Unmanned aerial vehicle unloading method and device, electronic equipment and storage medium

Abstract

The application discloses a UAV unloading method and device, an electronic device and a storage medium, relates to the technical field of UAV control, and comprises the following steps: controlling a UAV to perform a flight task according to a first navigation mode, sending a reaching notification message to a hangar system when reaching a preset hovering point and hovering satisfying a preset stable condition; when determining that a hangar system readiness state satisfies a preset requirement, sending a precision descending start instruction to the hangar system and switching a current navigation mode to a second navigation mode, so as to track a horizontal offset amount of the current UAV relative to the hangar system in the second navigation mode; according to the horizontal offset amount, height data and attitude data, performing horizontal position correction on the UAV and executing UAV landing control; when descending to a preset unloading preparation height and hovering stably, sending a hovering-in-place message to the hangar system; and when receiving a preparation completion signal, controlling the UAV to perform an on-board unloading operation.

Description

Technical Field

[0001] This disclosure relates to the field of unmanned aerial vehicle (UAV) control technology, specifically to a UAV unloading method, apparatus, electronic device, storage medium, and program product. Background Technology

[0002] Logistics drones, leveraging their low-altitude, straight-line transport capabilities, autonomous takeoff and landing, and independence from ground road networks, serve as an important supplement and upgrade to traditional ground logistics. However, in scenarios involving autonomous drone landing and automated unloading, landing positioning relies on a single RTK data source (raw satellite observation data collected by the drone's onboard GNSS module and differential correction data from the RTK base station) for both horizontal and vertical data. This positioning is susceptible to obstruction and multipath interference in the near-field of the hangar system, resulting in insufficient positioning accuracy and failing to meet precise landing requirements. Summary of the Invention

[0003] In view of this, this disclosure provides a method, apparatus, electronic device, storage medium, and program product for unloading cargo from a drone, in order to solve the problems that the data source used in existing drone landing control schemes is easily blocked and subject to multipath interference in the near field of the hangar system, resulting in insufficient positioning accuracy and inability to meet the requirements for precise landing.

[0004] Firstly, this disclosure provides a method for unloading cargo from a drone, applied to a drone flight controller in a drone cargo unloading control system. The drone flight controller is communicatively connected to an airborne photodiode array module, a hangar system, and a ground station in the drone cargo unloading control system. The method includes: when the self-test status of the airborne photodiode array module is determined to be normal, controlling the drone to execute a flight mission according to a first navigation mode; when the drone executes the flight mission according to the first navigation mode and reaches a preset hovering point corresponding to the hangar, determining whether the hovering meets preset stability conditions, wherein the horizontal distance between the preset hovering point and the center of the hangar system's take-off and landing platform is less than or equal to a preset horizontal distance, and the height between the preset hovering point and the hangar system's take-off and landing platform is less than or equal to a preset height; when the hovering meets the preset stability conditions, sending an arrival notification message to the hangar system, causing the hangar system to send a hangar system status message after receiving the arrival notification message; and when the drone reaches a preset hovering point according to the received hangar system status message, determining whether the hovering meets preset stability conditions. The system status message confirms that the hangar system's readiness status meets preset requirements. It then sends a fine-descent start command to the hangar system and switches the current navigation mode to a second navigation mode. In this second navigation mode, it continuously tracks the horizontal offset of the UAV relative to the hangar system. This horizontal offset is calculated by the onboard photodiode array module, activated in the second navigation mode, based on the sensed infrared beacons of the hangar system. Based on the continuously tracked horizontal offset of the UAV relative to the hangar system, the UAV's altitude relative to the hangar system's takeoff and landing platform, and the UAV's own attitude data, it performs horizontal position corrections and executes UAV landing control. When the UAV descends to the preset unloading preparation height and hovers stably, it sends a hovering position message to the hangar system. This causes the hangar system to perform a self-check preparation for unloading based on the hovering position message and send back a preparation completion signal. Upon receiving the preparation completion signal, it controls the UAV to perform the onboard unloading operation.

[0005] Secondly, this disclosure provides a drone unloading device, applied to a drone flight controller in a drone unloading control system. The drone flight controller is communicatively connected to an airborne photodiode array module, a hangar system, and a ground station in the drone unloading control system. The device includes: a first control module, used to control the drone to execute a flight mission according to a first navigation mode when the self-test status of the airborne photodiode array module is determined to be normal; a first judgment module, used to judge whether hovering meets preset stability conditions when the drone executes a flight mission according to the first navigation mode and reaches a preset hovering point corresponding to the hangar, wherein the horizontal distance between the preset hovering point and the center of the hangar system's take-off and landing platform is less than or equal to a preset horizontal distance, and the height between the preset hovering point and the center of the hangar system's take-off and landing platform is less than or equal to a preset height; and a first sending module, used to send a signal to the hangar system when hovering meets the preset stability conditions. The arrival notification message enables the hangar system to send a hangar system status message upon receiving the arrival notification message. The second sending module, when determining from the received hangar system status message that the hangar system's readiness status meets preset requirements, sends a fine-descent start command to the hangar system and switches the current navigation mode to a second navigation mode. In the second navigation mode, it continuously tracks the horizontal offset of the current UAV relative to the hangar system. This horizontal offset is calculated by the onboard photodiode array module activated in the second navigation mode based on the sensed infrared beacons of the hangar system. The second control module, based on the continuously tracked horizontal offset of the current UAV relative to the hangar system, the altitude data of the UAV relative to the hangar system's takeoff and landing platform, and the UAV's own attitude data, corrects the UAV's horizontal position and executes UAV landing control. The third sending module, when the UAV descends to a preset unloading preparation height and hovers stably, sends a hovering position message to the hangar system, enabling the hangar system to perform unloading self-check preparation based on the hovering position message and feed back a preparation completion signal. The third control module, upon receiving the preparation completion signal, controls the UAV to perform the onboard unloading operation.

[0006] Thirdly, this disclosure provides an electronic device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the unloading method of the drone described in the first aspect or any corresponding embodiment.

[0007] Fourthly, this disclosure provides a computer-readable storage medium storing computer instructions for causing a computer to execute the unloading method of the drone described in the first aspect or any corresponding embodiment.

[0008] Fifthly, this disclosure provides a computer program product, including computer instructions for causing a computer to execute the unmanned aerial vehicle unloading method of the first aspect or any corresponding embodiment described above.

[0009] The drone unloading method provided in this disclosure relies on the relative offset calculated by the PD array, the relative altitude of the laser, and the attitude data of the IMU to continuously correct the fuselage position in a closed loop, achieving centimeter-level precise descent. After the drone reaches the unloading height and hovers stably, it automatically unloads the cargo after coordinating with the hangar to confirm that the unloading conditions are complete. This segmented multi-source fusion navigation strategy can avoid the problem of GNSS / RTK obstruction failure around the hangar. Long-distance cruise relies on RTK to achieve stable navigation over a wide area, while near-field navigation relies on near-field sensing equipment to ensure alignment accuracy. At the same time, a multi-layered safety interlock is formed through hardware pre-testing, two-way interaction with the hangar, and multiple state checks before unloading, which greatly reduces the risks of misaligned landing, cargo falling, and equipment collision, and is suitable for the fully automated drone unloading operation requirements of unmanned stations. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the specific embodiments or related technologies of this disclosure, the accompanying drawings used in the description of the specific embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a schematic diagram corresponding to the drone unloading method according to the embodiments of this disclosure; Figure 2A , Figure 2B , Figure 2C , Figure 2D , Figure 2E This is a schematic diagram corresponding to the drone unloading method according to the embodiments of this disclosure; Figure 3 This is a structural block diagram of a drone unloading device according to an embodiment of the present disclosure; Figure 4 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present disclosure. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0013] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0014] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware, such as the electronic device, application, server, or storage medium performing the operations of this disclosed technical solution, based on the prompt message.

[0015] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0016] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0017] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0018] According to an embodiment of this disclosure, a method for unloading cargo using a drone is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0019] Logistics drones, leveraging their low-altitude, linear transport capabilities, autonomous takeoff and landing, and independence from ground road networks, serve as a crucial supplement and upgrade to traditional ground logistics. However, in related technologies, the autonomous landing and automated unloading scenarios for drones rely on a single RTK data source (raw satellite observation data collected by the drone's onboard GNSS module and differential correction data from the RTK base station) for horizontal and vertical data. This data is susceptible to obstruction and multipath interference in the hangar system's near-field, resulting in insufficient positioning accuracy and failing to meet precise landing requirements. Furthermore, the lack of a unified communication protocol between the drone, hangar system, onboard peripherals, and ground station leads to non-standard message formats and inconsistent timing, resulting in lost commands and response timeouts. The unloading process lacks closed-loop control, causing a disconnect between onboard peripheral actions and the hangar system's state, potentially leading to unloading failures and hardware malfunctions. Finally, the inadequate anomaly handling mechanism lacks standardized strategies for handling communication timeouts, data anomalies, and positioning failures, resulting in low system reliability and hindering unattended operation.

[0020] This embodiment provides a UAV unloading method, applied to the UAV flight controller in a UAV unloading control system. The UAV flight controller is communicatively connected to the onboard photodiode array module (PD array board), hangar system, and ground station in the UAV unloading control system. The UAV unloading control system is based on PX4 firmware secondary development, constructing a collaborative execution architecture with dual communication links, a five-stage state machine, and multi-source positioning fusion. This enables centimeter-level precise landing and automated unloading in near-field hangar scenarios and GNSS failure scenarios. The UAV terminal adopts the MAVLink 2.0 protocol (MicroAir Vehicle Link, a lightweight open-source binary communication protocol for UAVs) and integrates communication modules, serial communication modules, data source switching modules, process control modules, and exception handling modules. In the following embodiment, message numbers "2000~2007" belong to MAVLink custom service instruction IDs, where "2000" represents arrival notification, "2002" represents the start of precise descent, "2003" represents hovering in place, and "2007" represents other information. The "step0~step5" segments represent the execution status indicators of each step within the unloading sub-process, and the process progress labels reported by the flight control system to the ground station. Step 0 represents the unloading request and pre-inspection (the UAV flight control system initiates a request to check the position / confidence / interlock), corresponding to PD (Airborne Payload Unloading Controller) serial port 0x80: ACK (received and ready to execute); Step 1 represents the execution of the release action (driving the servo motor / electromagnet or robotic arm to perform the release); Step 2 represents the release result detection (weight / photoelectric / limit / visual confirmation that the cargo has been released); Step 3 represents the mechanism reset and safety lock (retracting the mounting parts, resetting the mechanical structure and locking); Step 4 represents the final status report and end cleanup (PD and FC confirm the success of all sub-steps, record the log and notify the hangar / ground station); Step 5 represents the final recovery / closing stage of the entire landing and unloading mission; Stage 0 [Self-check failure]: Report to Step 5. In the event of a failure, self-check is a prerequisite for the mission. A failure in the prerequisite self-check means that the entire mission cannot proceed to the final step 5 complete process. Reporting a step 5 failure indicates that the entire mission has failed. Stage 3 [Continued Tracking Anomaly]: Reporting a step 5 anomaly indicates a tracking failure during the precise landing, interrupting the landing process and preventing the completion of subsequent unloading and cleanup. Reporting a step 5 anomaly indicates that the mission has been abnormally terminated. Stage 5 [Flight Away and End]: Reporting a step 5 recovery event indicates that the UAV has successfully completed unloading and normal flight away. This means that stage 5 has been executed normally and the mission has successfully completed the entire process. Reporting a step 5 recovery event informs the ground station that the entire autonomous precise landing and unloading mission has been successfully closed.

[0021] The PD array board is installed at the bottom of the drone fuselage, facing downwards towards the hangar system's light source. It is used to calculate the drone's XY horizontal offset relative to the hangar system, providing near-field optical positioning data to assist in centimeter-level centering and landing. The drone flight controller is installed in the built-in shock-absorbing cabin in the middle of the drone fuselage. It is the core brain of the entire machine, integrating multi-sensor positioning, executing a five-stage operation state machine, scheduling dual-link communication, autonomously handling various faults, and controlling the flight and unloading mechanisms. The hangar system is installed inside the fixed take-off and landing station on the top surface of the hangar system platform. It is located at the ground docking station, emitting positioning light sources, communicating two-way with the drone, cooperating to complete automatic unloading, and locally outputting RTK differential data. The ground station is installed on the operation and maintenance office console or an outdoor portable terminal. It is responsible for top-level scheduling and management, issuing delivery tasks, monitoring the status of all equipment, receiving fault alarms, and remotely controlling the drone and the hangar system. Figure 1 This is a flowchart of a drone unloading method according to an embodiment of the present disclosure, such as... Figure 1 As shown, combined with Figure 2A , Figure 2B , Figure 2C , Figure 2D , Figure 2E The corresponding process sequence diagrams for each stage illustrate the solution described in the embodiments of this application. Figure 2A Corresponding to the "Phase 0 Self-inspection Gating" phase, Figure 2B Corresponding to "Phase 1: Reaching and stabilizing above the hangar system" and "Phase 2: Hangar system guidance and fine-tuning preparation," Figure 2C Corresponding to "Phase 3 Fine-tuning Tracking", Figure 2D Corresponding to "Phase 4 Unloading Process", Figure 2E This corresponds to "Phase 5: Departure and Termination"; the process includes the following steps: Step 101: When it is determined that the self-test status of the airborne photodiode array module is normal, control the UAV to execute the flight mission according to the first navigation mode.

[0022] For example, a flight mission refers to the flight task issued by the ground station to the flight controller, which is the mission mode or route mode in the flight controller. This mission is generally planned by the host computer (computer, etc.) operator and then uploaded to the flight controller. Before initiating a flight mission to the hangar, the UAV flight controller prioritizes the self-test process of the onboard photodiode array module. The onboard photodiode array module can perform tests such as power supply path detection, communication normality, and protocol parsing consistency. After the self-test is completed, the PD array module sends a self-test message containing the results of each test item back to the flight controller. The flight controller parses the message and only allows the subsequent flight mission to start if all test items are normal and the self-test status is determined to be normal. If the self-test is abnormal, the flight process is locked and an alarm is reported to the ground station, prohibiting the UAV from taking off. This avoids the risk of subsequent near-field precise descent lacking a reliable horizontal positioning data source and resulting in alignment misalignment and collision with the hangar due to PD array hardware failure.

[0023] In the first navigation mode of this application embodiment, the data sources fused include: RTK data source, IMU data source, and barometer data source. This mode is suitable for open airspace conditions where the UAV flies from the take-off point to the pre-set hovering area of ​​the hangar. RTK provides global high-precision horizontal coordinates and altitude reference to achieve long-distance flight path guidance. IMU outputs attitude and acceleration data at high frequency to maintain the stability of the fuselage and complete the inertial position recursion. The barometer provides long-term stable altitude constraints to suppress the cumulative drift of the IMU in the altitude direction. In the long-distance flight phase, there is no need to activate the PD array, which reduces the consumption of airborne computing power and avoids the inability to capture the infrared beacon of the hangar at long distances, which would cause invalid PD data to interfere with the positioning calculation.

[0024] In some optional implementations, prior to step 101, the method further includes: sending a self-test command to the airborne photodiode array module to initiate a self-test operation; determining the self-test status of the airborne photodiode array module based on the self-test results received from the airborne photodiode array module; and controlling the UAV to perform a flight mission when the self-test status of the airborne photodiode array module meets the requirements.

[0025] For example, combined Figure 2AAs shown, "Phase 0 Self-Check Gating" is a prerequisite for initiating all landing and unloading tasks. If the self-check fails, the system will prohibit entry into the subsequent flight and automatic unloading process. Specifically, it includes: "Node 1" corresponding to the UAV flight controller sending a self-check command to the onboard PD array board to initiate the PD positioning hardware self-check. The PD array board is the core sensing peripheral for near-field precision landing. The flight controller needs to confirm in advance that the positioning module's hardware, communication, and other functions are normal to ensure that XY offset positioning data can be output in the subsequent environment without GNSS obstruction; After the PD array board corresponding to "Node 2" performs the self-check, it feeds back the self-check status to the UAV flight controller. The timing mark "loop" represents the self-check retry mechanism. If communication is abnormal or the PD board does not respond, the flight controller will repeatedly send the self-check command for multiple retries to avoid misjudging the fault due to momentary communication disturbances; If the flight controller determines that the PD array board is working normally, it jumps to "Node 3", and the system prepares to switch to the multi-source fusion positioning mode, waiting for the UAV to arrive at the hangar system to carry out near-field precision landing. If the self-test fails, the flight controller determines that the PD array board is abnormal. The flight controller corresponding to "Node 4" reports the "Step 5 Failure" fault information to the ground station. The maintenance personnel receive the alarm at the ground station. "Node 5" remains in a fault-locked state in a loop. The system is forcibly intercepted and prohibited from starting landing and unloading actions, and continues to wait for manual troubleshooting.

[0026] Step 102: When the UAV performs a flight mission according to the first navigation mode and reaches the preset hovering point corresponding to the hangar, determine whether the hovering meets the preset stability conditions. The preset hovering point is less than or equal to the preset horizontal distance from the center of the hangar system take-off and landing platform and less than or equal to the preset height from the hangar system take-off and landing platform.

[0027] For example, after takeoff, the UAV operates in the first navigation mode, integrating RTK, IMU, and barometer data to complete global route navigation, relying on RTK global coordinates to guide the UAV towards the hangar. The system pre-configures distance thresholds: a preset horizontal distance and a preset unloading preparation altitude. When the flight control calculates that the horizontal coordinate difference between the UAV and the center of the hangar takeoff and landing platform is ≤ the preset horizontal distance and the vertical relative altitude is ≤ the preset altitude, it determines that the UAV has reached the preset hovering point and entered the effective sensing area where the hangar infrared beacon can be captured by the onboard PD array module. These dual distance thresholds constitute the near-field switching judgment boundary, avoiding frequent navigation mode switching and flight jitter caused by the UAV repeatedly drifting in and out of the sensing area. Only when both horizontal and altitude distance constraints are met can the PD array module stably capture the hangar infrared beacon and calculate the effective XY horizontal offset. If the range is exceeded, the infrared light signal intensity is insufficient, the field of view is blocked, the PD cannot output reliable positioning data, and the conditions for switching to the second navigation mode to carry out precise descent tracking are not met. Upon reaching the preset hovering point, the flight controller continuously and periodically collects IMU attitude data, RTK position data, and the aircraft's three-axis velocities, cyclically verifying preset stability conditions. These stability conditions include multi-dimensional constraints, such as the UAV's three-axis translational velocity being less than a velocity threshold, roll / pitch / yaw attitude angle fluctuations being less than an angle threshold, and continuous multi-cycle position offsets not exceeding permissible errors. If the aircraft shakes or drifts violently, directly initiating precise descent tracking will cause continuous jumps in horizontal offset data, resulting in the UAV continuously and significantly correcting its position, which can easily lead to alignment deviations and collisions with the hangar. Therefore, it is necessary to wait for the aircraft's attitude and position to fully converge and stabilize before sending an arrival notification message to the hangar to initiate a collaborative handshake between the two devices.

[0028] Specifically, such as Figure 2B As shown, after the Phase 0 self-check passes, the system enters "Phase 1: Reaching and stabilizing above the hangar system". In this phase, "2000", "2001", and "2002" are custom service message numbers based on MAVLink 2.0. Among them, "Node 6" represents the UAV flight control executing the flight mission, flying to the preset hovering point above the hangar system, and continuously cyclically judging whether the hovering position, attitude, and speed meet the stability threshold. Only when the hovering reaches the stability condition is the descent allowed to avoid initiating fine descent guidance while the UAV is shaking or drifting, thus preventing alignment errors. After "Node 7" hovers and stabilizes, the flight control sends the number "2000" representing the arrival notification message to the hangar system, informing the hangar system that the UAV has arrived above the target airspace and requesting the hangar system to prepare for fine descent, such as turning on the infrared positioning light source and being ready to receive communication.

[0029] Step 103: When the hovering meets the preset stability conditions, an arrival notification message is sent to the hangar system, so that the hangar system sends a hangar system status message after receiving the arrival notification message.

[0030] For example, the UAV flight controller can send a predefined MAVLink custom arrival notification message, corresponding to the service message numbered "2000," to the hangar system via the onboard wireless communication link. This message carries core identification information such as the UAV equipment number, current positioning status, and mission stage identifier, proactively informing the hangar that the UAV has safely arrived at the near-field operation area and that its attitude is stable, requesting the hangar to initiate preparations for precise landing. Upon receiving this notification, the hangar simultaneously performs internal self-checks, such as checking the power supply and illumination status of the infrared positioning beacon, the door mechanism, the unloading platform, and the wireless communication link to ensure they are functioning correctly, thus completing the readiness verification of all supporting equipment.

[0031] Specifically, such as Figure 2B In the "Phase 2: Hangar System Guidance and Precision Landing Preparation" shown, after the hangar system receives the message, "Node 8 (loop)" continuously sends "2001 Hangar System Status Message" to the UAV flight controller. The flight controller continuously receives this message, obtains the real-time readiness status of the hangar system's doors, light source, and unloading mechanism, and then enters the branch judgment "alt two branches": "Branch 1: Received valid hangar system status" in "Node 9" indicates that the flight controller confirms that the hangar system is ready and sends "2002 Start Precision Landing Command" to the hangar system; "Node 10" the UAV flight controller switches the navigation mode (first navigation mode) to the precision landing mode (second navigation mode), enables PD array board, IMU and laser ranging multi-source fusion positioning, continuously tracks the offset in closed loop, and prepares for close-range precision landing; "Branch 2: No valid hangar system status for a long time (such as communication interruption / hangar system failure)" in "Node 11" If the flight control system fails to receive a valid hangar system status message after a timeout, it determines that the coordination has failed; it triggers an RTL (Return-To-Launch) automatic return-to-home mechanism, repeatedly executing the return-to-home logic and abandoning the landing mission. The system relies on custom messages "2000 / 2001 / 2002" to complete a two-way handshake between the UAV and the hangar system, avoiding timing conflicts such as "UAV blindly descending while the hangar system is not ready." Only after confirming the hangar system's readiness does it switch to the near-field precision landing navigation mode relying on the PD array, avoiding premature positioning changes that could cause flight instability. If communication with the hangar system is lost, it does not attempt a forced landing but activates return-to-home protection to avoid the risk of collision due to the lack of an infrared light source from the hangar system and the inability of the PD array to locate the aircraft.

[0032] Step 104: When it is determined from the received hangar system status message that the hangar system readiness status meets the preset requirements, a fine descent start command is sent to the hangar system and the current navigation mode is switched to the second navigation mode to continuously track the horizontal offset of the current UAV relative to the hangar system in the second navigation mode. The horizontal offset is calculated by the airborne photodiode array module enabled in the second navigation mode based on the sensed infrared beacon of the hangar system.

[0033] For example, the flight controller sends a precision landing start command to the hangar, informing the hangar that the UAV has officially started the near-field landing process. Upon receiving the command, the hangar maintains stable operation of the infrared beacon and simultaneously locks the takeoff and landing platform status to prevent malfunctions of ground mechanisms during landing. In this embodiment, the data sources fused in the second navigation mode include: the horizontal offset data source output by the airborne photodiode array module, the IMU data source, the barometer, and the laser ranging data source. After switching to the second navigation mode, the airborne photodiode array module is officially activated. The module continuously collects the light signal from the hangar's infrared beacon and calculates the XY plane offset of the UAV relative to the center of the hangar's takeoff and landing platform by the light intensity difference of the photosensitive channels. One possible implementation is to collect the light intensity of all channels of the PD array. The coordinates of the light spot array are solved using the centroid formula. The specific formula is as follows:

[0034]

[0035] in, Let these be the coordinates of the barycenter of the light spot in the column direction within the PD array plane. Let these be the coordinates of the centroid of the light spot in the row direction within the PD array plane. Number the rows of the PD array. Number the columns of the PD array. For the first line, number List the light intensity values ​​output by the PD photosensitive unit; substitute them into the calibration zero point. The relative deviation of the array plane spot is obtained; combined with the real-time height H and equivalent focal length f, the original horizontal offset is initially solved, and the array mounting angle is compensated by the calibrated mounting rotation matrix R. Then, the XY offset of the UAV relative to the hangar center in the body coordinate system is output. The data is sent to the flight controller, as shown in the following formula:

[0036]

[0037] The flight controller obtains the horizontal offset based on the communication protocol with the PD array board. Unlike the RTK latitude and longitude coordinates that depend on long distances, this offset is a local relative coordinate and is not affected by GNSS obstruction or altitude reference error. It can be used for centimeter-level centering and alignment above the hangar.

[0038] Step 105: Based on the continuously tracked horizontal offset of the UAV relative to the hangar system, the altitude data of the UAV relative to the hangar system's take-off and landing platform, and the UAV's own attitude data, perform horizontal position correction for the UAV and execute UAV landing control.

[0039] For example, after entering the second navigation mode, the flight controller initiates a closed-loop tracking cycle, periodically reading the real-time horizontal offset output by the PD array as the core input for horizontal position correction. This is combined with the relative altitude from laser ranging and the UAV attitude data output by the IMU to participate in multi-source fusion filtering. Specifically, using the real-time XY offset of the PD array as the primary observation, combined with IMU attitude compensation for offset reading errors, the PI closed-loop controller outputs horizontal velocity commands to continuously pull the UAV closer to the hangar center, constantly reducing the horizontal offset and ensuring centered positioning throughout the landing. Using the relative altitude from laser ranging as the primary vertical data source, it can be supplemented with barometer and IMU altitude recursive data for auxiliary constraints, slowly decreasing altitude according to a preset descent speed, dynamically adjusting the descent rate; the lower the altitude, the flatter the descent speed.

[0040] Specifically, such as Figure 2CAs shown, after completing the handshake and switching to the fine descent navigation mode in Phase 2, the system enters "Phase 3: Fine Descent Tracking". In "Node 12", the UAV flight controller sends a command to the onboard PD array board to initiate the real-time horizontal offset calculation task, officially starting near-field horizontal alignment tracking. "Loop" represents continuous alignment tracking. "Node 13" represents the PD array board continuously sensing the infrared beacons of the hangar system, calculating the XY horizontal offset of the UAV relative to the center of the hangar system in real time, and continuously transmitting it back to the UAV flight controller. "Node 14" represents the flight controller cyclically executing the control logic, integrating the horizontal offset output by the PD, the laser ranging altitude, and the IMU attitude data, continuously correcting the horizontal position, and slowly reducing the altitude to achieve descent while aligning with the center. In the "ALT Dual Branch Decision," "Branch 1: Continuous Tracking Abnormality" indicates that the flight controller cannot obtain effective XY offset for a long time, the offset continues to exceed the limit, and the positioning data changes abruptly, indicating that the tracking has failed. At "Node 15," the flight controller reports a Step 5 abnormality alarm to the ground station. At "Node 16," the RTL automatic return to home is triggered, terminating the landing process and avoiding the risk of misaligned landing and collision with the hangar system. "Branch 2: Normal Tracking and Reaching Near-Ground Hovering" indicates that the UAV continues to track and align, and descends smoothly to the designated near-ground altitude above the hangar system, meeting the hovering conditions. At "Node 17," the flight controller sends a custom message "2003 Hovering in Place" to the hangar system, notifying the hangar system that the UAV has been accurately aligned and has reached the unloading preparation altitude, and the process moves to the next stage of the automatic unloading process.

[0041] Step 106: When the drone descends to the preset unloading preparation altitude and hovers stably, it sends a hovering in place message to the hangar system, so that the hangar system performs unloading self-check preparation based on the hovering in place message and sends back a preparation completion signal.

[0042] For example, under the control of the second navigation mode, the UAV continuously corrects its horizontal trajectory and descends slowly. The laser rangefinder measures the altitude relative to the hangar platform, which drops to the preset unloading preparation altitude. This altitude allows for a safety clearance, ensuring that the UAV is within the effective unloading area of ​​the hangar while avoiding the risk of collision due to the UAV being too close to the platform. The flight control continuously verifies multi-cycle data, ensuring that the three-axis horizontal velocity, roll / pitch / yaw fluctuations, and PD horizontal offset are all less than their respective thresholds. It is determined that there is no continuous drift or violent shaking of the UAV. Only when both altitude and stability conditions are met simultaneously will message interaction be executed. If the altitude is up to standard but the UAV continues to shake, the flight control will maintain small horizontal corrections until the hovering state converges. The flight control sends a hovering position message to the hangar via wireless communication. The message can carry the UAV's current altitude, positioning confidence, and mission stage identifier to notify the hangar that the UAV has reached the standard unloading operation position and is in stable alignment, allowing the start of the unloading process. After receiving the message, the hangar simultaneously conducts a full set of pre-unloading self-inspections, such as checking the locking mechanism of the take-off and landing platform, cargo docking device, safety protection components, and communication link status. After all the hangar self-inspections are passed, it sends a ready-to-complete signal to the drone as a ground-side permission mark for the unloading operation.

[0043] Step 107: Upon receiving the preparation completion signal, control the drone to perform the onboard unloading operation.

[0044] For example, when the preparation completion signal is received, it indicates that the ground hangar unloading mechanism has completed its self-inspection and is ready to receive cargo. Specifically, after the UAV flight control sends the "2003 Hover Position Message" to the hangar system, it enters "Phase 4 Unloading Process," as follows. Figure 2D As shown, relying on the MAVLink custom messages "2004~2006" for coordination, and simultaneously linking the airborne PD array positioning function, the process is divided into two branches: the "Hangar System Ready" normal unloading branch and the "Hangar System Abnormal / Waiting Timeout" fault protection branch. In the normal branch: In the "Hangar System Ready" process, after the "Node 18" hangar system receives the "2003 Hovering in Position Message" from the UAV, it completes its own door and unloading platform self-check preparation, and sends the "2004 Ready to Complete Signal" to the UAV flight controller, informing the UAV hangar system that the unloading conditions are met; "Node 19" flight controller receives the "2004 Ready to Complete Signal", sends the unloading start command to the airborne unloading mechanism, and adjusts the current precise alignment attitude to perform the airborne unloading operation in conjunction with the data source during the hovering unloading process to compensate for the offset caused by airflow.

[0045] In some optional implementations, step 107 includes: reporting an unloading start message to the ground station and receiving an unloading confirmation signal after the UAV's onboard unloading mechanism has performed the unloading action; sending unloading start success information to the ground station based on the unloading confirmation signal; controlling the UAV to execute a multi-step unloading process and synchronizing the status information of each step in the unloading process to the ground station; and when the multi-step unloading process is completed, reporting unloading success information to the ground station and sending an unloading completion message to the hangar system.

[0046] For example, at "Node 20", the flight controller reports "2005 step0 (unloading start message)" to the ground station, notifying the maintenance backend that the unloading action has officially started; at "Node 21", after the airborne unloading mechanism performs the action, it returns an unloading confirmation signal to the flight controller; at "Node 22", the flight controller receives the unloading confirmation signal and reports "step0 successful (unloading start successful information), step1 in progress" to the ground station, continuously pushing the unloading step status; at "Node 23", the airborne unloading mechanism completes the release of cargo and returns unloading completion feedback to the flight controller; at "Node 24", the flight controller summarizes all step results and reports "step1~step4 all successful" to the ground station; at "Node 25", the flight controller sends "2006 unloading completion message" to the hangar system, marking the completion of this round of automatic unloading procedures, and the system can enter the subsequent flight departure mission; in the "fault branch: hangar system abnormal or waiting timeout", if the flight controller cannot receive the hangar system's 2004 message for a long time... If a ready message is received, or if a hangar system malfunction is detected and the unloading conditions are not met, then the flight control system at "Node 26" will report the corresponding step as having failed to execute; "Node 27" will trigger the RTL (Automatic Return-to-Home) system to abandon the unloading mission.

[0047] The drone unloading method provided in this embodiment relies on the relative offset calculated by the PD array, the relative altitude of the laser, and the IMU attitude data to continuously correct the fuselage position in a closed loop, achieving centimeter-level precise descent. After the drone reaches the unloading height and hovers stably, it automatically unloads the cargo after coordinating with the hangar to confirm that the unloading conditions are complete. This segmented multi-source fusion navigation strategy can avoid the problem of GNSS / RTK obstruction failure around the hangar. Long-distance cruise relies on RTK to achieve stable navigation over a wide area, while near-field navigation relies on near-field sensing equipment to ensure alignment accuracy. At the same time, through hardware pre-testing, two-way interaction with the hangar, and multiple state checks before unloading, a multi-layered safety interlock is formed, which greatly reduces the risks of misaligned landing, cargo falling, and equipment collision, and is suitable for the fully automated drone unloading operation requirements of unmanned stations.

[0048] In some optional implementations, the method further includes: in response to receiving a drone take-off command, controlling the drone to initiate a take-off action and switching the current navigation mode from the second navigation mode back to the first navigation mode after reaching a preset airspace, wherein the drone take-off command is issued by the hangar system after receiving the unloading completion message; and sending an unloading process completion message to the ground station.

[0049] For example, upon receiving the takeoff command, the flight controller immediately releases the unloading attitude lock and initiates the ascent and takeoff process: first, it vertically lifts the UAV to a preset safe airspace altitude, which is outside the effective sensing range of the hangar's infrared beacon; during the ascent, the IMU and barometer are temporarily maintained to ensure attitude and altitude stability, while the PD array horizontal tracking is turned off, and no horizontal alignment correction is performed; after the UAV reaches the preset airspace, it automatically switches from the second navigation mode back to the first navigation mode. After the navigation mode switch is complete and the takeoff action is successfully initiated, the flight controller uploads a takeoff process completion message to the ground station. The message carries information such as the job number, the takeoff completion status, the current navigation mode, and the UAV's position. Upon receiving the message, the ground station updates the task log, marking the complete closed loop of this fully automated takeoff operation, facilitating backend recording and allowing maintenance personnel to view the entire operation status. If any abnormalities occur during takeoff or mode switching, this message will not be reported, and a ground alarm will be triggered simultaneously to indicate the fault.

[0050] Specifically, such as Figure 2E As shown, "Phase 5: Flight Departure and Completion" is the final stage of the entire autonomous landing and unloading mission. At "Node 28," after receiving the "2006 Unloading Complete Message" from the UAV, the hangar system sends a custom message "2007 Flight Departure Command" to the UAV flight controller, notifying the UAV that it can initiate the flight departure maneuver. At "Node 29," the UAV flight controller receives the flight departure command and executes the preset takeoff and departure strategies. It leaves the confined near-field area of ​​the hangar system and no longer relies on the hangar system's infrared light source and PD array board for positioning, maintaining flight control in a loop. At "Node 30," the flight controller reports the Step 5 recovery event to the ground station, informing the operations platform that the entire unloading process has successfully concluded, the UAV enters normal flight mode, and the ground station updates the mission status. At "Node 31," the UAV continues to execute the flight mission. Upon reaching the target location and landing, or after completing all delivery tasks, the entire landing-unloading process is terminated in a closed loop, awaiting the next mission scheduling.

[0051] In some optional implementations, the method further includes: if an abnormal collaborative data interaction is detected at any stage from the start of the hangar system entering the precision unloading state to the completion of unloading, an abnormal interaction process is triggered and the drone is controlled to return to its home location.

[0052] For example, once the hangar system is ready and the UAV switches to the second navigation mode to enter the precision unloading process, the flight control system will monitor the collaborative data interaction status between the UAV and the hangar system in real time until the unloading is completed. If any abnormal interaction occurs, such as communication interruption, message timeout, message verification error, hangar equipment failure feedback, or timing command disorder, the abnormal interaction process will be immediately initiated, terminating the current landing, alignment, and unloading-related actions. The UAV will be controlled to rise to a safe altitude and then automatically return to base. At the same time, an alarm message will be reported to the ground station. This forms a safety net in the high-risk operation area in the near field, avoiding safety accidents such as collisions and accidental unloading caused by air-to-ground coordination failure. For details, please refer to the above embodiment, which will not be repeated here.

[0053] As a specific implementation of this application, the UAV unloading control system in this embodiment sets up two independent communication links: The UAV communicates with the hangar and ground station using a custom interaction message with a message ID range of 2000–2007 based on the MAVLink2.0 protocol. The message uses CRC-16 / MCRF4XX algorithm verification, and the communication timeout threshold is set to 500ms. After a timeout, a maximum of 3 message retransmissions are supported. The UAV communicates with the PD array board using a private frame format with fixed header and tail identifiers AA55 and 55AA. The frame sequentially contains instructions, data length, data segment, and CRC16-CCITT checksum. Data is stored in little-endian mode. This link can retry a maximum of 2 times after a 500ms communication timeout. Secondly, by using a unified protocol stack and standard message model, heterogeneous device communication is transformed into a verifiable, orchestratable, and recoverable transactional communication process. Reliability is ensured through state machine constraints on timing, timeout retransmission and integrity checks, and an observable closed loop is achieved through end-to-end status reporting. Therefore, it effectively solves problems such as inconsistent protocols, non-standard formats, chaotic timing, and lost commands and timeout responses in existing technologies. When a drone flies near the hangar (near-field area), satellite / GNSS and RTK signals are easily blocked and become ineffective. This embodiment does not rely on satellite positioning but instead uses multi-sensor fusion of IMU, laser, PD array board, and barometer to perform relative positioning with the hangar center as the coordinate origin. It automatically switches to near-field precision landing mode when distance and altitude conditions are met, dynamically allocates sensor weights, and continuously corrects position deviations through closed-loop control, achieving centimeter-level precise landing. Specifically, the IMU outputs attitude position recursive data at high frequency; the barometer locks the vertical height reference; the laser ranging constrains the precise vertical height; the PD array board calculates the XY offset relative to the hangar center in real time and corrects the lateral positioning drift. The near-field mode start-stop logic calculation formula is shown in the following formula (1): (1) in, The straight-line distance between the drone and the center of the hangar on the horizontal plane; ( ) represents the current planar coordinates of the UAV; ( ) represents the plane coordinates of the hangar center. The horizontal distance threshold for entering near-field mode. To exit near-field mode at the horizontal threshold, The height threshold for entering near-field mode, To exit near-field mode, the height threshold is required. When horizontal distance ,high Furthermore, when the observation quality is satisfactory, the system enters near-field mode. By setting multiple thresholds, it eliminates jitter caused by frequent mode switching.

[0054] After entering the near field, the weight of laser and PD observation fusion is increased, and low-speed precision landing is performed. The offset is corrected by updating the estimated layer residuals and the closed-loop feedback of the control layer error, as shown in equations (2) and (3) below: (2) (3) in, This is the horizontal position deviation vector; XY offset of the UAV relative to the hangar center, output by the PD array board, represents how much the aircraft has deviated; Formula (3) is the horizontal speed control formula (PI controller). The final horizontal target speed output by the drone. This is the feedforward reference speed (the basic moving speed for a slow descent). This is a scaling factor used to correct the offset based on its magnitude. These are integral coefficients used to eliminate long-term residual offsets. To accumulate position deviations and eliminate static drift, this application employs a hierarchical weighted differential fusion method. During long-distance flight, the IMU, barometer, and RTK have high weights. After switching to near-field precision reduction mode, the fusion weights of laser ranging and PD array board are increased, so that when the flight control calculates the near-field position, more or only PD and laser data are considered, and the RTK / GNSS positioning weights are reduced (e.g., set to 0). When satellite signals are lost, the system can rely entirely on the high-weighted PD, laser, and IMU.

[0055] The specific precision landing and unloading process includes: system self-check and takeoff preparation: the UAV completes a self-check of the whole machine and the onboard sensing equipment, and after confirming that the status is normal, it takes off and enters the mission route; the UAV flies to the predetermined altitude above the hangar and hovers, sending a positioning signal to the hangar; After receiving the data, the hangar activates the laser guidance array and begins outputting positioning and ranging data. The predetermined altitude can be determined based on a preset mission waypoint; in this embodiment, the preset altitude can be selected as 80m. The UAV seamlessly switches from conventional positioning mode to laser ranging fusion positioning mode, entering near-field precise landing guidance. Specifically, the UAV does not switch to single laser positioning in the near field. Instead, after passing self-checks, confirming laser validity, and ensuring hangar readiness, it enters a combined state of data source and control strategy for the near-field precise landing phase. Using continuous estimation from the IMU and barometer as a base, it superimposes hangar laser altitude constraints and PD lateral offset constraints to achieve precise landing. When observations are abnormal or timing exceeds limits, it automatically enters a guard and backoff mechanism to ensure near-field precise landing and system safety. The UAV descends to a low-altitude safe position above the hangar and hovers, handshaking with the hangar to confirm its status. The hangar and onboard peripherals simultaneously enter the unloading ready state. The low-altitude "predetermined altitude" can be determined by the parameter CF_DOCK_HOVHGT. The system is configured such that a setting of 1m can be customized via ground station or host computer. The drone will complete the following sequence in a preset order: hatch opening, cargo mechanism operation, cargo release, mechanism recovery, and hatch closing, all within a closed-loop interlocked loop. Upon completion, it will report a successful unloading status. After confirming unloading with the hangar, the drone will switch back to normal positioning mode and execute takeoff and subsequent return-to-home missions. Anomaly handling measures include: discarding frames with errors or CRC errors and continuously reporting anomalies to the ground station; retrying communication timeouts according to rules, and entering a hovering / return-to-home strategy if the retry fails; automatically filtering positioning anomalies and using valid data from the previous frame; continuously adjusting hover stability if insufficient, and automatically returning to home if timeout occurs.

[0056] This application embodiment uses an inertial-multi-sensor fusion positioning underlying model to construct the UAV state vector, as shown in equation (4) below, which fully characterizes the UAV's position, velocity, attitude, and IMU sensor zero bias error: x = [p, v, q, b_a, b_g]^T (4) Where x is the UAV system state vector; p is the UAV's three-dimensional spatial position; v is the UAV's three-dimensional motion velocity; q is the UAV's attitude quaternion; b_a is the accelerometer zero bias error; and b_g is the gyroscope zero bias error.

[0057] The observation equation is established as shown in equation (5): (5) in, Let i be the output value observed by the i-th type of sensor. For the i-th type of sensor observation mapping model, the theoretical observation value corresponding to the state vector x is represented. The i-th type of sensor observes Gaussian noise, where i is the sensor / positioning mode number, covering multiple types of sensing devices such as GNSS, vision, UWB, and laser ranging.

[0058] The system incorporates observation data from various sensors, including GNSS, laser ranging, PD array, UWB, and vision. It relies on high-frequency IMU data to perform state inertial recursion, and then utilizes external sensor observation information to continuously correct accumulated errors in inertial calculations, outputting continuous high-precision pose estimation results. This provides a fundamental positioning data source for upper-level adaptive weight allocation, dynamic adjustment of filtering parameters, near-field fine descent, and unloading safety assessment.

[0059] In the entire fusion positioning system, the upper-level scheduling core module—the multimodal quality scoring and adaptive weight allocation model—receives the raw sensor data output from the inertial multi-sensor observation model and provides weight inputs to the fusion filtering covariance adaptive adjustment model. Specifically, this is achieved through a mapping function. Combining multi-dimensional quality index vectors of each positioning mode And the current mission phase of the UAV, to obtain the comprehensive quality score for each positioning mode; after normalizing the score, an exponential transformation is performed to obtain intermediate variables. Among them, the original score Normalization yields Then, intermediate variables are obtained through exponential function transformation. Then, through global normalization operation The final weights of each modality participating in pose fusion are output. The quality index vector integrates multi-dimensional parameters such as satellite status, visual feature quality, communication signal-to-noise ratio, and transmission delay. The mission phase identifier distinguishes between cruise, glide, and close-range alignment, enabling differentiated preference allocation for different operational phases. For example, the weights of close-range sensing methods such as PD arrays and laser ranging are increased in the close-range landing phase. The calculated weights are directly supplied to the filtering framework to dynamically adjust the observation confidence, ensuring stable and continuous positioning in obstructed environments.

[0060] An adaptive adjustment model for the covariance of the fusion filter is set up, utilizing the fusion weights of each mode output above. Dynamically adjust the filter parameters to satisfy the relationship , ;in The original observation noise covariance matrix is ​​used as the basis for weight correction to obtain the equivalent observation noise covariance. , The original perceived information matrix is ​​weighted to obtain the equivalent information matrix. When the quality of positioning modality perception deteriorates, the fusion weights... When the noise level decreases, the equivalent observation noise covariance increases and the equivalent information matrix decreases synchronously, prompting the filtering framework to automatically reduce the contribution of this mode to the pose fusion result, realize adaptive weight reduction and isolation of failed sensors, and ensure continuous and stable positioning output under complex conditions such as near-field GNSS occlusion.

[0061] To set up a confidence determination model for pre-release of cargo / landing actions, firstly through the function:

[0062] The local positioning confidence of a single machine is solved by integrating location estimation uncertainty, visual matching quality, communication latency, and unloading mechanism status, where φ is the local comprehensive confidence solution function. The trace of the position covariance matrix represents the uncertainty of the current position estimate; To estimate the covariance matrix for the 3D position of a UAV, Visual image matching overall quality score, For the wireless communication transmission delay between nodes, Airborne load and unloading actuator working status.

[0063] Then rely on the weighted relation The global composite location confidence is obtained by integrating the air-to-ground collaborative confidence score, whereby... , The weighted allocation coefficients are the local confidence and collaborative confidence scores. For the local confidence level of a single drone, The collaborative positioning reliability issued to the hangar / collaborative nodes.

[0064] The system employs triple constraints and logical decision-making.

[0065] ; in, To achieve a preset threshold for global composite confidence, The number of effective localization modes meets the redundancy requirement, i.e., the weights are higher than the threshold. There must be at least M available sensors to prevent a single sensor failure from directly causing danger; locks_ok means that all mechanical locks, electrical interlocks, and unloading mechanism self-tests are normal. If any one of these conditions is not met, the release of goods is prohibited.

[0066] Cargo release and terminal landing actions are only permitted when the global confidence level reaches the threshold, the number of effective sensing modes meets the redundancy standard, and the mechanical and electrical interlocks and unloading mechanisms are all in normal condition. Multiple conditions constrain each other to effectively avoid operational risks such as mis-unloading and collisions caused by positioning inaccuracies, communication disturbances, and mechanism failures.

[0067] A near-range differential correction model for the mobile landing platform is established, satisfying the following relation: ;in The original target location is broadcast wirelessly by the mobile landing platform. The position difference compensation amount is obtained by relying on close-range observations such as UWB, vision, and laser. This provides the precise coordinates of the landing target after differential compensation correction. Addressing the target position lag caused by the motion error of the mobile landing platform itself and the communication transmission delay between devices, the UAV relies on close-range multi-sensor observations to solve for compensation in real time, dynamically correcting the original target coordinates and outputting high-precision target positions for use in the terminal position closed-loop control. This effectively improves the accuracy of close-range alignment and unloading in mobile platform scenarios.

[0068] Set up a data-driven system error learning compensation model, with the following relationship: ;in The error prediction learning function is trained using historical test flight and positioning residual datasets. The model output includes real-time task multi-dimensional contextual features such as ambient wind speed, aircraft attitude perturbation, visual feature density, and historical filtering residuals. This is the overall system error compensation amount. The system incorporates the predicted error compensation amount into the inertial fusion filtering solution link to specifically compensate for various systematic fixed errors such as wind disturbance offset, visual scale drift, and sensor installation deviation, thereby continuously improving the pose estimation accuracy of the UAV during long-term operations and enhancing the long-term operational stability of the entire positioning system.

[0069] When the aircraft performs precision landing and cargo release missions, the system periodically and synchronously collects raw observation data from all sensors, including GNSS, visual cameras, UWB, IMU, lidar, and active beacons. It then sequentially calculates multi-dimensional quality indicators for each type of sensing device, covering parameters such as satellite signal quality, visual feature matching stability, ranging signal-to-noise ratio, and cross-node communication latency. Based on the quality indicator vectors of each modality and the real-time mission conditions, it calls a scoring mapping function to solve for the comprehensive quality score of each positioning modality. After exponential transformation and normalization, it obtains the adaptive fusion weights for each modality. After weight allocation, the system uses an extended Kalman filter or factor graph optimization framework to fuse all valid observations to uniformly estimate the UAV's three-dimensional position, velocity, and attitude quaternions. If the fusion weight of a single positioning modality is lower than a preset effective threshold, the current sensing of that modality is deemed invalid, its fusion contribution weight is automatically reduced, or the observation information is directly removed. Simultaneously, it requests supplementary external observation data from the ground station and cooperating UAV nodes to ensure positioning redundancy. After obtaining the high-precision pose estimation result of the local aircraft, the system packages and broadcasts the current pose, time synchronization stamp, and position covariance matrix to the ground station, mobile landing platform, and other cooperative aircraft. Simultaneously, it receives positioning data from external cooperative nodes and solves the cooperative pose estimate based on global precise time synchronization rules and covariance weighting criteria. The system integrates single-aircraft local confidence, multi-node cooperative confidence, visual matching score, communication latency, and the operating status of the airborne payload mechanism to calculate the global composite positioning confidence. When the global composite confidence exceeds the release action threshold, the number of effective positioning modes meets the minimum redundancy requirement, and all mechanical and electrical interlocks and unloading mechanisms are deemed normal, the system switches to the end-effector close-range alignment high-frequency fine-tuning control loop. It integrates visual, UWB, and mobile platform differential compensation information to continuously iteratively correct the horizontal and vertical position deviations of the UAV relative to the landing target. Once the aircraft attitude error and relative position error are continuously and stably less than the preset judgment threshold, the system executes the cargo release or final landing action. If the global composite confidence level fails to reach the action threshold within the specified time, or if abnormal conditions such as communication link interruption, multi-sensor mode synchronization failure, or load / interlock mechanism failure occur, the system will immediately terminate the end landing and cargo release process, and initiate a graded safety backoff strategy, including raising to a safe altitude to re-execute the alignment process, automatically returning to the take-off and landing origin, uploading fault alarms, and switching to manual remote control takeover mode.

[0070] The drone unloading method provided in this application adopts multi-source fusion positioning to achieve stable and reliable positioning results. It can maintain centimeter-level positioning accuracy even in near-field occlusion environments, with altitude error controlled within ±5cm and XY offset error not exceeding ±10cm. At the communication level, a dual-protocol parallel architecture is adopted, coupled with a unified message ID and verification mechanism to ensure stable communication. The normal communication latency is ≤50ms, and the maximum latency does not exceed 100ms. The overall operation process relies on a five-stage state machine to achieve closed-loop control. The stage interlocking mechanism restricts the entry into the next stage when the previous stage task is not completed, and the unloading success rate can reach 100%. The system can realize integrated anomaly detection, handling, and alarm reporting in all scenarios, and can autonomously handle various faults to meet the needs of unattended operation. At the same time, the whole solution is developed based on the PX4 standard framework, with strong adaptability, and can quickly be compatible with various drone models and different specifications of hangar platforms.

[0071] Employing a MAVLink 2.0 dual-link parallel communication collaborative design, relying on a unified protocol and verification mechanism, it effectively solves the problems of message disorder and command loss; it establishes a five-stage state machine closed-loop control logic, setting stage interlock constraints for the entire process of precise landing and unloading, realizing the synchronization of hangar and various peripheral device actions, and avoiding unloading failures; it achieves seamless switching of multi-source positioning modes, using RTK positioning during the long-distance cruise phase, and automatically switching to inertial, laser ranging, and barometer fusion positioning after entering the hangar near field, with no jumps or interruptions in the positioning process; it has the ability to autonomously handle anomalies in all scenarios, covering various communication, positioning, and hardware failures, and is suitable for unattended and high-risk scenario safety operation requirements; it has the advantage of general mass production adaptability, developed based on an open-source flight control framework, requiring no special hardware and low computing power, and facilitating low-cost and rapid adaptation to multiple aircraft models and platforms while ensuring operational accuracy.

[0072] This embodiment also provides a drone unloading device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0073] This embodiment provides a drone unloading device, applied to the drone flight controller in a drone unloading control system. The drone flight controller is communicatively connected to the airborne photodiode array module, hangar system, and ground station in the drone unloading control system. Figure 3 As shown, it includes: The first control module 201 is used to control the UAV to perform flight missions according to the first navigation mode when it is determined that the self-test status of the airborne photodiode array module is normal. The first judgment module 202 is used to determine whether hovering meets the preset stability conditions when the UAV performs a flight mission according to the first navigation mode and reaches the preset hovering point corresponding to the hangar. The preset hovering point is less than or equal to the preset horizontal distance from the center of the hangar system take-off and landing platform and less than or equal to the preset height. The first sending module 203 is used to send an arrival notification message to the hangar system when the hovering meets the preset stability conditions, so that the hangar system sends a hangar system status message after receiving the arrival notification message. The second sending module 204 is used to send a fine descent start command to the hangar system and switch the current navigation mode to the second navigation mode when it is determined from the received hangar system status message that the hangar system readiness status meets the preset requirements. In the second navigation mode, the module continuously tracks the horizontal offset of the current UAV relative to the hangar system. The horizontal offset is calculated by the airborne photodiode array module enabled in the second navigation mode based on the sensed infrared beacon of the hangar system. The second control module 205 is used to correct the horizontal position of the UAV and execute UAV landing control based on the continuously tracked horizontal offset of the UAV relative to the hangar system, the height data of the UAV relative to the take-off and landing platform of the hangar system, and the attitude data of the UAV itself. The third sending module 206 is used to send a hovering position message to the hangar system when the UAV descends to the preset unloading preparation height and hovers stably, so that the hangar system performs unloading self-check preparation based on the hovering position message and feeds back a preparation completion signal. The third control module 207 is used to control the drone to perform airborne unloading operations when it receives the preparation completion signal.

[0074] The drone unloading device provided in this embodiment relies on the relative offset calculated by the PD array, the relative altitude of the laser, and the attitude data of the IMU to continuously correct the fuselage position in a closed loop, achieving centimeter-level precise descent. After the drone reaches the unloading height and hovers stably, it automatically unloads the cargo after coordinating with the hangar to confirm that the unloading conditions are complete. This segmented multi-source fusion navigation strategy can avoid the problem of GNSS / RTK obstruction failure around the hangar. Long-distance cruise relies on RTK to achieve stable navigation over a wide area, while near-field relies on near-field sensing equipment to ensure positioning accuracy. At the same time, through hardware pre-testing, two-way interaction with the hangar, and multiple state verifications before unloading, a multi-layered safety interlock is formed, which greatly reduces the risks of misaligned landing, cargo falling, and equipment collision, and is suitable for the fully automated drone unloading operation requirements of unmanned stations.

[0075] The drone unloading device provided in this disclosure can execute the drone unloading method provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units described above are the same as in the corresponding embodiments described above, and will not be repeated here.

[0076] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure.

[0077] The following is a detailed reference. Figure 4 The diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present disclosure. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 501, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 502 or a program loaded from memory 508 into random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of the electronic device. The processor 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0078] Typically, the following devices can be connected to I / O interface 505: input devices 506 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 507 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 508 including, for example, magnetic tapes, hard disks, etc.; and communication devices 509. Communication device 509 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0079] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 509, or installed from a memory 508, or installed from a ROM 502. When the computer program is executed by the processor 501, it performs the functions defined in the drone unloading method of embodiments of this disclosure.

[0080] Figure 4The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0081] This disclosure also provides a computer-readable storage medium in which the methods described in this disclosure can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded over a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium may also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the drone unloading method shown in the above embodiments is implemented.

[0082] A portion of this disclosure can be applied to computer program products, such as computer program instructions, which, when executed by a computer, can invoke or provide methods and / or technical solutions according to this disclosure through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, and installation package files. Accordingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions; the computer compiling the instructions and then executing the corresponding compiled program; the computer reading and executing the instructions; or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0083] Although embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for unloading cargo using a drone, characterized in that, A drone flight controller applied in a drone unloading control system, wherein the drone flight controller is communicatively connected to the airborne photodiode array module, the hangar system, and the ground station in the drone unloading control system; the method includes: Once the self-test status of the airborne photodiode array module is confirmed to be normal, the UAV is controlled to execute the flight mission according to the first navigation mode. When the UAV performs a flight mission according to the first navigation mode and reaches the preset hovering point corresponding to the hangar, it is determined whether the hovering meets the preset stability conditions. The horizontal distance between the preset hovering point and the center of the hangar system take-off and landing platform is less than or equal to the preset horizontal distance, and the height between the preset hovering point and the hangar system take-off and landing platform is less than or equal to the preset height. When the hovering condition meets the preset stability condition, an arrival notification message is sent to the hangar system, so that the hangar system sends a hangar system status message after receiving the arrival notification message. When the hangar system readiness status is determined to meet the preset requirements based on the received hangar system status message, a fine descent start command is sent to the hangar system and the current navigation mode is switched to the second navigation mode to continuously track the horizontal offset of the current UAV relative to the hangar system in the second navigation mode. The horizontal offset is calculated by the airborne photodiode array module enabled in the second navigation mode based on the sensed infrared beacon of the hangar system. Based on continuously tracking the current horizontal offset of the UAV relative to the hangar system, the altitude data of the UAV relative to the take-off and landing platform of the hangar system, and the attitude data of the UAV itself, the horizontal position of the UAV is corrected and the UAV landing control is executed. When the drone descends to the preset unloading preparation altitude and hovers stably, it sends a hovering in place message to the hangar system, so that the hangar system performs unloading self-check preparation based on the hovering in place message and sends back a preparation completion signal. Upon receiving the prepared signal, control the drone to perform the onboard unloading operation.

2. The method according to claim 1, characterized in that, Before determining whether hovering meets preset stability conditions when the UAV performs a flight mission according to the first navigation mode and reaches the preset empty hovering point corresponding to the hangar, the method further includes: Send a self-test command to the airborne photodiode array module to initiate the self-test operation of the airborne photodiode array module; The self-test status of the airborne photodiode array module is determined based on the self-test results received from the airborne photodiode array module. When the self-test status of the airborne photodiode array module meets the requirements, the drone is controlled to perform flight missions.

3. The method according to claim 1, characterized in that, Upon receiving the preparedness completion signal, control the drone to perform onboard unloading operations, including: Report unloading start message to the ground station and receive unloading confirmation signal after the UAV's onboard unloading mechanism performs unloading action; Based on the unloading confirmation signal, a successful unloading start message is sent to the ground station; Control the drone to execute a multi-step unloading process and synchronize the status information of each step in the unloading process to the ground station; Once the multi-step unloading process is completed, a successful unloading message is reported to the ground station and an unloading completion message is sent to the hangar system.

4. The method according to claim 3, characterized in that, The method further includes: In response to receiving a drone take-off command, the system controls the drone to initiate a take-off maneuver and, upon reaching a preset airspace, switches the current navigation mode from the second navigation mode back to the first navigation mode. The drone take-off command is issued by the hangar system after receiving the unloading completion message. Send a message to the ground station indicating the end of the unloading process.

5. The method according to any one of claims 1-4, characterized in that, The method further includes: If any abnormality in collaborative data interaction is detected at any stage from the start of the hangar system entering the precision unloading state until the unloading is completed, the abnormal interaction process will be triggered and the drone will be controlled to return to its home location.

6. The method according to any one of claims 1-4, characterized in that, The data sources integrated in the first navigation mode include: RTK data source, IMU data source, and barometer data source; the data sources integrated in the second navigation mode include: horizontal offset data source output by the airborne photodiode array module, IMU data source, barometer, and laser ranging data source.

7. A drone unloading device, characterized in that, A drone flight controller used in a drone unloading control system, wherein the drone flight controller is communicatively connected to the airborne photodiode array module, the hangar system, and the ground station in the drone unloading control system; the device includes: The first control module is used to control the UAV to perform flight missions according to the first navigation mode when it is determined that the self-test status of the airborne photodiode array module is normal. The first judgment module is used to determine whether hovering meets the preset stability conditions when the UAV performs a flight mission according to the first navigation mode and reaches the preset hovering point corresponding to the hangar. The preset hovering point is less than or equal to the preset horizontal distance from the center of the hangar system take-off and landing platform and less than or equal to the preset height from the hangar system take-off and landing platform. The first sending module is used to send an arrival notification message to the hangar system when the hovering meets the preset stability conditions, so that the hangar system sends a hangar system status message after receiving the arrival notification message. The second sending module is used to send a fine descent start command to the hangar system and switch the current navigation mode to the second navigation mode when it is determined from the received hangar system status message that the hangar system readiness status meets the preset requirements. In the second navigation mode, the module continuously tracks the horizontal offset of the current UAV relative to the hangar system. The horizontal offset is calculated by the airborne photodiode array module enabled in the second navigation mode based on the sensed infrared beacon of the hangar system. The second control module is used to correct the horizontal position of the UAV and execute the UAV landing control based on the continuously tracked horizontal offset of the UAV relative to the hangar system, the height data of the UAV relative to the take-off and landing platform of the hangar system, and the attitude data of the UAV itself. The third sending module is used to send a hovering position message to the hangar system when the drone descends to the preset unloading preparation height and hovers stably, so that the hangar system performs unloading self-check preparation based on the hovering position message and sends back a preparation completion signal. The third control module is used to control the drone to perform onboard unloading operations when it receives the preparation completion signal.

8. An electronic device, characterized in that, include: A memory and a processor are interconnected, the memory storing computer instructions, and the processor executing the computer instructions to perform the unloading method of any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the unloading method of the drone as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes computer instructions for causing a computer to perform the unloading method for a drone as described in any one of claims 1 to 6.

Images