Cooperative distribution system and method of unmanned aerial vehicle and automatic hub station

Abstract

The present invention relates to the technical field of drone management, and discloses a collaborative distribution system and method for drones and automatic hub stations. The system adopts a multi-level architecture, including a data acquisition layer, a data processing and analysis layer, a business logic layer, an application interface layer, and an infrastructure layer. By integrating drone status, environment, and mission data, the system realizes intelligent task management and scheduling, improves distribution efficiency, and reduces error rate. By using a machine learning model, the system can dynamically plan the optimal distribution path according to real-time environmental data, and effectively manage drone resources to ensure efficient use of resources. The design of the user interface and the system monitoring interface improves user experience and system management capabilities. The implementation of the present invention will promote the development of drone distribution technology and bring innovative solutions to the logistics industry.

Description

Technical Field

[0001] The present invention relates to the technical field of drone management, and in particular to a collaborative distribution system and method for drones and automatic hub stations. Background Art

[0002] With the rapid development of e-commerce and logistics industries, fast and accurate delivery services have become the key to meeting customer needs. Traditional ground delivery methods are unable to cope with complex traffic conditions and remote areas, and the delivery efficiency is low and the cost is high. In order to meet these challenges, drone delivery technology has emerged. With its high speed and flexibility, it has significant advantages in improving delivery efficiency.

[0003] However, existing drone delivery technology still has many defects in practical applications. On the one hand, the management and scheduling of drone delivery tasks lack intelligence, and the receipt, allocation and status tracking of tasks often rely on manual operations, which is not only inefficient but also prone to errors. On the other hand, the path planning technology is not mature enough and cannot be dynamically adjusted according to real-time environmental data (such as weather conditions, traffic conditions, etc.), resulting in a suboptimal delivery path, which may even become infeasible due to environmental changes.

[0004] In addition, there are also problems with the collaborative work of drones and automatic hubs in the existing technology. The resource scheduling issues such as drone takeoff, landing, charging and maintenance have not been well solved, resulting in low drone utilization and high delivery costs. At the same time, the lack of effective user interface and system monitoring interface makes it difficult for users to conveniently place orders, query delivery status and provide feedback, and it is difficult for system administrators to effectively monitor and manage the system. To this end, we propose a collaborative delivery system and method for drones and automatic hubs. Summary of the invention

[0005] The purpose of the present invention is to provide a collaborative distribution system and method for drones and automatic hub stations to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: A coordinated delivery system of drones and automatic hub stations, comprising a data collection layer, a data processing and analysis layer, a business logic layer, an application interface layer and an infrastructure layer;

[0007] The data collection layer includes a drone status data module, an environmental data module and a task data module; the drone status data module is used to collect the drone's flight status, battery power and load information; the environmental data module is used to collect weather conditions and traffic conditions in the delivery area; the task data module is used to record detailed information on the delivery task, including the start and end points, cargo type and priority;

[0008] The data processing and analysis layer includes a data preprocessing module, a machine learning model module and a real-time data analysis module; the data preprocessing module is used to clean, integrate and format the collected data to prepare for subsequent analysis and model training; the machine learning model module includes a path planning model, a task allocation model and a resource scheduling model, which are used to make optimal decisions based on historical data and real-time data; the real-time data analysis module is used to perform real-time analysis on the preprocessed data to achieve dynamic scheduling;

[0009] The business logic layer includes a task management module, a path planning module and a resource scheduling module; the task management module is responsible for receiving, assigning and tracking the status of tasks; the path planning module is used to plan the optimal delivery path for the drone based on the output of the machine learning model; the resource scheduling module is used to manage the take-off, landing, charging and maintenance of the drone to ensure the effective use of resources;

[0010] The application interface layer includes a user interface module and a system monitoring interface module; the user interface module is used to provide the functions of the user-side APP, including placing orders, querying delivery status and providing feedback; the system monitoring interface module is used to provide an interface for the system administrator to monitor and manage the system, including viewing the delivery progress and adjusting system parameters;

[0011] The infrastructure layer includes a drone platform module and an automatic hub station module; the drone platform module includes drone hardware, a flight control system and a communication system; the automatic hub station module includes an intelligent dispatching center, an automated warehousing system and a drone take-off and landing platform.

[0012] Preferably, the data preprocessing module includes a data cleaning unit, a data integration unit and a data formatting unit; the data cleaning unit is used to denoise, remove duplication and process outliers on the collected raw data; the data integration unit is used to merge and associate the cleaned data according to preset rules to form a complete data set; the data formatting unit is used to convert the data set into an input format accepted by the machine learning model, including standardization processing of numerical data and encoding processing of categorical data.

[0013] Preferably, the data formatting unit uses a Z-score standardization method to subtract the mean value from each numerical data and divide it by its standard deviation, so that the processed data conforms to a standard normal distribution. The formula for standardization is: ,in, is the original data, is the mean of the original data, is the standard deviation of the original data, is the standardized data.

[0014] Preferably, the machine learning model module includes a path planning model unit, a task allocation model unit and a resource scheduling model unit;

[0015] The path planning model unit is used to plan the optimal delivery path for the drone based on historical delivery data and real-time environmental data, and is implemented in the following way:

[0016] The shortest path algorithm based on Dijkstra algorithm: For a given distribution area graph, take the starting point as the source point, calculate the shortest path to each node, and select the shortest path to the end point as the distribution path; the calculation formula is: dist[v] =min(dist[v], dist[u] + weight(u, v)), where dist[v] represents the shortest path length from the source point to vertex v, dist[u] represents the shortest path length from the source point to vertex u, and weight(u, v) represents the weight of the edge from vertex u to vertex v;

[0017] Path adjustment algorithm combined with real-time environmental factors: According to real-time weather conditions and traffic conditions, the path planned based on the Dijkstra algorithm is dynamically adjusted to avoid unfavorable factors; the calculation formula is: adjusted_weight(u, v) = original_weight(u, v) * (1 + adjustment_factor), where adjusted_weight(u, v) represents the adjusted edge weight, original_weight(u, v) represents the original edge weight, and adjustment_factor represents the adjustment factor calculated according to real-time environmental factors.

[0018] Preferably, the task allocation model unit is used to automatically allocate tasks to appropriate drones according to current task requirements and drone status, and is implemented in the following manner:

[0019] A benefit matrix of the task allocation problem is constructed, and the Hungarian algorithm is used to solve the task allocation plan with the maximum benefit. According to the real-time status of the UAV, the tasks allocated based on the Hungarian algorithm are dynamically adjusted to ensure the rationality and feasibility of the task allocation. The real-time status of the UAV includes battery power and load.

[0020] Preferably, the resource scheduling model unit is implemented using a time window-based resource scheduling algorithm, specifically: a take-off time window, a landing time window and a charging time window are set for each drone, each time window has a start time and an end time, indicating that the drone can perform corresponding operations within the time window; during the scheduling process, the current status and time window of each drone are checked, if the current time is within the take-off time window of a drone, the drone is scheduled to take off; if the current time is within the landing time window of a drone, the drone is scheduled to land; if the current time is within the charging time window of a drone, the drone is scheduled to charge.

[0021] Preferably, the task management module includes a task receiving unit, a task assigning unit and a task monitoring unit; the task receiving unit is used to receive delivery task requests from users or other systems, and store task information in a task database, and the task information includes the delivery starting point, delivery end point, cargo weight and delivery time requirements.

[0022] Preferably, the task allocation unit is used to allocate the received delivery task to a suitable drone according to the current drone status, task priority, and delivery area restriction factors, and the specific implementation method is as follows:

[0023] Drone status assessment: Obtain drone status information in real time, including battery power, flight speed, and load capacity, and assess the drone’s availability and mission execution capabilities based on this information;

[0024] Task prioritization: Prioritize the received delivery tasks according to the delivery time requirements and the importance of the goods, so that important and urgent tasks can be handled first;

[0025] Delivery area restriction handling: Consider the restrictions of the delivery area, including no-fly zones and altitude restrictions, to ensure that the assigned tasks are in line with the flight capabilities of the drone and legal and regulatory requirements;

[0026] Task allocation algorithm: A constraint satisfaction-based optimization algorithm is used to comprehensively consider the drone status, task priority, and delivery area restrictions, assign tasks to the most suitable drone, and generate a task execution plan.

[0027] Preferably, the task monitoring unit is used to monitor the execution of the task in real time, including the flight status, task progress and abnormal conditions of the UAV, and feed back the monitoring results to the task allocation unit for dynamic adjustment. The specific implementation method is as follows:

[0028] Flight status monitoring: Through the communication link between the drone and the automatic hub, the flight status information of the drone, including position, speed and altitude, can be obtained in real time;

[0029] Mission progress tracking: Calculate the mission progress in real time based on the UAV’s flight status and mission execution plan, and update relevant information in the mission database;

[0030] Abnormal situation handling: Abnormal situations found during the monitoring process, including drone failures and sudden weather changes, are handled promptly, and the task allocation unit is notified to reallocate or adjust the tasks.

[0031] Preferably, the user interface module includes a user interaction interface, an input processing unit and an output display unit; the user interaction interface is used to provide a graphical or command line interaction method, providing users with input of delivery task information, query of task status and setting of system parameters, and receiving system feedback information; the input processing unit is used to parse and process the information input by the user through the user interaction interface, including extraction of delivery task information, verification of task parameters and recognition of user instructions; the output display unit is used to display the results of system processing to the user in graphical or textual form, including task execution status, drone flight trajectory and delivery progress.

[0032] Preferably, a collaborative delivery method of a drone and an automatic hub station comprises the following steps:

[0033] Step 1: receiving the delivery task information, including the delivery starting point, delivery destination, cargo weight and delivery time requirements, input through the user interaction interface of the user interface module, and parsing and verifying by the input processing unit;

[0034] Step 2: Based on the received delivery task information, the task allocation unit of the task management module comprehensively considers the current status of the drone, task priority, and delivery area restrictions, allocates the task to the most suitable drone, and generates a task execution plan;

[0035] Step 3: The path planning model unit plans the optimal delivery path for the assigned drone based on historical delivery data and real-time environmental data, adjusts the path considering real-time environmental factors, and performs path smoothing and altitude planning;

[0036] Step 4: The drone takes off according to the planned delivery route and performs the delivery task. At the same time, the resource scheduling model unit manages the takeoff, landing, charging and maintenance resource scheduling of the drone to ensure the normal operation of the drone;

[0037] Step 5: The task monitoring unit of the task management module monitors the execution of the task in real time, including the flight status of the drone, task progress and abnormal conditions, and feeds back the monitoring results to the task allocation unit for dynamic adjustment;

[0038] Step 6: When the drone completes the mission, it returns to the automatic hub station for charging and maintenance. At the same time, the output display unit displays the mission execution status, drone flight trajectory and delivery progress information to the user for query and feedback.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] Through intelligent task management and scheduling, automatic task reception, allocation and status tracking are achieved, which significantly improves processing speed and reduces error rate.

[0041] The path planning module, combined with the machine learning model, can dynamically adjust the delivery path based on real-time environmental data to ensure the optimal path, thereby shortening the delivery time and improving overall delivery efficiency.

[0042] Reduced distribution costs: The resource scheduling module effectively manages the takeoff, landing, charging and maintenance of drones, improves the utilization rate of drones, and reduces the additional costs caused by improper resource scheduling. The automated warehousing system and drone take-off and landing platform reduce manual intervention and further reduce operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is the overall structure diagram of the present invention;

[0044] Figure 2 Flowchart for planning the optimal delivery route for the route planning model unit;

[0045] Figure 3 A working principle diagram of automatic task assignment for the task assignment model unit;

[0046] Figure 4 This is a step diagram of a collaborative delivery method between a drone and an automatic hub station described in the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] See also Figure 1-Figure 4 The present invention provides a technical solution: a coordinated delivery system of a drone and an automatic hub station, comprising:

[0049] Data collection layer:

[0050] Drone status data module: collects the drone’s flight status, battery power, and load information in real time through the drone’s sensors and flight control system.

[0051] Environmental data module: Utilizes data from weather stations and traffic monitoring systems to collect weather conditions and traffic conditions in the distribution area, providing real-time environmental information for route planning and task allocation.

[0052] Task data module: records detailed information of delivery tasks, including starting and ending points, cargo type and priority, to ensure the accuracy and completeness of task information.

[0053] Data processing and analysis layer:

[0054] Data preprocessing module: cleans, integrates and formats the collected data, removes noise and redundant data, and prepares for subsequent analysis and model training.

[0055] Machine learning model module: includes path planning model, task allocation model and resource scheduling model, and uses historical data and real-time data to train the model to make the best decision.

[0056] Real-time data analysis module: performs real-time analysis on pre-processed data, monitors data changes, and implements dynamic scheduling based on analysis results.

[0057] Business logic layer:

[0058] Task management module: responsible for receiving, allocating and tracking the status of tasks, ensuring that tasks can be correctly executed according to priorities and constraints.

[0059] Path planning module: Based on the output of the machine learning model, it plans the optimal delivery path for the drone and makes dynamic adjustments considering real-time environmental factors.

[0060] Resource scheduling module: manages the takeoff, landing, charging and maintenance of drones, optimizes resource utilization, and improves the operating efficiency of drones.

[0061] Application interface layer:

[0062] User interface module: provides user-side APP functions, including placing orders, checking delivery status and providing feedback, to improve user experience.

[0063] System monitoring interface module: provides system administrators with an interface for monitoring and managing the system, including viewing delivery progress and adjusting system parameters to ensure stable operation of the system.

[0064] Infrastructure layer:

[0065] UAV platform module: includes UAV hardware, flight control system and communication system, providing basic support for UAV flight and mission execution.

[0066] Automatic hub station module: includes an intelligent dispatching center, an automated warehousing system, and a drone take-off and landing platform to achieve automated management and maintenance of drones.

[0067] The present invention will be further described below in conjunction with Examples 1 to 3:

[0068] Embodiment 1:

[0069] The data preprocessing module is responsible for preparing high-quality data sets for subsequent analysis and model training, which specifically includes data cleaning unit, data integration unit and data formatting unit.

[0070] Data cleaning unit: This unit mainly performs preprocessing on the collected raw data, including denoising, deduplication and outlier processing. Denoising improves the signal-to-noise ratio of the data by filtering out random errors and interference in the data; deduplication eliminates duplicate records in the data to ensure the uniqueness of the data; outlier processing improves the overall quality of the data by identifying and removing outliers in the data. For missing values, linear interpolation algorithm is used for filling; for outliers, threshold judgment method is used to treat data points that exceed the preset threshold range as outliers and replace them with the mean of their neighboring points.

[0071] Data integration unit: After data cleaning, this unit will merge and associate the cleaned data according to preset rules. This includes matching and splicing data from different sources to form a complete data set and ensuring the consistency and relevance between data. Through data integration, data silos can be eliminated and the overall value of data can be improved.

[0072] Data formatting unit: This unit is responsible for converting the data set into the input format accepted by the machine learning model. It includes the standardization of numerical data and the encoding of categorical data. Specifically, for numerical data, the Z-score standardization method is used. This method calculates the difference between each numerical data and its mean and divides it by its standard deviation so that the processed data conforms to the standard normal distribution. The formula for standardization is:

[0073]

[0074] in, is the original data, is the mean of the original data, is the standard deviation of the original data, The data is standardized. This standardization helps to eliminate the differences between data of different dimensions and improve the training effect and prediction accuracy of the model. For categorical data, encoding processing, such as One-Hot Encoding, is used to convert categorical data into a numerical form that can be processed by the machine learning model.

[0075] The machine learning model module includes a path planning model unit, a task allocation model unit and a resource scheduling model unit, which are used to realize the planning of drone delivery paths, the allocation of tasks and the scheduling of resources respectively.

[0076] The path planning model unit is responsible for planning the optimal delivery path for the drone based on historical delivery data and real-time environmental data. The specific implementation method is as follows:

[0077] a. Shortest path algorithm based on Dijkstra algorithm:

[0078] For a given distribution area graph, the Dijkstra algorithm is used to calculate the shortest path to each node with the starting point as the source point, and the shortest path to the end point is selected as the distribution path. The core calculation formula of the Dijkstra algorithm is: dist[v] = min(dist[v], dist[u] + weight(u, v)). Among them, dist[v] represents the shortest path length from the source point to vertex v, dist[u] represents the shortest path length from the source point to vertex u, and weight(u, v) represents the weight of the edge from vertex u to vertex v. Through iterative calculation, the shortest path length from the source point to each vertex is continuously updated until the shortest path to the end point is found.

[0079] b. Path adjustment algorithm combined with real-time environmental factors:

[0080] Taking into account the impact of real-time weather conditions and traffic conditions on the delivery path, this unit also implements a path adjustment algorithm that combines real-time environmental factors. The algorithm dynamically adjusts the path planned based on the Dijkstra algorithm according to real-time environmental factors (such as wind direction, wind speed, rainfall, traffic congestion, etc.) to avoid unfavorable factors. The specific calculation formula is: adjusted_weight(u, v) = original_weight(u, v) * (1 + adjustment_factor). Among them, adjusted_weight(u, v) represents the weight of the adjusted edge, original_weight(u, v) represents the original edge weight, and adjustment_factor represents the adjustment factor calculated based on real-time environmental factors. By adjusting the weight of the edge, the planned path can be made more in line with the real-time environmental conditions, thereby improving the efficiency and safety of distribution.

[0081] The task allocation model unit is responsible for automatically allocating tasks to appropriate drones based on current task requirements and drone status. The specific implementation method is as follows:

[0082] a. Construct the benefit matrix of the task allocation problem:

[0083] First, a benefit matrix is ​​constructed based on the mission requirements and drone capabilities. The rows of the matrix represent the missions to be assigned, and the columns represent the available drones. Each element in the matrix represents the benefit value that can be obtained by assigning a specific mission to a specific drone. This value can be calculated based on factors such as the priority of the mission, the performance of the drone, and the historical delivery efficiency.

[0084] b. Use the Hungarian algorithm to solve the task allocation solution with the greatest benefit:

[0085] After constructing the benefit matrix, the Hungarian algorithm (also known as the KM algorithm or Kuhn-Munkres algorithm) is used to solve the task allocation solution with the maximum benefit. The Hungarian algorithm is a classic algorithm for solving the task allocation problem in polynomial time, which can find the task allocation solution that maximizes the total benefit.

[0086] c. Dynamic adjustment based on the real-time status of the drone:

[0087] Considering that the real-time status of the drone (such as battery power and load) may affect the rationality and feasibility of task allocation, this unit also implements a dynamic task adjustment mechanism based on the real-time status of the drone. Specifically, after deriving a preliminary task allocation plan based on the Hungarian algorithm, the system will monitor the state changes of the drone in real time and dynamically adjust the task allocation plan based on the state changes. For example, if a drone has low battery power or too much load, the system may reallocate its originally assigned tasks to other drones in better condition to ensure that the task can be completed smoothly.

[0088] The resource scheduling model unit is implemented using a time window-based resource scheduling algorithm, which aims to reasonably arrange the take-off, landing and charging time of the drone to ensure the effective use of resources and the efficient execution of tasks. The specific implementation method is as follows:

[0089] Set time windows: Set takeoff time windows, landing time windows, and charging time windows for each drone. Each time window consists of a start time and an end time, indicating that the drone can perform corresponding operations within the time window. The takeoff time window is used to specify when the drone can take off to perform a mission, the landing time window is used to specify when the drone needs to land and return to the base, and the charging time window is used to specify when the drone needs to be charged to ensure sufficient power.

[0090] Scheduling process: During the resource scheduling process, the system will check the current status and time window of each drone in real time. If the current time is within the take-off time window of a drone, and the drone is in good condition and has sufficient power, the system will schedule the drone to take off and perform the mission. If the current time is within the landing time window of a drone, or the drone has completed the mission and needs to return to charge due to low power, the system will schedule the drone to land and return to the base. Similarly, if the current time is within the charging time window of a drone, and the drone needs to be charged due to low power, the system will schedule the drone to perform charging operations.

[0091] Dynamic adjustment: The resource scheduling algorithm based on the time window also has a certain dynamic adjustment capability. In the actual operation process, if there are emergencies or unexpected events (such as sudden weather changes, traffic control, etc.), the system can dynamically adjust the take-off, landing and charging time windows of the drone according to the actual situation to ensure the rational use of resources and efficient execution of tasks. For example, if a drone encounters bad weather during a mission and needs to make an emergency landing, the system can temporarily adjust its landing time window and arrange for it to land as soon as possible; if a drone is low on power but is not currently within the charging time window, the system can also flexibly adjust the charging time window according to the actual situation and arrange for it to charge.

[0092] Embodiment 2:

[0093] The task management module is responsible for handling various operations related to the delivery task, including the task receiving unit, task allocation unit and task monitoring unit.

[0094] Task receiving unit: responsible for receiving delivery task requests from users or other systems. When receiving a task request, it will first parse the task information in the request, including key data such as the delivery starting point, delivery destination, cargo weight, and delivery time requirements. Then, this task information is stored in the task database for subsequent processing and query. The design of the task database takes into account the security, integrity, and query efficiency of the data to ensure that the task information can be retrieved and used accurately and quickly.

[0095] Task allocation unit: responsible for reasonably allocating the received delivery tasks to appropriate drones based on the current drone status, task priority, and restrictions in the delivery area. The specific implementation method is as follows:

[0096] Drone status assessment: The system obtains the status information of the drone in real time, including key indicators such as battery power, flight speed and load capacity. Based on this information, the system will conduct a comprehensive assessment of the drone's availability and mission execution capabilities. For example, the system will check whether the drone's battery power is sufficient to complete the delivery task, whether the flight speed meets the time requirements of the task, and whether the load capacity can carry the cargo. Through this evaluation process, the system can ensure that tasks are only assigned to drones that are in good condition and capable of performing the task.

[0097] Task prioritization: The system prioritizes the received delivery tasks based on the delivery time requirements and the importance of the goods. This sorting mechanism ensures that important and urgent tasks are handled first. For example, for medical supplies or high-value goods that need to be delivered urgently, the system will give these tasks a higher priority and ensure that they can be delivered by drone as soon as possible.

[0098] Delivery area restriction handling: When assigning tasks, the system will take into account the restrictions of the delivery area. This includes legal and regulatory requirements such as no-fly zones and altitude restrictions. The system will ensure that the assigned tasks are consistent with the flight capabilities of the drone and do not violate any relevant laws and regulations. For example, if a delivery area is set as a no-fly zone, the system will not assign any tasks to drones that need to fly over that area.

[0099] Task allocation algorithm: In order to achieve efficient and reasonable task allocation, the system adopts an optimization algorithm based on constraint satisfaction. This algorithm comprehensively considers multiple factors such as drone status, task priority, and delivery area restrictions. By calculating and comparing the costs and benefits of different allocation schemes, it finally allocates the task to the most suitable drone and generates a detailed task execution plan. This plan includes key information such as the drone's take-off time, flight route, landing time, and charging plan to ensure that the drone can successfully complete the task as planned.

[0100] The mission monitoring unit is responsible for real-time monitoring of mission execution. It will regularly retrieve mission information from the mission database and match and compare it with the real-time status of the drone. By monitoring the progress of the mission, the drone's location, power, load and other status information, the mission monitoring unit can promptly discover and resolve potential problems and risks. For example, if a drone fails or runs out of power during a mission, the mission monitoring unit will immediately issue an alarm and notify relevant personnel to handle it. At the same time, it will dynamically adjust the task allocation plan based on the urgency of the mission and the availability of the drone to ensure the smooth completion of the mission.

[0101] Embodiment 3:

[0102] The user interface module focuses on providing various functions of the user-side APP, ensuring that users can easily place orders, check delivery status and provide feedback. In order to achieve these functions, the user interface module is further subdivided into the user interaction interface, input processing unit and output display unit.

[0103] The user interface is a window for users to interact with the system, providing graphical or command line interaction. Users can use this interface to input delivery task information, such as the starting point, destination, cargo weight, etc.; query the current status of the task, such as whether it has taken off, estimated arrival time, etc.; and set system parameters, such as notification preferences, etc. At the same time, the interface will also receive and display system feedback information, such as successful task acceptance, drone location update, etc.

[0104] The input processing unit is responsible for parsing and processing the information input by the user through the interactive interface. It extracts key information of the delivery task, such as the delivery address, the type of goods, etc.; verifies the task parameters to ensure that the input information meets the system requirements; and identifies the user's instructions, such as placing an order, canceling a task, etc. Through this series of processing, the input processing unit can ensure that the information input by the user is accurately and effectively used for subsequent task processing.

[0105] Output display unit: responsible for displaying the results of system processing in a user-friendly way. Whether it is the task execution status, drone flight trajectory or delivery progress, the output display unit will clearly display it to the user in the form of graphics or text. In this way, the user can understand the latest developments of the task in real time and perform further operations or inquiries as needed.

[0106] A method for implementing a coordinated delivery method of a drone and an automatic hub station includes the following steps:

[0107] Step 1: receiving the delivery task information, including the delivery starting point, delivery destination, cargo weight and delivery time requirements, input through the user interaction interface of the user interface module, and parsing and verifying by the input processing unit;

[0108] Step 2: Based on the received delivery task information, the task allocation unit of the task management module comprehensively considers the current status of the drone, task priority, and delivery area restrictions, allocates the task to the most suitable drone, and generates a task execution plan;

[0109] Step 3: The path planning model unit plans the optimal delivery path for the assigned drone based on historical delivery data and real-time environmental data, adjusts the path considering real-time environmental factors, and performs path smoothing and altitude planning;

[0110] Step 4: The drone takes off according to the planned delivery route and performs the delivery task. At the same time, the resource scheduling model unit manages the takeoff, landing, charging and maintenance resource scheduling of the drone to ensure the normal operation of the drone;

[0111] Step 5: The task monitoring unit of the task management module monitors the execution of the task in real time, including the flight status of the drone, task progress and abnormal conditions, and feeds back the monitoring results to the task allocation unit for dynamic adjustment;

[0112] Step 6: When the drone completes the mission, it returns to the automatic hub station for charging and maintenance. At the same time, the output display unit displays the mission execution status, drone flight trajectory and delivery progress information to the user for query and feedback.

[0113] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0114] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The coordinated delivery system of drones and automatic hubs is characterized by: It includes data collection layer, data processing and analysis layer, business logic layer, application interface layer and infrastructure layer; The data collection layer includes a drone status data module, an environmental data module and a task data module; the drone status data module is used to collect the drone's flight status, battery power and load information; the environmental data module is used to collect weather conditions and traffic conditions in the delivery area; the task data module is used to record detailed information on the delivery task, including the start and end points, cargo type and priority; The data processing and analysis layer includes a data preprocessing module, a machine learning model module and a real-time data analysis module; the data preprocessing module is used to clean, integrate and format the collected data to prepare for subsequent analysis and model training; the machine learning model module includes a path planning model, a task allocation model and a resource scheduling model, which are used to make optimal decisions based on historical data and real-time data; the real-time data analysis module is used to perform real-time analysis on the preprocessed data to achieve dynamic scheduling; The business logic layer includes a task management module, a path planning module and a resource scheduling module; The task management module is responsible for receiving, assigning and tracking the status of tasks; the path planning module is used to plan the optimal delivery path for the drone based on the output of the machine learning model; the resource scheduling module is used to manage the takeoff, landing, charging and maintenance of the drone to ensure the effective use of resources; The application interface layer includes a user interface module and a system monitoring interface module; the user interface module is used to provide the functions of the user-side APP, including placing orders, querying delivery status and providing feedback; the system monitoring interface module is used to provide an interface for the system administrator to monitor and manage the system, including viewing the delivery progress and adjusting system parameters; The infrastructure layer includes a drone platform module and an automatic hub module; the drone platform module includes drone hardware, flight control system and communication system; the automatic hub module includes an intelligent dispatching center, an automated warehousing system and a drone take-off and landing platform; The machine learning model module includes a path planning model unit, a task allocation model unit and a resource scheduling model unit; The path planning model unit is used to plan the optimal delivery path for the drone based on historical delivery data and real-time environmental data, and is implemented in the following way: The shortest path algorithm based on Dijkstra algorithm: For a given distribution area graph, take the starting point as the source point, calculate the shortest path to each node, and select the shortest path to the end point as the distribution path; The calculation formula is: dist[v] = min(dist[v], dist[u] + weight(u,v)), where dist[v] represents the shortest path length from the source point to vertex v, dist[u] represents the shortest path length from the source point to vertex u, and weight(u,v) represents the weight of the edge from vertex u to vertex v; Path adjustment algorithm combined with real-time environmental factors: According to real-time weather conditions and traffic conditions, the path planned based on the Dijkstra algorithm is dynamically adjusted to avoid unfavorable factors; the calculation formula is: adjusted_weight(u,v)=original_weight(u,v)*(1+adjustment_factor), where adjusted_weight(u,v) represents the weight of the adjusted edge, original_weight(u,v) represents the original weight of the edge, and adjustment_factor represents the adjustment factor calculated according to real-time environmental factors; The task allocation model unit is used to automatically allocate tasks to appropriate drones according to current task requirements and drone status, and is implemented in the following manner: constructing a benefit matrix for the task allocation problem, and using the Hungarian algorithm to solve the task allocation plan with the maximum benefit; dynamically adjusting the tasks allocated based on the Hungarian algorithm according to the real-time status of the drone to ensure the rationality and feasibility of the task allocation; the real-time status of the drone includes battery power and load; The resource scheduling model unit is implemented by a time window-based resource scheduling algorithm, specifically: a take-off time window, a landing time window and a charging time window are set for each drone, each time window has a start time and an end time, indicating that the drone can perform corresponding operations within the time window; during the scheduling process, the current state and time window of each drone are checked, if the current time is within the take-off time window of a drone, the drone is scheduled to take off; if the current time is within the landing time window of a drone, the drone is scheduled to land; if the current time is within the charging time window of a drone, the drone is scheduled to charge; The task management module includes a task receiving unit, a task assigning unit and a task monitoring unit; the task receiving unit is used to receive a delivery task request from a user or other system and store the task information in a task database, the task information including the delivery starting point, delivery destination, cargo weight and delivery time requirements; The task allocation unit is used to allocate the received delivery task to the appropriate drone according to the current drone status, task priority, and delivery area restriction factors. The specific implementation method is as follows: Drone status assessment: Obtain drone status information in real time, including battery power, flight speed, and load capacity, and assess the drone’s availability and mission execution capabilities based on this information; Task prioritization: Prioritize the received delivery tasks according to the delivery time requirements and the importance of the goods, so that important and urgent tasks can be handled first; Delivery area restriction handling: Consider the restrictions of the delivery area, including no-fly zones and altitude restrictions, to ensure that the assigned tasks are in line with the flight capabilities of the drone and legal and regulatory requirements; Task allocation algorithm: A constraint satisfaction-based optimization algorithm is used to comprehensively consider the drone status, task priority, and delivery area restrictions, assign tasks to the most suitable drone, and generate a task execution plan.

2. The coordinated delivery system of drones and automatic hubs according to claim 1 is characterized by: The data preprocessing module includes a data cleaning unit, a data integration unit and a data formatting unit; the data cleaning unit is used to denoise, remove duplication and process outliers on the collected original data; the data integration unit is used to merge and associate the cleaned data according to preset rules to form a complete data set; the data formatting unit is used to convert the data set into an input format accepted by the machine learning model, including standardization processing of numerical data and encoding processing of categorical data.

3. The coordinated delivery system of drones and automatic hubs according to claim 2 is characterized by: The data formatting unit uses the Z-score standardization method to subtract the mean value from each numerical data and divide it by its standard deviation, so that the processed data conforms to the standard normal distribution. The formula for standardization is: Among them, X is the original data, μ is the mean of the original data, σ is the standard deviation of the original data, and Z is the standardized data.

4. The coordinated delivery system of drones and automatic hubs according to claim 1 is characterized by: The task monitoring unit is used to monitor the execution of the task in real time, including the flight status, task progress and abnormal conditions of the UAV, and feed back the monitoring results to the task allocation unit for dynamic adjustment. The specific implementation method is as follows: Flight status monitoring: Through the communication link between the drone and the automatic hub, the flight status information of the drone, including position, speed and altitude, can be obtained in real time; Mission progress tracking: Calculate the mission progress in real time based on the UAV’s flight status and mission execution plan, and update relevant information in the mission database; Abnormal situation handling: Abnormal situations found during the monitoring process, including drone failures and sudden weather changes, are handled promptly, and the task allocation unit is notified to reallocate or adjust the tasks.

5. The coordinated delivery system of drones and automatic hubs according to claim 1 is characterized by: The user interface module includes a user interaction interface, an input processing unit and an output display unit; the user interaction interface is used to provide a graphical or command line interaction method, providing users with input of delivery task information, query of task status and setting of system parameters, and receiving system feedback information; the input processing unit is used to parse and process the information input by the user through the user interaction interface, including extraction of delivery task information, verification of task parameters and recognition of user instructions; the output display unit is used to display the results of system processing to the user in the form of graphics or text, including task execution status, UAV flight trajectory and delivery progress.

6. A method for coordinated delivery of drones and automatic hubs, applied to the coordinated delivery system of drones and automatic hubs as claimed in claim 1, characterized in that: The following steps are involved: Step 1: receiving the delivery task information, including the delivery starting point, delivery destination, cargo weight and delivery time requirements, input through the user interaction interface of the user interface module, and parsing and verifying by the input processing unit; Step 2: Based on the received delivery task information, the task allocation unit of the task management module comprehensively considers the current status of the drone, task priority, and delivery area restrictions, allocates the task to the most suitable drone, and generates a task execution plan; Step 3: The path planning model unit plans the optimal delivery path for the assigned drone based on historical delivery data and real-time environmental data, adjusts the path considering real-time environmental factors, and performs path smoothing and altitude planning; Step 4: The drone takes off according to the planned delivery route and performs the delivery task. At the same time, the resource scheduling model unit manages the takeoff, landing, charging and maintenance resource scheduling of the drone to ensure the normal operation of the drone; Step 5: The task monitoring unit of the task management module monitors the execution of the task in real time, including the flight status of the drone, task progress and abnormal conditions, and feeds back the monitoring results to the task allocation unit for dynamic adjustment; Step 6: When the drone completes the mission, it returns to the automatic hub station for charging and maintenance. At the same time, the output display unit displays the mission execution status, drone flight trajectory and delivery progress information to the user for query and feedback.

Images