Low-altitude logistics unmanned aerial vehicle full-life-cycle management and control system based on digital twinning

Abstract

The invention discloses a digital twinning-based low-altitude logistics unmanned aerial vehicle full-life-cycle management and control system, relates to the technical field of low-altitude logistics management and control, and aims to solve the technical problem that the current low-altitude logistics unmanned aerial vehicle state management and control maintenance is limited, and the digital twinning-based low-altitude logistics unmanned aerial vehicle full-life-cycle management and control system comprises a physical unmanned aerial vehicle and a real-time interactive mapping virtual digital twinning body corresponding to the physical unmanned aerial vehicle, and the predictive maintenance unit calculates the residual service life of each component by adopting a time sequence prediction algorithm based on the potential fault diagnosis result and the historical operation data of the unmanned aerial vehicle. According to the invention, through real-time interactive mapping of a physical unmanned aerial vehicle and a virtual digital twinborn body, a multi-source data fusion driving module is combined to dynamically simulate the performance state and the working environment of the unmanned aerial vehicle, and based on a potential fault result identified by a fault intelligent diagnosis unit and historical operation data of the unmanned aerial vehicle, the real-time interactive mapping is realized; a time sequence prediction algorithm is adopted to accurately calculate the remaining service life of each part, a traditional rigid regular overhaul mode is replaced, and maintenance resource waste and fault sudden risks are avoided.

Description

Technical Field

[0001] This invention relates to the field of low-altitude logistics management and control technology, and more specifically, to a low-altitude logistics drone full lifecycle management and control system based on digital twins. Background Technology

[0002] Currently, low-altitude logistics drones face the challenge of full lifecycle management in complex operating environments. Existing management systems often focus on monitoring single aspects, making it difficult to achieve comprehensive mapping and dynamic tracking of the physical drone's status. Due to the lack of deep integration with digital twin technology, multi-source data such as drone flight attitude and power operation cannot be efficiently integrated, resulting in insufficient real-time simulation and precise control of flight status. Consequently, it is difficult to identify potential faults in advance, seriously affecting the stability and safety of logistics tasks.

[0003] Meanwhile, existing drone maintenance largely relies on scheduled inspections, lacking targeted predictive maintenance mechanisms and failing to effectively predict potential faults based on localized data. Traditional maintenance methods ignore individual drone operational differences and the dynamic changes in component wear, making it impossible to accurately estimate the remaining lifespan of core components using time-series prediction algorithms, and also difficult to generate personalized maintenance plans. This not only wastes maintenance resources but also increases flight risks due to sudden faults, failing to meet the needs of large-scale, routine low-altitude logistics operations. Therefore, we propose a digital twin-based full lifecycle management system for low-altitude logistics drones. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology, adapt to the needs of reality, and provide a digital twin-based full life cycle management system for low-altitude logistics drones to solve the current technical problem of limited status management and maintenance of low-altitude logistics drones.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: a low-altitude logistics drone full lifecycle management system based on digital twins, comprising a physical drone and a corresponding virtual digital twin with real-time interactive mapping. The system further includes a multi-source data fusion driving module that collects data from the physical drone and drives the virtual digital twin to dynamically simulate the performance status and operating environment of the physical drone in a virtual state. The virtual digital twin integrates a fault intelligent diagnosis unit for identifying potential faults in the physical drone, a predictive maintenance unit for formulating maintenance strategies, a task simulation and extrapolation unit for simulating flight missions, and a risk assessment unit for evaluating flight mission risks and outputting optimized scheduling strategies. The predictive maintenance unit, based on potential fault diagnosis results and historical drone operation data, uses a time-series prediction algorithm to calculate the remaining service life of each component and generates a personalized maintenance plan. The maintenance plan includes maintenance time nodes, a list of maintenance components, and maintenance operation specifications. The task simulation and extrapolation unit supports multi-scenario parameter configuration and can simulate the flight mission execution process under different wind speeds, precipitation, visibility, and airspace congestion conditions, and outputs the extrapolation results of mission completion time, energy consumption, and path deviation.

[0006] Preferably, the predictive maintenance unit realizes real-time data transmission between the physical UAV and the corresponding virtual digital twin based on the multi-source data fusion driving module. The predictive maintenance unit performs localized data coverage estimation on the operational data output by the physical UAV, that is, it performs data estimation on the virtual digital twin based on the output of the multi-source data fusion driving module to obtain potential fault data of the physical UAV. This localized data coverage estimation is based on the following formula:

[0007] in for Time of the first The remaining service life of each core component For time series prediction functions, For the first The inherent local preset parameters of each component, For the first Total design life of each component For the first Each component and Momentary fusion data For the first Lifetime loss function of each component This refers to the current state data output from the physical drone to the virtual digital twin based on the multi-source data fusion driving module.

[0008] Preferably, the virtual digital twin further includes a physical property modeling module, a real-time data synchronization module, and a simulation calculation module. The physical property modeling submodule locally replicates the structural parameters, performance parameters, and material properties of the physical UAV based on its factory data. The real-time data synchronization module enables the physical UAV to synchronize data with the virtual digital twin based on the multi-source data fusion driving module, and performs data extraction simulation based on the simulation calculation module, outputting the current state data to the fault intelligent diagnosis unit.

[0009] Preferably, the data collected by the multi-source data fusion driving module includes the flight attitude data of the physical UAV, power system operation data, battery status data, load weight data, meteorological data of the operating area, airspace control data, and geographical environment data, and is output based on the real-time data synchronization module in the virtual digital twin.

[0010] Preferably, the intelligent fault diagnosis unit has a built-in fault feature database. By comparing real-time simulation data with the benchmark data in the fault feature database, and combining deep learning algorithms, it can identify and locate potential faults in the UAV's power system, flight control system, navigation system, and communication system.

[0011] Preferably, the system further includes a closed-loop management module based on the entire life cycle. The closed-loop management module includes a production assembly traceability unit, which records the manufacturer, batch information, assembly process parameters and quality inspection data of each component of the physical UAV, forming a traceable production file and storing it locally. The file is connected to the data of the physical attribute modeling module.

[0012] Preferably, the risk assessment unit performs risk assessment based on the data output by the multi-source data fusion driving module, the task simulation and deduction unit, and the fault intelligent diagnosis unit. It provides threshold reminders based on existing localized risk data of physical drones, such as meteorological data, control data, and status data of the operation area, and the above data is output based on the multi-source data fusion driving module.

[0013] Preferably, the data output by the mission simulation and deduction unit to the risk assessment unit includes flight mission simulation results under different scenarios, and the data output by the fault intelligent diagnosis unit to the risk assessment unit includes potential fault data identified by it. The risk assessment unit performs node synchronization based on the above data and assesses the risks based on flight safety risks, mission execution risks, and operational risks.

[0014] Preferably, the task simulation and extrapolation unit simulates the repetition points of the simulation route based on the historical operational data of the physical UAV. It then fuses the simulation data with the assessment data output by the risk assessment unit for the current area under the real-time state of the physical UAV, and calculates the accuracy of the task simulation data based on the fused data.

[0015] Preferably, the physical UAV performs data fusion based on the accuracy rate output by the mission simulation and deduction unit and the remaining service life of the virtual digital twin components deduced by the predictive maintenance unit, and determines whether the physical UAV can complete the mission based on the fused data, including the deduction of the remaining service life of the physical UAV components.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention utilizes real-time interactive mapping between a physical UAV and a virtual digital twin, combined with a multi-source data fusion-driven module to dynamically simulate the UAV's performance status and operating environment. The core of this invention is a predictive maintenance unit that addresses key issues. Based on potential fault results identified by the intelligent fault diagnosis unit and historical UAV operational data, it employs a time-series prediction algorithm to accurately estimate the remaining lifespan of each component, generating a personalized maintenance plan that includes maintenance time nodes, component lists, and operating procedures. This replaces the traditional rigid periodic maintenance model, avoiding wasted maintenance resources and the risk of sudden failures. Simultaneously, the task simulation and deduction unit's multi-scenario simulation and the risk assessment unit's scheduling optimization scheme collaboratively achieve refined management throughout the entire lifecycle, significantly improving the stability and safety of logistics tasks and meeting the needs of large-scale, routine low-altitude logistics operations.

[0017] 2. This invention also utilizes a physical property modeling module to replicate the structure, performance, and material properties of the UAV based on factory data. This provides high-fidelity basic data such as the inherent local preset parameters of components and the total design life for "data coverage estimation," avoiding estimation errors caused by deviations between the model and the physical object. A real-time data synchronization module ensures seamless integration of physical and virtual data, providing data coverage estimation with... Real-time status data is used to ensure the timeliness of calculations. Data coverage calculations rely on a proprietary formula, combined with... By integrating component data with life loss functions, the remaining lifespan of core components can be accurately calculated. This avoids the problem of general models ignoring individual differences. The results are output to the simulation calculation module, which optimizes the simulation accuracy of fault scenarios, reduces redundant data processing, improves system computing efficiency, and provides high-quality data support for subsequent maintenance and diagnosis.

[0018] 3. This invention also provides the risk assessment unit with key data such as mission completion time and energy consumption through the flight mission results output by the mission simulation and simulation unit for different scenarios. Combined with potential fault data from the intelligent fault diagnosis unit, this helps the unit comprehensively assess flight safety, mission execution, and operational risks, avoiding omissions in risk assessment due to incomplete data. The simulation and simulation, combined with historical operational data, simulates route repetition points, accurately calculating the accuracy of mission simulation data, improving the credibility of simulation results, and laying a reliable foundation for subsequent decision-making. At the same time, the physical UAV, relying on the accuracy of the simulation and simulation, integrates the remaining service life of components to determine mission feasibility, avoiding blindly executing missions and causing safety hazards. The synergy of these three aspects makes control more precise, further ensuring the stability and safety of low-altitude logistics operations. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0020] like Figure 1 As shown, this invention relates to a low-altitude logistics drone full lifecycle management system based on digital twins, including a physical drone and a corresponding virtual digital twin with real-time interactive mapping. The system also includes a multi-source data fusion driving module that collects data from the physical drone and drives the virtual digital twin to dynamically simulate the performance status and operating environment of the physical drone in a virtual state. The virtual digital twin integrates a fault intelligent diagnosis unit for identifying potential faults of the physical drone, a predictive maintenance unit for formulating maintenance strategies, a task simulation and extrapolation unit for simulating flight missions, and a risk assessment unit for evaluating flight mission risks and outputting optimized scheduling strategies. The predictive maintenance unit uses a time-series prediction algorithm to calculate the remaining service life of each component based on potential fault diagnosis results and historical drone operation data, and generates a personalized maintenance plan. The maintenance plan includes maintenance time nodes, a list of maintenance components, and maintenance operation specifications. The task simulation and extrapolation unit supports multi-scenario parameter configuration and can simulate the flight mission execution process under different wind speeds, precipitation, visibility, and airspace congestion conditions, and output the extrapolation results of mission completion time, energy consumption, and path deviation.

[0021] This invention utilizes real-time interactive mapping between a physical UAV and a virtual digital twin, combined with a multi-source data fusion-driven module to dynamically simulate the UAV's performance status and operating environment. The core of the solution lies in a predictive maintenance unit that addresses key issues. Based on potential faults identified by the intelligent fault diagnosis unit and historical UAV operational data, it employs a time-series prediction algorithm to accurately estimate the remaining lifespan of each component, generating a personalized maintenance plan that includes maintenance timelines, component lists, and operating procedures. This replaces the traditional rigid, periodic maintenance model, avoiding wasted maintenance resources and the risk of sudden failures. Simultaneously, the task simulation and deduction unit's multi-scenario simulation and the risk assessment unit's scheduling optimization scheme collaboratively achieve refined management throughout the entire lifecycle, significantly improving the stability and safety of logistics tasks and meeting the needs of large-scale, routine low-altitude logistics operations.

[0022] Specifically, the predictive maintenance unit uses a multi-source data fusion driving module to achieve real-time data transmission between the physical UAV and its corresponding virtual digital twin. Based on this, the predictive maintenance unit performs localized data coverage estimation on the operational data output by the physical UAV. That is, it performs data estimation on the virtual digital twin based on the output of the multi-source data fusion driving module to obtain potential fault data of the physical UAV. This localized data coverage estimation is based on the following formula:

[0023] in for Time of the first The remaining service life of each core component For time series prediction functions, For the first The inherent local preset parameters of each component, For the first Total design life of each component For the first Each component and Momentary fusion data For the first Lifetime loss function of each component This refers to the current state data output from the physical drone to the virtual digital twin based on the multi-source data fusion driving module; Its localized data coverage and calculation formulas can incorporate personalized data such as component-specific local parameters and total design life. By combining real-time fusion data with life loss functions, it avoids the problem of general models ignoring individual differences, making the calculation of the remaining life of core components more accurate. At the same time, relying on real-time data transmission, it dynamically updates the calculation results, improving timeliness and providing reliable data support for personalized maintenance plans, reducing the risk of maintenance inaccuracies.

[0024] Based on the above formula, taking the potential fault detection of a physical drone battery as an example, the following application scenarios are derived: Real-time data (Current battery level 30%), historical data (The factory-set lifespan of the same model battery is 1000 cycles), fusion weighting Real-time coefficient Data integration ; Battery design life Loop, currently 800 loops used, lifetime decay function "Each 10% reduction in battery capacity corresponds to 20 cycles of loss," the calculation is based on.

[0025] The battery cycle life was determined to be 157.6, and the failure threshold was [not specified]. Loop, output The battery was determined to be faulty.

[0026] Furthermore, the virtual digital twin also includes a physical property modeling module, a real-time data synchronization module, and a simulation calculation module. The physical property modeling sub-module locally replicates the structural parameters, performance parameters, and material properties of the physical UAV based on the factory data. The real-time data synchronization module realizes the data synchronization operation between the physical UAV and the virtual digital twin based on the multi-source data fusion driving module, and performs data extraction simulation based on the simulation calculation module, and outputs the current state data to the fault intelligent diagnosis unit. The three modules of the pseudo-digital twin enhance the system's functionality from the ground up. The physical property modeling module replicates the UAV's structure, performance, and material properties to build a high-fidelity virtual model and reduce simulation errors. The real-time data synchronization and simulation calculation module works in conjunction to achieve seamless data integration and filter key data, improving computational efficiency and ensuring that the data output to the fault diagnosis unit is accurate and timely.

[0027] This invention also utilizes a physical property modeling module to replicate the structure, performance, and material properties of the UAV based on factory data. This provides high-fidelity basic data such as the inherent local preset parameters of components and the total design life for "data coverage estimation," avoiding estimation errors caused by deviations between the model and the physical object. A real-time data synchronization module ensures seamless integration of physical and virtual data, providing data coverage estimation with... Real-time status data is used to ensure the timeliness of calculations. Data coverage calculations rely on a proprietary formula, combined with... By integrating component data with life loss functions, the remaining lifespan of core components can be accurately calculated. This avoids the problem of general models ignoring individual differences. The results are output to the simulation calculation module, which optimizes the simulation accuracy of fault scenarios, reduces redundant data processing, improves system computing efficiency, and provides high-quality data support for subsequent maintenance and diagnosis.

[0028] Furthermore, the multi-source data fusion driving module collects data including the flight attitude data of the physical UAV, power system operation data, battery status data, payload weight data, meteorological data of the operating area, airspace control data, and geographical environment data, and outputs it based on the real-time data synchronization module in the virtual digital twin. The multi-source data fusion-driven module collects comprehensive data, covering equipment data such as flight attitude and power system, as well as environmental data such as meteorological, airspace, and geographic information. It can capture information about the UAV's operation and work environment from all angles, avoiding the disconnect between virtual simulation and reality due to data gaps. It provides multi-dimensional data for fault diagnosis, mission simulation, and other tasks, improving the comprehensiveness of management and control.

[0029] Furthermore, the intelligent fault diagnosis unit has a built-in fault feature database. By comparing real-time simulation data with the benchmark data in the fault feature database, and combining deep learning algorithms, it can identify and locate potential faults in the UAV's power system, flight control system, navigation system, and communication system. The intelligent fault diagnosis unit has a built-in fault feature database. Combined with deep learning algorithms, it can accurately identify and locate potential faults in systems such as power and flight control by comparing real-time simulation data with benchmark data. This avoids the problems of missed or false diagnoses in traditional diagnostics, detects potential equipment problems in advance, reduces the probability of sudden failures, and provides accurate fault information for predictive maintenance.

[0030] It is worth noting that the system also includes a closed-loop management module based on the entire life cycle. The closed-loop management module includes a production and assembly traceability unit, which records the manufacturers, batch information, assembly process parameters and quality inspection data of each component of the physical drone, forming a traceable production file and storing it locally. Its data is connected with the physical property modeling module. The production assembly traceability unit of the closed-loop management module records and stores data such as component manufacturers and batches locally, and also communicates with the data of the physical property modeling module. This enables traceability of the entire production lifecycle of the drone, facilitates the tracing of component quality issues, provides production background data for fault analysis and maintenance plan formulation, and improves the control chain.

[0031] It is worth mentioning that the risk assessment unit conducts risk assessment based on the data output by the multi-source data fusion driving module, the task simulation and deduction unit, and the fault intelligent diagnosis unit. It provides threshold alerts based on existing localized risk data of physical drones, such as meteorological data, control data, and status data of the operation area. All of the above data is output based on the multi-source data fusion driving module. The risk assessment unit is based on multi-module data evaluation and relies on localized risk data threshold alerts. It can integrate environmental status data, mission simulation results, and fault data to avoid the one-sidedness of single-data evaluation; threshold alerts can provide timely warnings of dangerous scenarios such as weather and air traffic control, preventing drones from operating in high-risk environments and indirectly improving mission safety.

[0032] It is worth mentioning that the data output from the mission simulation and simulation unit to the risk assessment unit includes flight mission simulation results under different scenarios, and the data output from the fault intelligent diagnosis unit to the risk assessment unit includes the potential fault data it identifies. The risk assessment unit performs node synchronization based on the above data and assesses the risks based on flight safety risks, mission execution risks, and operational risks. The risk assessment unit combines mission simulation results with potential fault data to classify and assess risks according to flight safety, mission execution, and operational risks. This avoids confusion in risk assessment dimensions and enables the targeted identification of different risk points, such as energy consumption risks and equipment risks, providing a clear basis for outputting optimized scheduling strategies and improving the accuracy of risk management.

[0033] It is worth noting that the mission simulation and extrapolation unit simulates the repetition points of the simulation route based on the historical operational data of physical drones. It then integrates the simulation data with the assessment data output by the risk assessment unit for the current area under the real-time state of the physical drones, and calculates the accuracy of the mission simulation data based on the integrated data. The task simulation and extrapolation unit combines historical operational data to simulate recurring points on the route, calculating the accuracy of the task simulation data. This avoids the distortion caused by relying solely on real-time data, making the simulation more closely resemble actual operational patterns, improving the credibility of the simulation results, reducing task planning errors, and providing a reliable reference for subsequent task feasibility assessments.

[0034] It is worth noting that the physical drone is fused with data from the real rate output by the mission simulation and extrapolation unit and the remaining service life of the virtual digital twin components extrapolated by the predictive maintenance unit. The fused data is used to determine whether the physical drone can complete the mission, including extrapolating the remaining service life of the physical drone's components. The feasibility of a mission is determined by combining the realism of the mission simulation with the remaining lifespan of the components. This avoids blindly executing missions; for example, low simulation realism can lead to path deviations, and insufficient component lifespan can cause malfunctions. Combining these two factors can eliminate potential unsafe executions in advance, further reducing flight risks and ensuring the stable completion of logistics missions. Based on a multi-source data fusion risk assessment formula, it encompasses four major stages: data preprocessing, node synchronization, multi-dimensional risk calculation, and comprehensive assessment. This allows for the implementation of control measures for the stable operation and monitoring of physical drones. The formula is as follows:

[0035] when This indicates no risk; This represents a linear quantification of risk; This represents extremely high risk, among which These are the original inference parameters. The parameter is the safety threshold; for example, a path deviation rate ≤ 5% is considered safe. =5, The parameter is the danger threshold; for example, a path deviation rate ≥ 30% is considered dangerous. =30.

[0036] This invention also provides the risk assessment unit with key data such as mission completion time and energy consumption through flight mission results output by the mission simulation and simulation unit for different scenarios. Combined with potential fault data from the intelligent fault diagnosis unit, it helps the unit comprehensively assess flight safety, mission execution, and operational risks, avoiding omissions in risk assessment due to incomplete data. The simulation and simulation, combined with historical operational data, simulates route repetition points, accurately calculating the accuracy of mission simulation data, improving the credibility of simulation results, and laying a reliable foundation for subsequent decision-making. At the same time, the physical UAV, based on the accuracy of the simulation and simulation, integrates the remaining service life of components to determine mission feasibility, avoiding blindly executing missions and causing safety hazards. The synergy of these three aspects makes management and control more precise, further ensuring the stability and safety of low-altitude logistics operations.

[0037] Working Principle: This embodiment provides a digital twin-based full lifecycle management system for low-altitude logistics drones. During operation, a multi-source data fusion module collects equipment data such as the physical drone's flight attitude, power system, and battery status, as well as environmental data such as weather and airspace control in the operating area. Seamless data transmission is achieved through a real-time data synchronization module of the virtual digital twin. Simultaneously, a physical attribute modeling module replicates the structure and performance parameters of the physical drone based on its factory data, constructing a high-fidelity virtual model. A simulation calculation module extracts and processes the synchronized data before outputting it to a fault intelligent diagnosis unit. This unit compares real-time simulation data with a built-in fault feature database, using deep learning to identify potential faults. A predictive maintenance unit, based on fault results and historical operating data, uses localized data coverage calculation formulas, incorporating inherent component parameters and total design life data, to accurately calculate the remaining service life of core components and generate personalized maintenance plans. A task simulation and extrapolation unit, combining historical operating data, simulates route repetition points, outputting multi-scenario task extrapolation results, which, along with fault data, are sent to a risk assessment unit for categorized assessment of flight safety, task execution, and operational risks. Ultimately, the physical drones combine simulation accuracy and component remaining lifespan to determine mission feasibility, while the production and assembly traceability unit provides component production data, forming a closed loop for full lifecycle management.

[0038] The embodiments disclosed in this invention are preferred embodiments, but are not limited thereto. Those skilled in the art can easily understand the spirit of this invention based on the above embodiments and make different extensions and variations, but as long as they do not depart from the spirit of this invention, they are all within the protection scope of this invention.

Claims

1. A low-altitude logistics drone full lifecycle management system based on digital twins, comprising a physical drone and a corresponding virtual digital twin with real-time interactive mapping, characterized in that, The system also includes a multi-source data fusion driving module that collects data from physical UAVs and drives a virtual digital twin to dynamically simulate the performance status and operating environment of the physical UAV in a virtual state. The virtual digital twin integrates a fault intelligent diagnosis unit for identifying potential faults in the physical UAV, a predictive maintenance unit for formulating maintenance strategies, a task simulation unit for simulating flight missions, and a risk assessment unit for evaluating flight mission risks and outputting optimized scheduling strategies. The predictive maintenance unit uses a time-series prediction algorithm to calculate the remaining service life of each component based on potential fault diagnosis results and historical UAV operation data, and generates a personalized maintenance plan. The maintenance plan includes maintenance time nodes, a list of maintenance components, and maintenance operation specifications. The task simulation unit supports multi-scenario parameter configuration and can simulate the flight mission execution process under different wind speeds, precipitation, visibility, and airspace congestion conditions, and output the simulation results of mission completion time, energy consumption, and path deviation.

2. The low-altitude logistics drone full lifecycle management system based on digital twin as described in claim 1, characterized in that, The predictive maintenance unit, based on a multi-source data fusion driving module, enables real-time data transmission between the physical UAV and its corresponding virtual digital twin. It also performs localized data coverage estimation on the operational data output by the physical UAV, specifically by performing data estimation on the virtual digital twin based on the output of the multi-source data fusion driving module, to obtain potential fault data of the physical UAV. This localized data coverage estimation is based on the following formula: ； in for Time of the first The remaining service life of each core component For time series prediction functions, For the first The inherent local preset parameters of each component, For the first Total design life of each component For the first Each component and Momentary fusion data For the first Lifetime loss function of each component This refers to the current state data output from the physical drone to the virtual digital twin based on the multi-source data fusion driving module.

3. The low-altitude logistics drone full lifecycle management system based on digital twin as described in claim 1, characterized in that, The virtual digital twin also includes a physical property modeling module, a real-time data synchronization module, and a simulation calculation module. The physical property modeling submodule locally replicates the structural parameters, performance parameters, and material properties of the physical UAV based on its factory data. The real-time data synchronization module enables the physical UAV to synchronize data with the virtual digital twin based on the multi-source data fusion driving module, and performs data extraction simulation based on the simulation calculation module, outputting the current state data to the fault intelligent diagnosis unit.

4. A low-altitude logistics drone full lifecycle management system based on digital twins as described in claim 3, characterized in that, The data collected by the multi-source data fusion driving module includes the flight attitude data of the physical UAV, power system operation data, battery status data, load weight data, meteorological data of the operating area, airspace control data, and geographical environment data, and is output based on the real-time data synchronization module in the virtual digital twin.

5. A low-altitude logistics drone full lifecycle management system based on digital twins as described in claim 4, characterized in that, The intelligent fault diagnosis unit has a built-in fault feature database. By comparing real-time simulation data with the benchmark data in the fault feature database, and combining deep learning algorithms, it can identify and locate potential faults in the UAV's power system, flight control system, navigation system, and communication system.

6. A low-altitude logistics drone full lifecycle management system based on digital twins as described in claim 3, characterized in that, The system also includes a closed-loop management module based on the entire life cycle. The closed-loop management module includes a production assembly traceability unit, which records the manufacturer, batch information, assembly process parameters and quality inspection data of each component of the physical UAV, forming a traceable production file and storing it locally. The file is connected to the data of the physical attribute modeling module.

7. A low-altitude logistics drone full lifecycle management system based on digital twins as described in claim 6, characterized in that, The risk assessment unit performs risk assessment based on the data output by the multi-source data fusion driving module, the task simulation and deduction unit, and the fault intelligent diagnosis unit. It provides threshold alerts based on existing localized risk data of physical drones, such as meteorological data, control data, and status data of the operation area. All of the above data is output by the multi-source data fusion driving module.

8. A low-altitude logistics drone full lifecycle management system based on digital twins as described in claim 7, characterized in that, The data output by the mission simulation and deduction unit to the risk assessment unit includes flight mission simulation results under different scenarios. The data output by the fault intelligent diagnosis unit to the risk assessment unit includes potential fault data it identifies. The risk assessment unit performs node synchronization based on the above data and assesses the risks based on flight safety risks, mission execution risks, and operational risks.

9. A low-altitude logistics drone full lifecycle management system based on digital twins as described in claim 8, characterized in that, The task simulation and deduction unit simulates the repetition points of the simulation route based on the historical operational data of physical UAVs. It then merges the simulation data with the assessment data output by the risk assessment unit for the current area under the real-time state of the physical UAVs, and calculates the accuracy of the task simulation data based on the merged data.

10. A low-altitude logistics drone full lifecycle management system based on digital twins as described in claim 9, characterized in that, The physical UAV is fused with data from the accuracy rate output by the mission simulation and deduction unit and the remaining service life of the virtual digital twin components deduced by the predictive maintenance unit. The fused data is used to determine whether the physical UAV can complete the mission, including the deduction of the remaining service life of the physical UAV components.

Images