Unmanned aerial vehicle distribution goods loading and unloading traceability method and system based on double-sensor cooperation

Abstract

The present application relates to the technical field of goods traceability, and proposes a UAV distribution goods loading and unloading traceability method and system based on double-sensor cooperation, which comprises the following steps: double-sensor cooperation pre-binding of the outbound goods label information and the UAV identification code to obtain an initial association record; infrared guidance homing of the UAV to obtain a loading verification data set of the UAV; time-weight mutual information integral judgment of the loading verification data set and the initial association record to obtain a binary loading confirmation instruction of the UAV; loading state judgment of the UAV to obtain a unloading operation process data set of the UAV; Hash closed-loop verification of the loading verification data set and the unloading operation process data set to obtain a unloading operation verification result of the UAV; Hash chain binding association of the unloading operation verification result and the binary loading confirmation instruction to obtain a complete closed-loop traceability file of the UAV; and the present application can improve the UAV distribution goods loading and unloading traceability efficiency.

Description

Technical Field

[0001] This invention relates to the field of cargo traceability technology, and in particular to a method and system for tracing the loading and unloading of cargo delivered by drones based on dual-sensor collaboration. Background Technology

[0002] The current drone delivery cargo handling traceability system lacks a dual-sensor collaborative working mechanism. Outbound cargo label information and drone identification codes cannot be accurately pre-bound, and the consistency verification process for cargo tracking numbers, destination codes, and drone unique identifiers is missing. This prevents the generation of authentic and valid initial association records and binding relationship certificates, making it difficult to guarantee the accuracy of cargo-drone matching. Furthermore, the drone loading phase lacks infrared guidance and homing functionality, preventing the ground-based drone housing from emitting modulated infrared light signals, hindering spot calculation and attitude deviation correction. This results in inaccurate drone landing path adjustments, unstable hovering and locking status acquisition, and the cargo hold's built-in RFID reader failing to scan the cargo manifest in a timely manner. Consequently, loading verification data cannot be fully collected, and the generation of the loading verification dataset lacks comprehensive data support.

[0003] The existing drone loading and unloading traceability process lacks the ability to make time-weighted mutual information integration judgments. It cannot perform time-series matching and judgment between the loading verification dataset and the initial associated records, and cannot generate accurate binary loading confirmation instructions. It cannot quickly distinguish between qualified and abnormal loading states, and abnormal loading states cannot trigger alarms and manual review processes in a timely manner. The unloading operation lacks a hash-based closed-loop verification mechanism, making it impossible to perform time-series linkage deviation analysis and hash operation comparison of loading and unloading data. The data verification results are unreliable. Furthermore, the lack of hash chain binding and blockchain notarization means that the unloading operation verification results and binary loading confirmation instructions cannot form a secure association link, and a complete closed-loop traceability file cannot be generated. The efficiency, data security, and reliability of loading and unloading traceability fail to meet the practical application requirements of intelligent drone delivery. Summary of the Invention

[0004] The present invention provides a method for tracing the loading and unloading of goods delivered by drones based on dual-sensor collaboration, comprising: S1. Perform dual-sensor collaborative pre-binding on the outbound goods label information and the drone identification code to obtain the initial association record of the drone; S2. Based on the initial association record, perform infrared guidance and repositioning of the UAV to obtain the loading verification dataset of the UAV; S3. Perform time-weighted mutual information integration on the loading verification dataset and the initial association record to obtain the binary loading confirmation command of the UAV; S4. Based on the binary loading confirmation command, determine the loading status of the UAV to obtain the unloading operation process dataset of the UAV; S5. Perform hash closed-loop verification on the loading verification dataset and the unloading operation process dataset to obtain the unloading operation verification result of the UAV; S6. The verification result of the unloading operation is linked with the binary loading confirmation instruction by hash chain to obtain the complete closed-loop traceability file of the UAV.

[0005] Preferably, the outbound cargo label information and the drone identification code are pre-bound using dual-sensor collaborative methods to obtain an initial association record between the drone and the cargo, including: Read the label information of the outbound goods to obtain the shipment order number and destination code of the outbound goods; The consistency of the cargo tracking number, the destination code, and the unique identifier of the drone is verified to obtain the binding relationship certificate between the drone and the cargo; The binding relationship certificate is stored to obtain the initial association record between the drone and the cargo.

[0006] Preferably, the step of performing infrared guidance and repositioning of the UAV to obtain the UAV's loading verification dataset includes: Based on the initial association record, the UAV's ground nest infrared beacon is activated, and the modulated infrared light signal of the infrared beacon is transmitted; The modulated infrared light signal is subjected to spot resolution to obtain the pose deviation data of the UAV; Based on the pose deviation data, a flight path correction command is injected into the flight control unit of the UAV to obtain the corrected landing path of the UAV. Based on the corrected landing path, the hovering lock status of the UAV after landing is detected to obtain a hovering lock completion signal; Based on the hovering lock completion signal, scan the radio frequency identification reader built into the drone's cargo compartment to obtain the cargo list data inside the compartment; The cargo manifest data inside the cabin is verified against the cargo order number in the initial association record to obtain the loading verification dataset of the UAV.

[0007] Preferably, the step of performing a time-weighted mutual information integral decision on the loading verification dataset and the initial associated record to obtain the binary loading confirmation instruction for the UAV includes: During the infrared-guided return process of the UAV, temporal features are extracted from the loading verification dataset to obtain an instantaneous loading verification feature vector; The instantaneous loading verification feature vector includes the UAV's horizontal offset, vertical height difference, and RFID signal strength at the corresponding moment; The cargo order number and the unique identifier of the UAV in the initial association record are feature-encoded to obtain the feature vector of the initial association record; The loading confirmation index of the UAV is obtained by performing time-weighted mutual information integration on the instantaneous loading verification feature vector and the initial associated record feature vector; The formula for calculating the loading confirmation index is as follows: ; in, The loading confirmation index, It is a unit step function. For the infrared-guided return process The instantaneous loading verification feature vector collected. This represents the total duration of infrared-guided repositioning. The initial associated record feature vector, This is the conditional mutual information estimate for the UAV. For the drone at time Historical data sets from a previous period of time, To load and verify the feature vector instantaneously, This is an estimate of the conditional entropy. For An exponential function with the natural constant as its base, where The preset time decay coefficient, and , The entropy of the initial recorded feature vector is... The preset confirmation threshold, The default value is a small positive number to prevent smooth terms with a denominator of zero; When the loading confirmation index is 1, a first binary loading confirmation command indicating that the loading is qualified is generated; when the loading confirmation index is 0, a second binary loading confirmation command indicating that the loading is abnormal is generated.

[0008] Preferably, after obtaining the loading confirmation index of the UAV, the method further includes: If a first-valued loading confirmation command indicating that the loading is qualified is generated, the loading of the UAV is triggered to obtain the cargo in the UAV's cargo hold; A second full scan of the cargo in the cargo hold is performed using radio frequency identification to obtain a second scan cargo list; The consistency of the secondary scanned goods list with the goods order number in the initial associated record is compared to obtain a consistency result; Based on the comparison results, the electronic lock of the UAV cargo compartment is locked to obtain a locking completion signal; Based on the locking completion signal, the onboard sensors of the UAV are used to collect status data to obtain real-time status data of the UAV, including cargo hold locking status and takeoff ready signal. Upon receiving the real-time status data, the system uploads the storage information for the drone's loading completion status record and receives a data reception confirmation signal. Based on the data reception confirmation signal, the data records of the drone from loading completion to infrared deactivation are recorded, and the unloading operation process dataset of the drone is obtained by summarizing the data.

[0009] Preferably, after obtaining the loading confirmation index of the UAV, the method further includes: If the binary loading confirmation instruction is the second binary loading confirmation instruction, an alarm message indicating a loading abnormality is generated; Based on the alarm information, the loading action of the drone is prohibited, and a manual review work order is obtained.

[0010] Preferably, the step of performing a hash-based closed-loop verification of the time-series linkage deviation between the loading verification dataset and the unloading operation process dataset to obtain the unloading operation verification result of the UAV includes: The first key indicator is extracted from the loading verification dataset to obtain the first received signal strength indication time series; The second key indicator is extracted from the unloading operation process dataset to obtain the second received signal strength indication time series; A timing deviation analysis is performed on the first received signal strength indication time series and the second received signal strength indication time series to obtain a timing linkage deviation sequence. A hash operation is performed on the time-series linkage deviation sequence to obtain the unloading operation hash digest; The unloading operation hash digest is compared with the preset loading verification hash benchmark value to obtain the unloading operation verification result of the UAV.

[0011] Preferably, after the preset loading verification hash base value, the method further includes: The load verification hash base value is combined with the load verification dataset to form a hash base data packet; The hash base data packet is digitally signed to obtain the signed loading verification hash base value; The signed loading verification hash base value is stored on the blockchain to obtain the stored transaction identifier of the drone.

[0012] Preferably, the verification result of the unloading operation is hash-chained with the binary loading confirmation instruction to obtain a complete closed-loop traceability file for the drone, including: The unloading operation verification result and the binary loading confirmation instruction are feature-encoded to obtain traceability data to be bound. Perform an initial hash operation on the traceability data to be bound to obtain the traceability hash value; The traceability hash value and the unloading operation process dataset are packaged together to obtain a traceability data block; The traceability data blocks are chained together to obtain a chained traceability data structure; The chain-based traceability data structure, the loading verification dataset, the unloading operation process dataset, and the drone's evidence transaction identifier are associated and stored to obtain a complete closed-loop traceability file for the drone.

[0013] This invention also provides a drone delivery cargo loading and unloading traceability system based on dual-sensor collaboration, the system comprising: The dual-sensor pre-binding module is used to perform dual-sensor collaborative pre-binding of outbound cargo label information and UAV identification code to obtain the initial association record of the UAV. The infrared guidance and repositioning module is used to guide the UAV to its repositioning based on the initial association record, and obtain the loading verification dataset of the UAV. The time-weighted mutual information decision module is used to perform time-weighted mutual information integration decision on the loading verification dataset and the initial association record to obtain the binary loading confirmation instruction of the UAV; The loading status determination module is used to determine the loading status of the UAV based on the binary loading confirmation command, and obtain the unloading operation process dataset of the UAV. The hash loop verification module is used to perform hash loop verification on the loading verification dataset and the unloading operation process dataset to obtain the unloading operation verification result of the UAV. The chain-based traceability file module is used to perform hash chain binding and association between the unloading operation verification result and the binary loading confirmation instruction to obtain the complete closed-loop traceability file of the UAV.

[0014] As can be seen from the above technical solution, the UAV delivery cargo loading and unloading traceability method and system based on dual-sensor collaboration provided by the present invention has the following beneficial effects: 1. This invention relies on a dual-sensor collaborative mechanism to accurately pre-bind outbound cargo label information with UAV identification codes, fully verify the consistency of cargo order number, destination code, and UAV unique identifier, and generate standardized initial association records and binding relationship certificates for matching cargo and UAVs. By transmitting modulated infrared light signals through ground-based UAV nest infrared beacons, UAV attitude deviation data is obtained through spot calculation and the landing path is corrected to achieve precise hovering and locking of the UAV. Simultaneously, a full scan and data verification of cargo in the cargo hold is completed through an RFID reader, comprehensively collecting and forming a loading verification dataset, which greatly improves the standardization of the UAV loading process and ensures the integrity of loading data collection and the accuracy of loading status determination.

[0015] 2. This invention completes the temporal matching and judgment of the loading verification dataset and the initial associated record through time-weighted mutual information integration, accurately generates binary loading confirmation instructions, realizes rapid identification and differentiation of loading status, and completes the temporal linkage deviation analysis and hash operation verification of loading and unloading data based on hash closed-loop verification, ensuring that the unloading operation verification results are true and reliable. Through hash chain binding and blockchain notarization, a full-process data security association link is constructed, and traceability data, verification data, and operation data are uniformly associated and stored to form a complete closed loading and unloading traceability file, which significantly improves the overall execution efficiency of drone delivery cargo loading and unloading traceability, while ensuring the security, immutability and full traceability of traceability data. Attached Figure Description

[0016] Other objects and results of the invention will become more apparent and readily understood by referring to the following description taken in conjunction with the accompanying drawings, and with a more complete understanding of the invention. In the drawings: Figure 1 This is a flowchart illustrating the drone delivery cargo loading and unloading traceability method based on dual-sensor collaboration, according to an embodiment of the present invention. Figure 2 This is a functional block diagram of the drone delivery cargo loading and unloading traceability system based on dual-sensor collaboration, according to an embodiment of the present invention. Detailed Implementation To address the aforementioned problems, this invention provides a method and system for tracing the loading and unloading of goods delivered by drones based on dual-sensor collaboration. The specific embodiments of this invention will be described in detail below with reference to the accompanying drawings.

[0017] To illustrate the method and system for tracing the loading and unloading of goods via drones based on dual-sensor collaboration provided by this invention, Figure 1 An exemplary illustration is provided for the drone delivery cargo loading and unloading traceability method based on dual-sensor collaboration in this invention. Figure 2 An exemplary illustration is provided for the drone delivery cargo loading and unloading traceability system based on dual-sensor collaboration in an embodiment of the present invention.

[0018] The following description of exemplary embodiments is merely illustrative and is in no way intended to limit the invention or its application or use. Techniques and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques and equipment should be considered part of the specification.

[0019] Reference Figure 1 The diagram shown is a flowchart illustrating a drone-based cargo loading and unloading traceability method based on dual-sensor collaboration, according to an embodiment of the present invention. In this embodiment, the drone-based cargo loading and unloading traceability method based on dual-sensor collaboration includes: S1. Perform dual-sensor collaborative pre-binding on the outbound goods label information and the drone identification code to obtain the initial association record of the drone; In this embodiment of the invention, the outbound cargo label information and the drone identification code are pre-bound using dual-sensor collaborative methods to obtain an initial association record between the drone and the cargo, including: Read the label information of the outbound goods to obtain the shipment order number and destination code of the outbound goods; The consistency of the cargo tracking number, the destination code, and the unique identifier of the drone is verified to obtain the binding relationship certificate between the drone and the cargo; The binding relationship certificate is stored to obtain the initial association record between the drone and the cargo.

[0020] The electronic tags affixed to outbound goods are read directly by radio frequency identification (RFID) sensors. The pre-stored tracking number and destination code within the electronic tag are completely extracted. The extracted tracking number is a fixed-length character combination, and the extracted destination code is a fixed code that corresponds one-to-one with the delivery area. There are no data omissions or errors in the extraction process.

[0021] The extracted cargo tracking number and destination code are compared character by character with the drone's factory-preset unique identifier. If the cargo tracking number, destination code, and drone's unique identifier are completely identical, a binding relationship certificate containing the cargo tracking number, destination code, and drone's unique identifier is generated. If any character is inconsistent, the binding relationship certificate is not generated.

[0022] The generated binding relationship certificate is written into a dedicated data storage module, and fixed storage is completed by using the combination of the cargo order number and the drone's unique identifier as the storage index. After storage is completed, an initial association record between the drone and the cargo is formed.

[0023] S2. Based on the initial association record, perform infrared guidance and repositioning of the UAV to obtain the loading verification dataset of the UAV; In this embodiment of the invention, the step of performing infrared guidance and repositioning on the UAV to obtain the UAV's loading verification dataset includes: Based on the initial association record, the UAV's ground nest infrared beacon is activated, and the modulated infrared light signal of the infrared beacon is transmitted; The modulated infrared light signal is subjected to spot resolution to obtain the pose deviation data of the UAV; Based on the pose deviation data, a flight path correction command is injected into the flight control unit of the UAV to obtain the corrected landing path of the UAV. Based on the corrected landing path, the hovering lock status of the UAV after landing is detected to obtain a hovering lock completion signal; Based on the hovering lock completion signal, scan the radio frequency identification reader built into the drone's cargo compartment to obtain the cargo list data inside the compartment; The cargo manifest data inside the cabin is verified against the cargo order number in the initial association record to obtain the loading verification dataset of the UAV.

[0024] The call to the stored initial association record triggers the power supply circuit of the ground nest, enabling the infrared beacon of the UAV's ground nest to enter the working state. The infrared beacon emits modulated infrared light signals covering the positioning area according to the preset signal modulation rules, and the signal spot maintains a fixed shape and propagation range.

[0025] The onboard infrared signal receiving component of the UAV collects the modulated infrared light signal emitted by the ground nest in real time. The collected light spot coordinates are matched point-to-point with the onboard preset return reference light spot coordinates to determine the UAV's horizontal offset value, vertical height value, and body orientation deflection value. The combination of the above three types of values ​​forms the UAV's attitude deviation data.

[0026] The attitude deviation data is transmitted to the command access port of the UAV flight control unit. The flight control unit adjusts the UAV's flight attitude and descent trajectory based on the attitude deviation data, removes path content that deviates from the return reference in the original landing route, and generates a corrected landing route that perfectly matches the return requirements.

[0027] After the drone flies to the designated position on the ground-based drone nesting platform according to the corrected landing path, its position and attitude remain fixed. The position sensing module of the landing platform detects the positioning contact point on the drone's body, confirms that the drone is in hovering and locking state, and then generates a hovering and locking completion signal.

[0028] After receiving the hovering lock completion signal, the radio frequency identification (RFID) reader installed inside the drone's cargo compartment is activated. The reader sends identification instructions to the electronic tags of all goods in the cargo compartment, reads the corresponding cargo order numbers of all goods, and organizes them in sequence to form a complete cargo list data in the cargo compartment.

[0029] The cargo manifest number in the cargo manifest data is compared character by character with the cargo manifest number stored in the initial associated record. The comparison results, cargo matching status and complete scanning and identification information are recorded. All the comparison and recording information are integrated to form the UAV loading verification dataset.

[0030] S3. Perform time-weighted mutual information integration on the loading verification dataset and the initial association record to obtain the binary loading confirmation command of the UAV; In this embodiment of the invention, the step of performing a time-weighted mutual information integration decision on the loading verification dataset and the initial associated record to obtain the binary loading confirmation instruction for the UAV includes: During the infrared-guided return process of the UAV, temporal features are extracted from the loading verification dataset to obtain an instantaneous loading verification feature vector; The instantaneous loading verification feature vector includes the UAV's horizontal offset, vertical height difference, and RFID signal strength at the corresponding moment; The cargo order number and the unique identifier of the UAV in the initial association record are feature-encoded to obtain the feature vector of the initial association record; The loading confirmation index of the UAV is obtained by performing time-weighted mutual information integration on the instantaneous loading verification feature vector and the initial associated record feature vector; The formula for calculating the loading confirmation index is as follows: ; in, The loading confirmation index, For unit step function, For the infrared-guided return process The instantaneous loading verification feature vector collected. This represents the total duration of infrared-guided repositioning. The initial associated record feature vector, This is the conditional mutual information estimate for the UAV. For the drone at time Historical data sets from a previous period of time, To load and verify the feature vector instantaneously, This is an estimate of the conditional entropy. For An exponential function with the natural constant as its base, where The preset time decay coefficient, and , The entropy of the initial recorded feature vector is... The preset confirmation threshold, The default value is a small positive number to prevent smooth terms with a denominator of zero; When the loading confirmation index is 1, a first binary loading confirmation command indicating that the loading is qualified is generated; when the loading confirmation index is 0, a second binary loading confirmation command indicating that the loading is abnormal is generated.

[0031] After obtaining the loading confirmation index of the drone, the following is also included: If a first-valued loading confirmation command indicating that the loading is qualified is generated, the loading of the UAV is triggered to obtain the cargo in the UAV's cargo hold; A second full scan of the cargo in the cargo hold is performed using radio frequency identification to obtain a second scan cargo list; The consistency of the secondary scanned goods list with the goods order number in the initial associated record is compared to obtain a consistency result; Based on the comparison results, the electronic lock of the UAV cargo compartment is locked to obtain a locking completion signal; Based on the locking completion signal, the onboard sensors of the UAV are used to collect status data to obtain real-time status data of the UAV, including cargo hold locking status and takeoff ready signal. Upon receiving the real-time status data, the system uploads the storage information for the drone's loading completion status record and receives a data reception confirmation signal. Based on the data reception confirmation signal, the data records of the drone from loading completion to infrared deactivation are recorded, and the unloading operation process dataset of the drone is obtained by summarizing the data.

[0032] After obtaining the loading confirmation index of the drone, the following is also included: If the binary loading confirmation instruction is the second binary loading confirmation instruction, an alarm message indicating a loading abnormality is generated; Based on the alarm information, the loading action of the drone is prohibited, and a manual review work order is obtained.

[0033] Throughout the entire process of the UAV performing infrared-guided repositioning, real-time information from the loading verification dataset is collected at fixed time intervals of 100 milliseconds. The horizontal offset of the UAV relative to the repositioning reference point, the vertical height difference relative to the repositioning reference plane, and the RFID signal strength fed back by the RFID reader are extracted at that moment. The three data extracted at the same moment are combined to form an instantaneous loading verification feature vector.

[0034] According to the preset character encoding rules, the cargo order number and UAV unique identifier stored in the initial association record are subjected to a unified format feature conversion process. The converted cargo order number feature data and UAV unique identifier feature data are combined to form the feature vector of the initial association record.

[0035] Using the entire duration of infrared-guided repositioning as the statistical range, all historical data collected before each moment in the repositioning process are retrieved as a reference. First, the correlation between the instantaneous loading verification feature vector and the initial association record feature vector is determined. Then, the dispersion of the instantaneous loading verification feature vector under historical data and the dispersion of the initial association record feature vector itself are determined. Combined with the rule of gradually reducing the weight of early data over time, the correlation degree throughout the process is weighted and accumulated statistically. At the same time, a small numerical correction is added to avoid abnormal statistical results, and finally the loading confirmation index of the UAV is obtained.

[0036] The loading confirmation index is compared with the preset confirmation threshold. When the loading confirmation index reaches the qualified judgment standard set by the confirmation threshold, the result of the loading confirmation index is determined to be 1. Based on the result, a first binary loading confirmation command indicating that the loading is qualified is generated.

[0037] The loading confirmation index is compared with the preset confirmation threshold. If the loading confirmation index does not meet the qualified judgment standard set by the confirmation threshold, the result of the loading confirmation index is determined to be 0. Based on this result, a second binary loading confirmation command indicating loading abnormality is generated.

[0038] The loading confirmation index is based on instantaneous loading verification feature vectors collected at fixed time intervals throughout the entire process of UAV infrared-guided return. This vector is formed by combining three types of measured data: UAV real-time horizontal offset, vertical height difference, and RFID signal strength.

[0039] The initial associated record feature vector comes from the combined data formed by encoding the cargo order number and the UAV's unique identifier in a unified format within the initial associated record.

[0040] The total infrared-guided repositioning time is derived from the statistical results of the actual runtime of the UAV from the start of the ground-based infrared beacon to the completion of hovering and locking.

[0041] The historical dataset consists of all the homing pose data and RFID data collected by the UAV that were stored before the data acquisition at a single moment.

[0042] The time decay coefficient comes from a fixed value preset by the system and is used to gradually reduce the weight of early collected data as time progresses throughout the statistical process.

[0043] The confirmation threshold is derived from the pre-set loading qualification standard value of the system and is used to define the boundary between qualified and abnormal loading confirmation index.

[0044] The smoothing term comes from a fixed, small value preset by the system, used to avoid calculation anomalies where the denominator is zero during the entire statistical calculation process.

[0045] This statistical calculation method is used to perform a full-process weighted correlation analysis between the instantaneous loading verification feature vector and the initial associated record feature vector, and to quantitatively evaluate the matching compliance of the drone loading process.

[0046] This statistical calculation method combines time decay rules to perform weighted accumulation processing on the collected data throughout the process, thereby improving the accuracy and stability of loading status timing determination.

[0047] This statistical calculation method transforms the quantitative statistical results into binary judgment results through a step-by-step judgment rule, directly determining whether the drone's loading status is qualified or abnormal.

[0048] This statistical calculation method, through a comprehensive analysis of correlation and data dispersion, ensures that the loading confirmation index can accurately reflect the binding and matching status between cargo and drones.

[0049] The loading confirmation index output by this statistical calculation method serves as the core basis for generating binary loading confirmation commands, enabling automated and accurate judgment of the UAV loading status.

[0050] After receiving the first and second value loading confirmation command indicating that the loading is qualified, the system immediately triggers the automatic loading execution mechanism of the drone to start operation. The automatic loading execution mechanism accurately pushes the goods that perfectly match the initial association record into the drone's cargo compartment. After the loading action is completed, the cargo in the drone's cargo compartment is obtained.

[0051] The built-in RFID reader in the drone's cargo hold is activated to perform a full scan of all goods inside the cargo hold without any blind spots. The RFID reader reads the cargo tracking number information from the electronic tag of each item, and then organizes all the cargo tracking numbers into a fixed sequence to obtain a secondary scan cargo list.

[0052] Each group of order numbers in the secondary scanned goods list is compared character by character with the order numbers stored in the initial associated record. When all the characters of the order numbers match completely, a comparison result is generated between the secondary scanned goods list and the order numbers in the initial associated record.

[0053] Based on the matching results, a locking drive command is sent to the electronically controlled latch of the drone cargo compartment. The actuator motor of the electronically controlled latch drives the latch tongue to move and fully engage in the latch positioning groove. After the cargo compartment locking detection module recognizes that the latch tongue is fully engaged, it generates and outputs a locking completion signal.

[0054] Upon receiving the locking completion signal, the system collects the output information of the UAV's onboard sensors in real time, reads the closed state fed back by the cargo hold locking sensor and the standby state fed back by the takeoff preparation sensor, and integrates the two types of status information to obtain real-time UAV status data that includes the cargo hold locking status and takeoff ready signal.

[0055] After receiving the real-time status data of the drone, the system uploads the drone's completed loading status record to the warehouse management system through a dedicated communication channel. After the warehouse management system completes the data verification and archiving operations, it returns a data reception confirmation signal to the control terminal.

[0056] After receiving the data reception confirmation signal, all action data, status data, and timing data of the UAV from the loading completion node to the infrared guidance signal release node are recorded. After all process data of this stage are completely collected in time sequence, the UAV unloading operation process dataset is obtained.

[0057] After the system recognizes that the current binary loading confirmation instruction is the second binary loading confirmation instruction, it extracts the abnormal matching information from the loading verification dataset and the initial associated record, and combines the abnormal status, the corresponding UAV identification code, the cargo order number and the occurrence time information to generate alarm information representing the loading abnormality.

[0058] After receiving an alarm message indicating an abnormal loading, the system disconnects the electrical connection of the drone loading drive circuit, blocks the transmission channel of the loading trigger signal to the automatic loading actuator, prohibits the execution of all loading actions, and integrates the abnormality details, drone information, and cargo information to generate a manual review work order containing complete verification elements.

[0059] S4. Based on the binary loading confirmation command, determine the loading status of the UAV to obtain the unloading operation process dataset of the UAV; The system receives and parses the type identifier of the binary loading confirmation command, and uses the type identifier as the core basis for determining the loading status of the UAV, thus completing the preliminary work for determining the loading status.

[0060] When the binary loading confirmation command is the first binary loading confirmation command, the UAV is determined to be in a qualified loading state. The following operations are executed in sequence: material loading trigger, second full scan of radio frequency identification, cargo order number consistency comparison, cargo compartment electronic lock locking, airborne sensor status acquisition, and warehouse information uploading. The execution sequence, execution status, and equipment feedback results of each step are recorded throughout the process.

[0061] When the binary loading confirmation command is the second binary loading confirmation command, the drone is determined to be in an abnormal loading state. The abnormal loading alarm information is generated, the loading action is prohibited, and the manual review work order is generated. The entire process records the basis for the abnormal judgment, the content of the alarm information, and the complete information of the generated review work order.

[0062] All operational data, timing data, status data, and feedback data generated under both qualified and abnormal loading conditions are uniformly collected and standardized to form a dataset of the drone unloading operation process.

[0063] S5. Perform hash closed-loop verification on the loading verification dataset and the unloading operation process dataset to obtain the unloading operation verification result of the UAV; In this embodiment of the invention, the step of performing hash-based closed-loop verification on the loading verification dataset and the unloading operation process dataset to obtain the unloading operation verification result of the UAV includes: The first key indicator is extracted from the loading verification dataset to obtain the first received signal strength indication time series; The second key indicator is extracted from the unloading operation process dataset to obtain the second received signal strength indication time series; A timing deviation analysis is performed on the first received signal strength indication time series and the second received signal strength indication time series to obtain a timing linkage deviation sequence. A hash operation is performed on the time-series linkage deviation sequence to obtain the unloading operation hash digest; The unloading operation hash digest is compared with the preset loading verification hash benchmark value to obtain the unloading operation verification result of the UAV.

[0064] After the preset load verification hash baseline value, it also includes: The load verification hash base value is combined with the load verification dataset to form a hash base data packet; The hash base data packet is digitally signed to obtain the signed loading verification hash base value; The signed loading verification hash base value is stored on the blockchain to obtain the storage transaction identifier of the drone.

[0065] Extract the received signal strength values ​​corresponding to the RFID from the loading verification dataset according to a fixed acquisition time sequence, and arrange the values ​​at each time point in chronological order to form the first received signal strength indication time series.

[0066] Extract the corresponding RFID received signal strength values ​​from the unloading operation data set according to the same acquisition time sequence as the loading stage, and arrange the values ​​at each time point in chronological order to form a second received signal strength indication time series.

[0067] The values ​​of the same time node in the first received signal strength indication time series and the second received signal strength indication time series are compared one by one. The numerical difference of each node is calculated and recorded. The numerical differences of all nodes are combined in time series to form a time series linkage deviation sequence.

[0068] All data from the time-series linkage deviation sequence are integrated and feature-compressed according to preset hash processing rules to generate a unique and tamper-proof hash digest of the unloading operation.

[0069] The unloading operation hash digest is matched with the loading verification hash base value that is preset and stored in the system. If the match is completely consistent, the unloading operation is deemed to have passed the verification. If there is a difference in the match, the unloading operation is deemed to have failed the verification. This determination result is the unloading operation verification result of the UAV.

[0070] According to the preset fixed data encapsulation structure, the loading and verification hash base value is used as the core data field, and the entire contents of the loading and verification dataset are used as the associated data field. The data splicing and encapsulation process is completed in sequence to form a complete hash base data packet.

[0071] The hash base data packet is signed using a pre-configured asymmetric encryption private key. The generated digital signature is then bound and merged with the loading verification hash base value to obtain the signed loading verification hash base value.

[0072] The signed loading verification hash base value is sent to the designated consortium blockchain node. After the blockchain node completes the data verification and block packaging, it generates a unique transaction number corresponding to this evidence storage operation. This number is the evidence storage transaction identifier for the drone.

[0073] S6. The verification result of the unloading operation is linked with the binary loading confirmation instruction by hash chain to obtain the complete closed-loop traceability file of the UAV.

[0074] In this embodiment of the invention, the hash chain binding association between the unloading operation verification result and the binary loading confirmation instruction yields a complete closed-loop traceability file for the drone, including: The unloading operation verification result and the binary loading confirmation instruction are feature-encoded to obtain traceability data to be bound. Perform an initial hash operation on the traceability data to be bound to obtain the traceability hash value; The traceability hash value and the unloading operation process dataset are packaged together to obtain a traceability data block; The traceability data blocks are chained together to obtain a chained traceability data structure; The chain-based traceability data structure, the loading verification dataset, the unloading operation process dataset, and the drone's evidence transaction identifier are associated and stored to obtain a complete closed-loop traceability file for the drone.

[0075] According to the preset fixed character encoding rules, the pass or fail status of the unloading operation verification result and the qualified or abnormal identifier of the binary loading confirmation instruction are converted into standardized coded data respectively. The two standardized coded data are combined and spliced ​​to obtain the traceability data to be bound.

[0076] According to the preset hash processing specifications, irreversible feature extraction and data compression are performed on all the content of the traceability data to be bound, generating a unique traceability hash value that corresponds to the traceability data to be bound and cannot be tampered with.

[0077] Using the traceability hash value as the core identifier field of the block, and taking all the time-series data, operation data, and status data of the unloading operation process dataset as the main content of the block, the traceability data block is obtained by integrating and encapsulating the data according to a fixed data encapsulation structure.

[0078] According to the preset chain structure construction rules, a one-way unique connection is established between the traceability data block and the related data block of the preceding link, so that each data block forms an immutable connection relationship according to the loading and unloading operation sequence, resulting in a chain traceability data structure.

[0079] Using a chain-based traceability data structure as the core framework, fixed indexes are established with the loading verification dataset, the unloading operation process dataset, and the drone's evidence transaction identifier. All related data are archived and stored in a unified format to obtain a complete closed-loop traceability file for the drone.

[0080] like Figure 2 The diagram shown is a functional block diagram of a drone delivery cargo loading and unloading traceability system 100 based on dual-sensor collaboration provided in an embodiment of the present invention. It includes a dual-sensor pre-binding module 101, an infrared guidance and homing module 102, a time-weighted mutual information judgment module 103, a loading status judgment module 104, a hash closed-loop verification module 105, and a chain-type traceability file module 106.

[0081] In this embodiment, the functions of each module are as follows: The dual-sensor pre-binding module 101 is used to perform dual-sensor collaborative pre-binding of outbound cargo label information and UAV identification code to obtain the initial association record of the UAV. The infrared guidance and repositioning module 102 is used to guide the UAV to its repositioning based on the initial association record, and obtain the loading verification dataset of the UAV. The time-weighted mutual information decision module 103 is used to perform time-weighted mutual information integration decision on the loading verification dataset and the initial association record to obtain the binary loading confirmation instruction of the UAV. The loading status determination module 104 is used to determine the loading status of the UAV based on the binary loading confirmation command, and obtain the unloading operation process dataset of the UAV. The hash closed-loop verification module 105 is used to perform hash closed-loop verification on the loading verification dataset and the unloading operation process dataset to obtain the unloading operation verification result of the UAV. The chain-based traceability file module 106 is used to perform hash chain binding association between the unloading operation verification result and the binary loading confirmation instruction to obtain the complete closed-loop traceability file of the UAV.

[0082] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for tracing the loading and unloading of goods delivered by drones based on dual-sensor collaboration, characterized in that, The method includes: S1. Perform dual-sensor collaborative pre-binding on the outbound goods label information and the drone identification code to obtain the initial association record of the drone; S2. Based on the initial association record, perform infrared guidance and repositioning of the UAV to obtain the UAV's loading verification dataset, including: Based on the initial association record, the UAV's ground nest infrared beacon is activated, and the modulated infrared light signal of the infrared beacon is transmitted; The modulated infrared light signal is subjected to spot resolution to obtain the pose deviation data of the UAV; Based on the pose deviation data, a flight path correction command is injected into the flight control unit of the UAV to obtain the corrected landing path of the UAV. Based on the corrected landing path, the hovering lock status of the UAV after landing is detected to obtain a hovering lock completion signal; Based on the hovering lock completion signal, scan the radio frequency identification reader built into the drone's cargo compartment to obtain the cargo list data inside the compartment; The cargo manifest data inside the cabin is verified against the cargo order number in the initial association record to obtain the loading verification dataset of the UAV; S3. Perform time-weighted mutual information integration on the loading verification dataset and the initial associated record to obtain the binary loading confirmation instruction for the UAV, including: During the infrared-guided return process of the UAV, temporal features are extracted from the loading verification dataset to obtain an instantaneous loading verification feature vector; The instantaneous loading verification feature vector includes the UAV's horizontal offset, vertical height difference, and RFID signal strength at the corresponding moment; The cargo order number and the unique identifier of the UAV in the initial association record are feature-encoded to obtain the feature vector of the initial association record; The loading confirmation index of the UAV is obtained by performing time-weighted mutual information integration on the instantaneous loading verification feature vector and the initial associated record feature vector; The formula for calculating the loading confirmation index is as follows: ； in, The loading confirmation index, For unit step function, For the infrared-guided return process The instantaneous loading verification feature vector collected. This represents the total duration of infrared-guided repositioning. The initial associated record feature vector, This is the conditional mutual information estimate for the UAV. For the drone at time Historical data sets from a previous period of time, To load and verify the feature vector instantaneously, This is an estimate of the conditional entropy. For An exponential function with the natural constant as its base, where The preset time decay coefficient, and , The entropy of the initial recorded feature vector is... The preset confirmation threshold, The default value is a small positive number to prevent smooth terms with a denominator of zero; When the loading confirmation index is 1, a first binary loading confirmation command indicating that the loading is qualified is generated; when the loading confirmation index is 0, a second binary loading confirmation command indicating that the loading is abnormal is generated. S4. Based on the binary loading confirmation command, determine the loading status of the UAV to obtain the unloading operation process dataset of the UAV; S5. Perform hash closed-loop verification on the loading verification dataset and the unloading operation process dataset to obtain the unloading operation verification result of the UAV; S6. The verification result of the unloading operation is linked with the binary loading confirmation instruction by hash chain to obtain the complete closed-loop traceability file of the UAV.

2. The method for tracing the loading and unloading of goods via drone delivery based on dual-sensor collaboration as described in claim 1, characterized in that, The outbound cargo label information and the drone identification code are pre-bound using dual-sensor collaborative methods to obtain the initial association record between the drone and the cargo, including: Read the label information of the outbound goods to obtain the shipment order number and destination code of the outbound goods; The consistency of the cargo tracking number, the destination code, and the unique identifier of the drone is verified to obtain the binding relationship certificate between the drone and the cargo; The binding relationship certificate is stored to obtain the initial association record between the drone and the cargo.

3. The method for tracing the loading and unloading of goods via drone delivery based on dual-sensor collaboration as described in claim 1, characterized in that, After obtaining the loading confirmation index of the drone, the following is also included: If a first-valued loading confirmation command indicating that the loading is qualified is generated, the loading of the UAV is triggered to obtain the cargo in the UAV's cargo hold; A second full scan of the cargo in the cargo hold is performed using radio frequency identification to obtain a second scan cargo list; The consistency of the secondary scanned goods list with the goods order number in the initial associated record is compared to obtain a consistency result; Based on the comparison results, the electronic lock of the UAV cargo compartment is locked to obtain a locking completion signal; Based on the locking completion signal, the onboard sensors of the UAV are used to collect status data to obtain real-time status data of the UAV, including cargo hold locking status and takeoff ready signal. Upon receiving the real-time status data, the system uploads the storage information for the drone's loading completion status record and receives a data reception confirmation signal. Based on the data reception confirmation signal, the data records of the drone from loading completion to infrared deactivation are recorded, and the unloading operation process dataset of the drone is obtained by summarizing the data.

4. The method for tracing the loading and unloading of goods via drone delivery based on dual-sensor collaboration as described in claim 1, characterized in that, After obtaining the loading confirmation index of the drone, the following is also included: If the binary loading confirmation instruction is the second binary loading confirmation instruction, an alarm message indicating a loading abnormality is generated; Based on the alarm information, the loading action of the drone is prohibited, and a manual review work order is obtained.

5. The method for tracing the loading and unloading of goods via drone delivery based on dual-sensor collaboration as described in claim 1, characterized in that, The step of performing a hash-based closed-loop verification of the time-series linkage deviation between the loading verification dataset and the unloading operation process dataset to obtain the unloading operation verification result of the UAV includes: The first key indicator is extracted from the loading verification dataset to obtain the first received signal strength indication time series; The second key indicator is extracted from the unloading operation process dataset to obtain the second received signal strength indication time series; A timing deviation analysis is performed on the first received signal strength indication time series and the second received signal strength indication time series to obtain a timing linkage deviation sequence. A hash operation is performed on the time-series linkage deviation sequence to obtain the unloading operation hash digest; The unloading operation hash digest is compared with the preset loading verification hash benchmark value to obtain the unloading operation verification result of the UAV.

6. The method for tracing the loading and unloading of goods via drone delivery based on dual-sensor collaboration as described in claim 5, characterized in that, After the preset load verification hash baseline value, it also includes: The load verification hash base value is combined with the load verification dataset to form a hash base data packet; The hash base data packet is digitally signed to obtain the signed loading verification hash base value; The signed loading verification hash base value is stored on the blockchain to obtain the storage transaction identifier of the drone.

7. The method for tracing the loading and unloading of goods via drone delivery based on dual-sensor collaboration as described in claim 1, characterized in that, The verification result of the unloading operation is linked to the binary loading confirmation instruction using a hash chain to obtain a complete closed-loop traceability file for the drone, including: The unloading operation verification result and the binary loading confirmation instruction are feature-encoded to obtain traceability data to be bound. Perform an initial hash operation on the data to be bound to obtain the traceability hash value; The traceability hash value and the unloading operation process dataset are packaged together to obtain a traceability data block; The traceability data blocks are chained together to obtain a chained traceability data structure; The chain-based traceability data structure, the loading verification dataset, the unloading operation process dataset, and the drone's evidence transaction identifier are associated and stored to obtain a complete closed-loop traceability file for the drone.

8. A drone delivery cargo loading and unloading traceability system based on dual-sensor collaboration, characterized in that: The system for implementing the drone delivery cargo loading and unloading traceability method based on dual-sensor collaboration as described in claim 1 includes: The dual-sensor pre-binding module is used to perform dual-sensor collaborative pre-binding of outbound cargo label information and UAV identification code to obtain the initial association record of the UAV. The infrared guidance and repositioning module is used to guide the UAV to its repositioning based on the initial association record, and obtain the loading verification dataset of the UAV. The time-weighted mutual information decision module is used to perform time-weighted mutual information integration decision on the loading verification dataset and the initial association record to obtain the binary loading confirmation instruction of the UAV; The loading status determination module is used to determine the loading status of the UAV based on the binary loading confirmation command, and obtain the unloading operation process dataset of the UAV. The hash loop verification module is used to perform hash loop verification on the loading verification dataset and the unloading operation process dataset to obtain the unloading operation verification result of the UAV. The chain-based traceability file module is used to perform hash chain binding and association between the unloading operation verification result and the binary loading confirmation instruction to obtain the complete closed-loop traceability file of the UAV.

Images