and parts without departing from the spirit or essential characteristics of the invention. Other variations and modifications can be made to the embodiments disclosed herein without departing from the scope and spirit of the invention.

[0076] 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 variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A travel and delivery scheduling method based on ride-hailing-drone collaboration, characterized in that, include: Step A: Based on the large language model, preprocess the parcel delivery data, ride-hailing GPS trajectory data, and passenger order data; The pre-processed ride-hailing GPS trajectory data is associated with passenger order data to generate driver order records. At the same time, based on the driver order records, the parcel delivery data within the preset delivery area is initially extracted. Step B: Combining the behavioral characteristics of different drivers on ride-hailing platforms, and based on their historical order records, establish a driver response model through machine learning; Based on the driver response model, real-time orders are analyzed to predict the probability of each driver accepting an order given by the ride-hailing platform and the cost of empty driving. Step C: Determine the initial ride-hailing-drone collaborative dispatch plan, which includes a ride-hailing order dispatch plan and a drone delivery plan; among which, For ride-hailing order dispatching schemes, ride-hailing platforms always adopt a one-to-many dispatching strategy, selecting the multiple drivers with the highest order acceptance probability to dispatch orders. For drone delivery solutions, based on the initial extracted parcel delivery data, ride-hailing platforms select a parcel for each passenger order that minimizes the drone's empty-load distance. Step D: Optimize the initial ride-hailing-drone collaborative scheduling scheme; With the goal of minimizing the empty driving distance of ride-hailing vehicles and the empty carrying distance of drones, a travel and delivery scheduling model is constructed. Based on heuristic algorithms, a ride-hailing-drone collaborative scheduling scheme with the minimum objective function value of the travel and delivery scheduling model is found.

2. The method for travel and delivery scheduling based on ride-hailing-drone collaboration according to claim 1, characterized in that, In step C, the one-to-many dispatch strategy means that a single passenger order can be assigned by the ride-hailing platform. Each driver can only accept one order at a time.

3. The method for travel and delivery scheduling based on ride-hailing-drone collaboration according to claim 1, characterized in that, In step C, the unloaded distance of the drone includes: the distance from the ride-hailing vehicle to the parcel delivery location and the distance from the parcel's final destination to the ride-hailing vehicle meeting point.

4. The method for travel and delivery scheduling based on ride-hailing-drone collaboration according to claim 1, characterized in that, In step A, preprocessing includes filtering out abnormal data and repairing missing data. Filtering out abnormal data refers to identifying and removing spatial positioning offsets and speed anomalies from package delivery data, ride-hailing GPS trajectory data, and passenger order data. Repairing missing data refers to filling in missing timestamps or geographic location data in ride-hailing GPS trajectory data.

5. A travel and delivery scheduling method based on ride-hailing-drone collaboration according to claim 1, characterized in that, In step A, linking the preprocessed ride-hailing GPS trajectory data with passenger order data includes: matching the driver and order IDs to generate driver order records; these driver order records include passenger order feature sets and ride-hailing driver attribute sets: the passenger order feature set includes the order start coordinates, destination coordinates, fare information, estimated mileage, and service period; the ride-hailing driver attribute set includes the vehicle's starting location, the driver's historical order frequency, and dispatch response distance.

6. A travel and delivery scheduling method based on ride-hailing-drone collaboration according to claim 1, characterized in that, In step C, the ride-hailing dispatch plan and the drone delivery plan are determined as follows: Ride-hailing platforms generate real-time first For each passenger order, select the one with the highest probability of acceptance. One driver is used as the ride-hailing order dispatching scheme; For the real-time generated first Each passenger order determines the set of parcel delivery orders within the drone's flight time. Calculate the empty distance for drone delivery of the package order, and select the package order with the shortest empty distance as the drone delivery solution.

7. A travel and delivery scheduling method based on ride-hailing-drone collaboration according to claim 1, characterized in that, In step D, with the goal of minimizing the empty driving distance of ride-hailing vehicles and the empty carrying distance of drones, a travel and delivery scheduling model is constructed, which is expressed as follows: ； in, , , and These represent the number of passenger orders, ride-hailing drivers, drones, and candidate meeting points, respectively. Indicates the first Is the passenger order assigned to the first...? A ride-hailing driver; , They represent the first The, the The ride-hailing driver accepted the first The probability value of each passenger's order; Indicates the first The ride-hailing driver was performing the first Does the first passenger order enable the first One drone; Indicates the first The ride-hailing driver and the first Are the drones at the rendezvous point? Complete the first One passenger order; For the first A ride-hailing driver at the dispatch time Location, For the first The pick-up point for each passenger's order; , For the first The origin and delivery points for drone-delivered parcels; For the first The first drone One candidate rendezvous point location; , These are the distance functions for ride-hailing vehicles and drones, respectively. The time required for drones to load and unload packages. For the first The completion time of each passenger's order. This refers to the maximum time a ride-hailing driver can stay at the meeting point. This represents the longest flight time of a drone when fully charged. and These are the speed parameters for ride-hailing vehicles and drones, respectively. , These are the weighting coefficients; This represents the objective function value of the travel and delivery scheduling model.

8. A travel and delivery scheduling method based on ride-hailing-drone collaboration according to claim 1, characterized in that, In step D, based on the hill-climbing algorithm HC in the heuristic algorithm, we find the ride-hailing-drone collaborative scheduling scheme that minimizes the objective function value of the travel and delivery scheduling model; include: Step D11: The ride-hailing platform generates the first... For each passenger order, calculate the objective function value of the travel and delivery scheduling model corresponding to the initial ride-hailing-drone collaborative scheduling scheme; Step D12: By partially adjusting the matching relationship between passenger orders and ride-hailing drivers, a new dispatch plan is generated. The partial adjustment of the matching relationship is as follows: The dispatch objects of two passenger orders are swapped to explore neighborhood solutions. Each local adjustment only changes a finite number of items in the current scheduling scheme, thus forming a new scheduling scheme. Step D13: Calculate the objective function value corresponding to the new scheduling scheme and compare it with the objective function value of the current scheduling scheme; if the objective function value corresponding to the new scheduling scheme is less than the objective function value corresponding to the current scheduling scheme, the ride-hailing platform replaces the current scheduling scheme with the new scheduling scheme; otherwise, retain the current scheduling scheme and continue to search for a new scheduling scheme in the remaining neighborhood. Step D14: Repeat steps D11 to D13 until the objective function value of the scheduling scheme no longer changes or the iteration number is reached. The scheduling scheme obtained at this time is the scheduling scheme with the minimum objective function value.

9. A travel and delivery scheduling method based on ride-hailing-drone collaboration according to claim 1, characterized in that, Once the dispatch plan is determined, drones parked above ride-hailing vehicles are assigned to deliver goods. After delivery, the drones fly back to the destination where they received the passenger order at the departure time to meet the ride-hailing vehicle. During idle periods, the drones are parked above the ride-hailing vehicles to charge.

Images