Modularized unmanned aerial vehicle parallel operation system

Abstract

The invention discloses a modular unmanned aerial vehicle parallel operation system, and belongs to the technical field of unmanned aerial vehicle logistics. The unmanned aerial vehicle connection platform is arranged on the rack and comprises an unmanned aerial vehicle landing position and an unmanned aerial vehicle take-off position which are physically separated; the bearing and transferring mechanism is arranged in the rack; the integrated operation cabin is arranged in the rack, and a fixed operation station, a goods temporary storage mechanism, a battery storage and replacement mechanism, an operation mechanical arm and a visual positioning system are arranged in the integrated operation cabin; the bearing transfer mechanism is used for integrally transferring the unmanned aerial vehicle landing at the landing position to a fixed operation station in the integrated operation cabin; the operation mechanical arm is arranged on the fixed operation station and performs at least two operations of unloading, loading and battery replacement on the unmanned aerial vehicle in parallel; the visual positioning system provides operation positioning for the operation mechanical arm. According to the invention, through the integrated and modular design of overall transfer and concurrent operation, the system can be flexibly deployed in various logistics network topologies in a plug-and-play manner, and is convenient for rapid expansion.

Description

Technical Field

[0001] This invention relates to the field of drone logistics technology, and in particular to a modular drone parallel operation system. Background Technology

[0002] With the continuous development of drone logistics technology and the deepening of its application scenarios, the level of automation and operational efficiency of ground transfer nodes have become key factors restricting the overall effectiveness of the logistics network. To adapt to the needs of large-scale, high-frequency drone logistics operations, automated drone nests (or drone airports), as core ground infrastructure, need to have efficient fully automated processing capabilities for drone reception, cargo transfer, energy replenishment, and re-takeoff.

[0003] Currently, most common automated drone docking technologies focus on basic functions such as self-parking, charging, and cargo loading / unloading for individual drones. In typical workflows, a sequential operation mode is usually adopted: after the drone lands, a robotic arm or other actuator first removes the delivered cargo, then the battery is replaced or charged, and finally, new cargo is loaded for delivery. While this step-by-step approach automates the process, the sequential execution of each step results in a long overall processing cycle for a single drone at each node, limiting node throughput and making it difficult to meet the efficiency requirements of high-volume logistics scenarios.

[0004] Furthermore, the design of existing automated warehouses is often tightly integrated with the specific logistics network architecture they serve, lacking versatility and flexibility. Most systems are custom-developed for specific network topologies (such as single-hub radial systems), failing to form highly standardized, modular basic operating units. This deficiency makes it difficult for existing warehouses to adapt to diverse network forms, such as star, linear, or grid (including cellular) logistics networks, thus limiting the rapid deployment, replication, and network expansion of ground infrastructure, and increasing the complexity and cost of network construction and subsequent adjustments.

[0005] Therefore, current technology faces two major bottlenecks: first, the serial operation mode of existing automated drone nests leads to low node processing efficiency; second, their design is too tightly coupled with specific networks, lacking standardized modules that can serve as general building blocks. There is an urgent need for an innovative drone nest design that can not only significantly improve single-node processing efficiency through parallel operations, but also serve as a plug-and-play standardized module, flexibly integrating into logistics networks with different topologies, thereby providing fundamental support for building efficient and scalable drone logistics networks. Summary of the Invention

[0006] The purpose of this invention is to provide a modular unmanned aerial vehicle (UAV) parallel operation system to solve the problems of low efficiency, insufficient reliability, lack of standardization and insufficient resource utilization in existing automated UAV systems.

[0007] To achieve the above objectives, this application adopts the following technical solution:

[0008] This application discloses a modular unmanned aerial vehicle (UAV) parallel operation system, comprising:

[0009] frame;

[0010] The drone docking platform is mounted on the frame and includes physically separate drone landing positions and drone takeoff positions;

[0011] A load-bearing and transfer mechanism is disposed inside the frame;

[0012] An integrated work cabin is located inside the frame and includes a fixed work station, a cargo buffer mechanism, a battery storage and swapping mechanism, a robotic arm, and a vision positioning system.

[0013] The carrying and transfer mechanism is used to transfer the UAV that has landed at the landing position to a fixed work station in the integrated work cabin; the robotic arm is used to perform at least two of the following operations on the UAV in parallel at the fixed work station: unloading, loading, and battery replacement; the visual positioning system is used to provide operation positioning for the robotic arm; and the carrying and transfer mechanism is also used to transfer the UAV to the takeoff position after the operation is completed.

[0014] Preferably, the landing position and the takeoff position are independent platforms arranged horizontally side by side, and both the landing position and the takeoff position are equipped with:

[0015] V-shaped or conical guide grooves formed on the surface of the docking platform;

[0016] An electromagnetic locking mechanism built into the docking platform and corresponding to the guide groove;

[0017] And a liftable interface module located in the central area of ​​the guide groove, the interface module being provided with data communication contacts and power supply contacts.

[0018] Preferably, the load-bearing and transfer mechanism includes:

[0019] A horizontal moving unit includes at least one linear guide rail fixed to the bottom of the integrated work cabin, a moving platform slidably connected to the guide rail, and a first driving unit that drives the moving platform to move along the guide rail.

[0020] A vertical lifting unit, the base of which is fixedly installed on the mobile platform;

[0021] A carrying tray is fixedly installed on top of the vertical lifting unit;

[0022] The carrying tray is driven to move up and down by the vertical lifting unit and moves horizontally with the mobile platform.

[0023] Preferably, the top surface of the carrying tray is provided with a contoured bracket, and the contoured bracket is provided with a retractable positioning pin and an electromagnetic locking module; the first drive unit of the horizontal moving unit is a servo motor and a gear rack mechanism or a synchronous belt mechanism connected thereto; the vertical lifting unit is a cross scissor lift or multiple sets of synchronous electric push rods.

[0024] Preferably, the cargo buffering mechanism is an automated warehouse based on a shuttle and telescopic forks. The automated warehouse includes multi-level racks, a shuttle that moves horizontally on a track in front of the racks, a vertically lifting platform mounted on the shuttle, and telescopic forks mounted on the platform that can reach into the racks to pick up and place standard cargo boxes. The cargo buffering mechanism also includes a fixed transfer platform for the robotic arm to transfer cargo boxes.

[0025] Preferably, the battery storage and swapping mechanism includes a matrix-style multi-compartment drawer charging rack, each independent compartment of which is used to dock and manage a standardized quick-swap battery module.

[0026] Preferably, the robotic arm is a six-degree-of-freedom tandem joint robotic arm, and its end effector is equipped with a tool quick-change device for changing cargo grippers and battery insertion / removal tools.

[0027] Preferably, the end effector of the robotic arm is an integrated duplex structure, which integrates an electric parallel gripper for picking up goods and a battery operating head for inserting and removing batteries.

[0028] Preferably, the battery insertion / removal tool includes a guide pin, an electrical connector, and a rotary jaw locking mechanism; the rotary jaw locking mechanism is controlled by a micro servo motor via a worm gear-driven annular cam.

[0029] Preferably, the integrated work cabin is also equipped with a visual positioning system, which includes a global vision camera fixedly installed on the top of the cabin and a local vision camera integrated into the end of the robotic arm.

[0030] Preferably, the visual positioning system is configured to: perform global scanning and coarse positioning of the fixed work station using the global vision camera, and perform local fine positioning and real-time visual servo control using the local vision camera when the robotic arm approaches the target.

[0031] Preferably, a local controller is also included;

[0032] The local controller is communicatively connected to the load-bearing transfer mechanism, the robotic arm, the cargo buffer mechanism, and the battery storage and swapping mechanism, and is configured to:

[0033] Receive mission instructions and real-time status data of the UAV;

[0034] Based on the task instructions and the real-time status data of the UAV, a sequence of operation instructions is generated to control the robotic arm to perform operations.

[0035] Preferably, the local controller is also configured to: autonomously decide whether to include a battery replacement operation in the sequence of operation instructions based on the real-time battery data of the UAV and a preset safety threshold; and / or to perform real-time compensation on the motion trajectory of the robotic arm based on feedback from the visual positioning system.

[0036] Preferably, the integrated work cabin is also equipped with an openable and closable environmental isolation door; the system also includes a trajectory planning module that is communicatively connected to the load-bearing transfer mechanism, the working robotic arm and the vision positioning system. The trajectory planning module is configured to plan a collision-free motion trajectory for the working robotic arm with the goal of minimizing the total operation time, so that the unloading, loading and battery replacement operations at least partially overlap in time.

[0037] A method for operating a modular unmanned aerial vehicle (UAV) parallel operation system based on any one of the above-mentioned methods includes the following steps:

[0038] The drone landed at the designated landing site;

[0039] The carrying and transfer mechanism transfers the entire UAV to a fixed work station within the integrated work cabin;

[0040] At the fixed work station, the robotic arm, with the assistance of the vision positioning system, performs at least two of the following tasks in parallel on the drone: unloading, loading, and battery replacement;

[0041] The carrying and transfer mechanism moves the completed UAV out of the integrated operation cabin and transports it to the takeoff position;

[0042] The drone takes off from the launch position.

[0043] The modular UAV parallel operation system provided in this application achieves the following beneficial effects through the integrated and modular design of "overall transfer" and "parallel operation":

[0044] 1. Revolutionary improvement in transfer efficiency: By eliminating the time spent on drone self-movement through "overall transfer", and by combining "parallel operation" to overlap the time of multiple core processes, the transfer time of a single transfer is shortened from several minutes to seconds, greatly improving the node throughput capacity.

[0045] 2. High operational accuracy and reliability: The UAV is operated stably in a fixed work position. Combined with high-precision visual positioning and servo control, it can achieve reliable and accurate loading, unloading and battery swapping. The enclosed cabin isolates it from external interference.

[0046] 3. Strong modularity and network adaptability: The system is designed as a fully functional standardized module with a unified interface, which can be flexibly deployed in various logistics network topologies in a plug-and-play manner, facilitating rapid expansion.

[0047] 4. Intelligent resource utilization: The local controller can make optimal decisions based on real-time data and resource status (such as on-demand battery swapping), optimize battery life and machine turnover, and improve overall operational economy. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of the structure of a modular unmanned aerial vehicle (UAV) parallel operation system provided in an embodiment of this application;

[0050] Figure 2 This is a flowchart of an operation method for a modular unmanned aerial vehicle (UAV) parallel operation system provided in an embodiment of this application;

[0051] Figure 3 This is a schematic diagram of the drone reaching the landing position in an embodiment of this application;

[0052] Figure 4 This is a schematic diagram of a robotic arm performing a battery replacement operation on a drone in an embodiment of this application;

[0053] Figure 5 This is a schematic diagram of the UAV reaching the takeoff position in an embodiment of this application;

[0054] Among them, 101, rack; 102, landing position; 103, takeoff position; 104, position sensor; 105, guide rail; 106, horizontal movement unit; 107, carrying pallet; 108, integrated work cabin; 109, door; 110, working robotic arm; 111, arriving cargo buffer rack; 112, waiting cargo buffer rack; 113, fully charged battery compartment; 114, matrix-type multi-compartment drawer charging rack; 115, visual positioning system; 116, electromagnetic locking mechanism. Detailed Implementation

[0055] To make the technical solution of this application clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The terms "first," "second," etc., in the claims and specification of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate. This is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of units is not necessarily limited to those units, but may include other units not explicitly listed or inherent to these processes, methods, products, or apparatuses.

[0056] Example 1

[0057] This embodiment provides a modular unmanned aerial vehicle (UAV) parallel operation system, including:

[0058] Rack 101;

[0059] The drone docking platform is set on the rack 101 and includes a physically separated drone landing position 102 and drone take-off position 103.

[0060] The load-bearing and transfer mechanism is located inside the frame 101;

[0061] An integrated work cabin 108 is located inside the frame 101 and includes a fixed work station, a cargo buffer mechanism, a battery storage and swapping mechanism, a robotic arm 110, and a vision positioning system 115.

[0062] The carrier-transfer mechanism is used to transfer the UAV that landed at landing position 102 to a fixed work station in the integrated work cabin 108; the robotic arm 110 is used to perform at least two of the following operations on the UAV in parallel at the fixed work station: unloading, loading and unloading and battery replacement; the visual positioning system 115 is used to provide operation positioning for the robotic arm 110; the carrier-transfer mechanism is also used to transfer the UAV to the take-off position 103 after the operation is completed.

[0063] See Figure 1 The modular parallel UAV system of this embodiment is mainly a protective frame 101. The frame 101 has a rectangular box structure, and the internal space is divided into upper and lower layers. The front of the upper layer is the UAV docking area, and the lower layer is the integrated operation cabin 108 area.

[0064] A docking platform is fixedly mounted on the top steel frame of the rack 101. A key feature of this platform is the presence of two horizontally parallel, physically separated, and structurally independent dedicated landing positions 102 and 103. These two positions are arranged side-by-side on the top of the rack 101 with a gap between them. This design achieves complete spatial decoupling of the landing and takeoff processes, allowing one UAV to perform pre-takeoff self-checks and take off from takeoff position 103 while another UAV performs landing and locking from landing position 102. The two operations do not interfere with each other in terms of physical space and operational sequence, significantly improving the throughput per unit time and fundamentally eliminating the risk of process conflicts.

[0065] In one specific embodiment, both the landing position 102 and the takeoff position 103 are provided with:

[0066] V-shaped or conical guide grooves formed on the surface of the docking platform;

[0067] An electromagnetic locking mechanism 116 is built into the docking platform and corresponds to the guide groove;

[0068] And a liftable interface module located in the central area of ​​the guide groove, the interface module is equipped with data communication contacts and power supply contacts.

[0069] To achieve rapid, accurate, and reliable locking of the drone, the platform structure of the dedicated landing position 102 and takeoff position 103 incorporates a three-level collaborative scheme:

[0070] 1) Primary Positioning (Guidance): A ring-shaped LED guide light and multiple photoelectric sensors are embedded on the platform surface. When the UAV lands, its flight control system communicates with the light group to achieve initial photoelectric and electromagnetic guidance and centering.

[0071] 2) Secondary positioning (mechanical precision positioning): V-shaped or conical guide grooves are milled or cast on the platform surface. These guide grooves match the conical guide pins at the bottom of the UAV landing gear (or vice versa, the platform has pins and the UAV landing gear has grooves). After the UAV lands, under its own weight or slight driving force, the guide pins slide along the inclined surface into the bottom of the guide groove, achieving mechanical self-centering from centimeter to millimeter level.

[0072] 3) Three-level locking (rigid fixation): An electromagnetic locking mechanism 116 is connected to the lower or side end of each guide groove on the internal structural components of the platform. When the UAV positioning sensor 104 (such as a miniature limit switch) is triggered, the electromagnet is energized, driving a self-locking claw or a rising wedge pin to lock the UAV's landing gear pin from the side or press against the locking hole from below, achieving secure positioning. The locking status is fed back to the local controller in real time by a Hall sensor.

[0073] At the center of each platform is a liftable interface module. This module is driven by a small servo motor or cylinder and can be lifted vertically.

[0074] This module integrates data communication contacts and power supply contacts. The data communication contacts are a set of gold-plated elastic probes (such as Pogo Pins) used to connect to the corresponding contact array on the UAV's underside, establishing a wired high-speed data channel. This is used to quickly download the next mission command after locking, upload flight logs, synchronize status information, and perform a more reliable pre-flight command verification than wireless communication.

[0075] The power supply contacts are another set of elastic probes with higher power capacity, used to provide DC power when the drone battery is severely depleted or when deep system maintenance is required, ensuring that the drone's core systems do not lose power.

[0076] After the drone is mechanically locked, the local controller drives the interface module to rise and dock. The docking is completed and confirmed by the contact impedance detection circuit and the alignment sensor 104 on the module.

[0077] The carrying and transfer mechanism is located in the integrated operation cabin 108 below the docking platform, and its mechanical mounting base is fixed to the bottom frame of the operation cabin. Its core task is to perform precise reciprocating horizontal movement between the area below the landing opening and the area below the takeoff opening to complete the "pick-up-transport-placement" of the entire UAV.

[0078] In one specific embodiment, the load-bearing and transfer mechanism includes:

[0079] The horizontal moving unit 106 includes at least one linear guide rail 105 fixed to the bottom of the integrated work cabin 108, a moving platform slidably connected to the guide rail 105, and a first driving unit that drives the moving platform to move along the guide rail 105.

[0080] The vertical lifting unit has its base fixedly mounted on the mobile platform;

[0081] The carrying pallet 107 is fixedly installed on the top of the vertical lifting unit;

[0082] The carrying tray 107 is driven to move up and down by the vertical lifting unit and moves horizontally with the mobile platform.

[0083] Specifically, the horizontal movement unit 106 includes two parallel high-precision linear guide rails 105 laid and fixed to the bottom frame 101 of the work cabin. A rigid moving platform (base) is connected to these two guide rails 105 via a linear slider. The first drive unit is a servo motor, which drives a gear in a precision gear and rack mechanism to rotate via a coupling. The gear meshes with a rack fixed to the frame 101, thereby converting the rotational motion into linear motion of the moving platform. Alternatively, a servo motor can be used to drive a synchronous pulley, which pulls the moving platform via a synchronous belt. The servo motor is equipped with an absolute encoder, which, together with the limit switches at the ends of the guide rails 105, achieves closed-loop control of the platform's horizontal position.

[0084] The base of the vertical lifting unit is bolted to the mobile platform and moves horizontally with the platform. This unit can be a cross-scissor lift: composed of multiple sets of connecting rods hinged by pins, with rollers at both ends rolling within guide rails 105 at the bottom of the mobile platform and the upper support tray 107. A lifting servo motor drives the central connecting shaft of the scissor lift via a worm gear reducer or ball screw, thereby controlling the entire frame's expansion (ascent) and contraction (descent). An alternative solution is to use multiple sets (e.g., four) of synchronous electric push rods, with their cylinders fixed to the mobile platform. The tops of the push rods are connected to the support tray 107, and a controller ensures synchronous lifting.

[0085] The support tray 107 is a rigid, lightweight component (such as an aluminum alloy frame) that is fixedly mounted on top of the vertical lifting unit via a mounting plate. The top surface of the support tray 107 is machined or fitted with a contoured bracket that precisely matches the profile of the target UAV's landing gear. This contoured bracket integrates a retractable tapered locating pin and an electromagnetic locking module.

[0086] The retractable conical positioning pin, driven by a miniature cylinder or motor, can pop out upwards. It is used to guide and mechanically center the drone by first inserting into the corresponding guide hole at the bottom of the drone's landing gear when the tray is raised to lift the drone.

[0087] The electromagnetic locking module uses an electromagnet or an electromagnet-driven mechanical latch. Once the tray rises to a position where it is fully in contact with the bottom surface of the drone's landing gear (confirmed by a contact sensor), the electromagnetic locking module is energized, generating a strong magnetic force or driving the latch to firmly attach or lock the drone's landing gear to the tray, forming a rigid connection. The locking force is fed back by a sensor.

[0088] After the drone is transported into and secured to a fixed work station within the integrated work cabin 108 by the carrying and transfer mechanism, the automated system integrated within the cabin will perform cargo loading and unloading and battery replacement operations in parallel. This is the key to compressing transit time to minutes or even seconds.

[0089] In one specific embodiment, the integrated work cabin 108 is also equipped with a robotic arm 110, a cargo buffer mechanism, and a battery storage and replacement mechanism; the cargo buffer mechanism and the battery storage and replacement mechanism are arranged adjacent to each other and are both located within the working range of the robotic arm 110.

[0090] The cargo buffer mechanism is used to temporarily store arriving and pending shipment goods. Preferably, this mechanism is an automated warehouse based on a shuttle and telescopic forks. It includes:

[0091] Multi-layer racks: constructed from aluminum alloy profiles and fixed to the two side walls inside the work compartment, with one side serving as the arrival goods buffer rack 111 and the other side as the waiting goods buffer rack 112.

[0092] Standardized cargo box: with bottom guide grooves and side RFID tags.

[0093] Shuttle: Moves horizontally on a ground track in front of the shelf, driven by a servo motor.

[0094] Telescopic forks: Installed on the lifting platform of the shuttle, they are driven by a motor and can extend and retract in both directions, allowing them to be precisely inserted into the grooves at the bottom of the cargo box.

[0095] The working principle of cargo storage is as follows: the robotic arm 110 places the unloaded cargo box onto the fixed transfer platform, the shuttle moves to the transfer platform, the forks extend to lift the cargo box, and it is stored in the designated empty space of the cargo buffer rack 111 according to the instructions.

[0096] The working principle of cargo retrieval is as follows: the control system commands a cargo box to be shipped from a certain location in the cargo buffer rack 112. The shuttle moves to the location, the forks extend to retrieve the box, and then it is transported to the transfer table for the robotic arm 110 to grab.

[0097] It's important to note that the fixed work station is not an independent platform, but rather defined by the precise stopping position of the carrier pallet 107 within the integrated work cabin 108. This position is defined by mechanical limit blocks on the ground and four sets of high-precision photoelectric sensors. When the carrier pallet 107 moves to this position, the photoelectric sensors are triggered, confirming that it has reached the "fixed work station." Simultaneously, the carrier pallet 107 descends to a preset "working height," which positions the UAV cargo door 109 and battery compartment interface precisely at the optimal operating height of the robotic arm 110.

[0098] A battery storage and swapping mechanism is used for rapid standardization of quick-swap battery modules for drones. Preferably, a matrix-style multi-compartment drawer charging rack 114 is adopted. It includes:

[0099] The main body of the charging rack is fixed inside the work compartment on the opposite side or below the cargo buffer mechanism, and consists of a matrix of multiple independent drawer compartments.

[0100] Charging drawers: Each drawer is an independent unit with built-in flexible power contacts, communication contacts (for communicating with the battery BMS), locking mechanisms (such as electromagnetic locks), and cooling fans.

[0101] Battery Management System (BMS): Integrated into the local controller, it monitors the real-time status (SOC, SOH, temperature) of the batteries in each drawer.

[0102] The battery storage and swapping mechanism also includes a fully charged battery compartment 113 located on the opposite side of the matrix-type multi-compartment drawer charging rack 114, for storing fully charged batteries.

[0103] The robotic arm 110 is the core component for performing specific operations. It is preferably a six-DOF tandem articulated robotic arm. Its base is fixedly mounted on a support beam at the center of the top of the work cabin via a flange-mounted inverted mounting method. All joints are driven by an integrated servo joint module, featuring high repeatability (e.g., ±0.1mm) and a sufficient working radius to cover fixed workstations, handover platforms, and battery compartments.

[0104] Furthermore, its end flange is equipped with a male connector for a tool quick-change device, which can automatically connect to female connectors of different tool heads. The tool head includes at least a cargo gripper and a battery insertion / removal tool head. The cargo gripper is a servo-driven, parallel two-finger gripper with replaceable fingers and an internal pressure sensor. The battery insertion / removal tool head has a latch, electrical connector, and guide mechanism that matches the battery module.

[0105] In a more integrated design, the end effector employs a single duplex structure, internally integrating two independent drive systems in parallel. The upper part houses the drive motor and transmission mechanism for the electric parallel grippers; the lower part contains the battery insertion / removal head, which includes a linear telescopic motor for overall insertion / removal and a rotary gripper drive motor for locking / releasing the battery. Physically, both mechanisms are always in place. During operation, the controller electronically switches between enabling one system and locking the other, achieving "function switching" without physical replacement time.

[0106] In one specific embodiment, the integrated work cabin 108 is also equipped with a visual positioning system 115, which includes a global vision camera fixed inside the cabin and a local vision camera installed at the end of the work robot arm 110.

[0107] The visual positioning system 115 provides high-precision "visual guidance" for the robotic arm 110 to ensure operational accuracy. The system adopts a composite mode of global guidance + local fine-tuning, including a global vision camera fixed to the top of the cabin and a local vision camera installed at the end of the robotic arm 110.

[0108] Specifically, the global vision camera is a high-resolution binocular stereo camera, fixedly installed at the center of the top of the work cabin, used to quickly perform three-dimensional point cloud scanning after the drone enters, and obtain the rough pose of key components such as the drone, cargo door 109, and battery compartment.

[0109] This local vision camera is a high frame rate 2D industrial camera that is directly integrated and mounted on the end effector (or tool head) of the robotic arm for millimeter-level real-time positioning and visual servo control when approaching a target.

[0110] The processing unit of the vision positioning system 115 is a dedicated industrial vision controller that runs vision algorithms and communicates with the robotic arm 110 in real time. Through pre-completed "hand-eye calibration", the camera-recognized coordinates can be converted into motion commands in the robotic arm's base coordinate system or the tool coordinate system.

[0111] Its working principle is as follows: After the carrying and transfer mechanism transports the entire UAV and fixes it in the fixed working position in the center of the working cabin, the binocular stereo vision cameras installed diagonally on the top of the cabin are immediately triggered. These cameras quickly scan the entire working area in three dimensions, and combined with a pre-trained deep learning model, identify the AprilTag visual mark on the UAV cargo door 109, the specific geometric feature pattern of the battery compartment interface, and the markings of the surrounding cargo buffer racks in one go. The system uses precise hand-eye calibration data (which is determined once during installation, establishing a fixed transformation relationship between the camera coordinate system, the robotic arm base coordinate system, and the physical space coordinate system) to calculate in real time the three-dimensional spatial coordinates and attitude angles of these key feature points in the robotic arm base coordinate system, thereby providing the robotic arm controller with a "global map" covering the entire working area and the initial action target.

[0112] Based on this global positioning information, the robotic arm begins to move. For relatively coarse cargo gripping and placement operations, the robotic arm mainly relies on the coordinates provided by global vision for "visual positioning-open-loop motion." For example, when the robotic arm's end effector switches to a cargo gripper and moves to the vicinity of cargo door 109, a high-frame-rate 2D industrial camera integrated on the end effector is activated to perform final fine-tuning confirmation of the cargo door 109 latch, and then performs the opening and gripping actions. When the robotic arm moves the cargo to the buffer rack, the end effector continuously identifies the positioning marks on the rack slots to ensure accurate placement of the cargo. However, for battery insertion and removal operations that require extremely high alignment accuracy, the vision system switches to a more refined "real-time vision servo" mode. When the robotic arm's end effector is equipped with a battery insertion and removal tool and moves to the vicinity of the battery compartment, the end effector camera begins to continuously capture images of the battery interface at a high frame rate, and the vision processor calculates in real time the sub-millimeter positional deviation and minute angular offset between the guide pin on the insertion head, the electrical connector, and the battery interface. These deviation data are fed back to the robotic arm controller in milliseconds to form a closed-loop control, dynamically adjusting the motion trajectory of the robotic arm's end effector until the deviation approaches zero, thereby ensuring perfect alignment between the plug and the battery interface, and then performing smooth plugging, unplugging and locking actions.

[0113] Throughout the process, global vision is responsible for "guiding the direction" and providing coarse positioning and spatial perception; local vision is responsible for "completing the last millimeter" of precise positioning and dynamic correction. The two work together to enable the robotic arm to quickly reach the work area and complete various operations on the drone with extremely high reliability and precision. This is the key technical guarantee for achieving fully automatic and highly efficient parallel operation.

[0114] In one embodiment, the system further includes a local controller;

[0115] The local controller is communicatively connected to the load-bearing transfer mechanism, the robotic arm 110, the cargo buffer mechanism, and the battery storage and swapping mechanism, and is configured as follows:

[0116] Receive mission instructions and real-time status data of the UAV;

[0117] Based on the task instructions and the real-time status data of the UAV, a sequence of operation instructions is generated to control the robotic arm 110 to perform operations.

[0118] Specifically, after the drone enters the fixed work station, the local controller follows a strict state machine logic to drive the entire operation process:

[0119] First, the local controller performs task parsing and precise modeling. Upon receiving a work instruction from the upper-level scheduling system (such as unloading cargo A, loading cargo B, and replacing the battery), the local controller decomposes it into an indivisible sequence of atomic actions, such as "move to cargo A," "grab," "move to buffer rack X," and "release." Simultaneously, it loads high-precision 3D geometric models of the work cabin, drone, rack, battery compartment, and robotic arm itself, serving as the basis for subsequent collision detection and spatial planning.

[0120] Next, dependency analysis and parallelism identification are performed. Based on the logical sequence of atomic actions (e.g., a task-dependent directed acyclic graph must be constructed before movement), a task dependency is defined. Subsequently, considering the constraints of resources such as the robotic arm and multi-functional end effector, a graph search-based list scheduling algorithm is used to automatically analyze which actions are logically independent and do not occupy the same critical resources or space, thereby identifying action pairs that can be executed in parallel and their potential time windows. For example, on the return path of the robotic arm transporting goods A to the shelf, its idle phase can be planned to simultaneously move towards the location of goods B.

[0121] Then, the trajectory planning module uses a spatiotemporal trajectory joint optimization algorithm to plan a joint optimal trajectory that allows the robotic arm (or multiple robotic arms / multiple tool heads) to move smoothly and overlaps highly in time (for example, when the robotic arm is transporting old goods back to the warehouse, its path has been planned to pass near the new goods pickup point; or in a single-arm system, the gripper operation and the preparatory movement of the battery tool head are interspersed in time by optimizing the trajectory).

[0122] In one specific embodiment, the local controller is also configured to: autonomously decide whether to include a battery replacement operation in the sequence of operation instructions based on the real-time battery data of the UAV and a preset safety threshold; and / or to perform real-time compensation on the motion trajectory of the robotic arm 110 based on feedback from the visual positioning system 115.

[0123] The local controller's decision on whether to replace the drone battery is not a simple threshold judgment, but a dynamic process that integrates multi-source information and performs intelligent arbitration based on preset rules and real-time status. This mechanism ensures an optimal balance between the timeliness, safety, and economy of battery swapping decisions.

[0124] Specifically, before the drone arrives, the local controller receives a cloud-based instruction package containing the estimated arrival time, a list of recommended services, and a suggested battery swapping threshold. Meanwhile, after the drone lands and physically docks, the controller continuously acquires millisecond-level real-time data from its battery management system (BMS) via wired and wireless channels, including precise state of charge (SOC), state of health (SOH), voltage, temperature, and fault codes.

[0125] Based on the above information, the controller executes a hierarchical arbitration logic. First, a spatiotemporal consistency check is performed, comparing the drone's actual arrival time and battery status with the cloud-based prediction. If the deviation is significant, independent judgment is made based on real-time data. Next, the highest-priority safety check is executed: regardless of other conditions, if the real-time SOC falls below the absolute safety threshold (e.g., 10%), or the BMS reports a serious fault, the controller will unconditionally force a battery replacement operation into the work sequence, potentially simplifying other non-urgent tasks. If the safety red line is not triggered, the controller will further assess the feasibility of local resources and workflow, such as checking if sufficient fully charged batteries are available and whether the currently planned robotic arm operation time window allows for the battery replacement step. Finally, before the robotic arm performs the operation, the vision positioning system 115 performs a final physical verification of the battery compartment; if interface damage or abnormal battery ID is found, the battery replacement will be aborted.

[0126] Suppose a drone arrives from "Cell A". The cloud predicts its State of Charge (SOC) is 22% and suggests replacement, but the local controller reads an actual SOC of 28%. Verification shows this SOC is above the safety threshold and the battery is in normal condition. Simultaneously, the controller detects a shortage of fully charged batteries in the local cell and a heavy workload ahead. Based on this, the controller might make an optimized decision not to replace the battery, but instead schedule unloading and loading in the job sequence, feeding this decision and its rationale back to the cloud system. Through this collaborative "cloud prediction, local decision-making" model, the system optimizes resource utilization and operational efficiency while ensuring flight safety.

[0127] In one embodiment, the system further includes a trajectory planning module that is communicatively connected to the load-bearing transfer mechanism, the robotic arm 110, and the vision positioning system 115. The trajectory planning module is configured to plan a collision-free motion trajectory for the robotic arm 110 with the goal of minimizing the total operation time, so that the unloading, loading, and battery replacement operations at least partially overlap in time.

[0128] The trajectory planning module works as follows: With the shortest total operation time as the core optimization objective, the position, velocity, and other parameters of each joint of the robotic arm 110 at discrete time steps are set as variables. Kinematic limits, dynamic limits, temporal requirements of actions, and collision-free requirements between robotic arms and between the robotic arm and the environment are collectively constructed as constraints, forming a large-scale nonlinear optimization problem. Using numerical optimization solvers such as sequential quadratic programming, the system calculates multiple optimal joint motion trajectories that are synchronized in time and completely collision-free in space, enabling multiple actions to be executed with high overlap on the time axis.

[0129] During the trajectory optimization process, the visual positioning system 115 continuously monitors the real-time pose of targets such as drones and cargo gripping points. Once the actual pose detected deviates from the pose assumed during planning beyond a safety threshold, the system immediately triggers millisecond-level local replanning: taking the current actual state of the robotic arm as the new starting point and the original task objective as the endpoint, it quickly re-solves a new trajectory that satisfies all constraints in a very short time, thereby achieving closed-loop compensation control for sudden deviations and ensuring that the operation is always accurate and reliable.

[0130] The system provided in this embodiment is a fully functional, standard-interface independent "node module". Its standardized physical interface and control protocol allow it to be used as a basic operating unit without deep customization and flexibly integrated into various logistics network architectures. For example, it can be used as an intermediate station in a relay network, a regional hub in a distribution network, or a dedicated relay node in a more complex cellular micro-hub system, which greatly facilitates the rapid construction and expansion of large-scale, standardized networks.

[0131] Example 2

[0132] like Figure 2 As shown, the operation method of the modular UAV parallel operation system provided in Example 1 includes the following steps:

[0133] S110, the drone landed at landing position 102;

[0134] S120, the carrying and transfer mechanism transfers the UAV as a whole to a fixed work station inside the integrated work cabin 108;

[0135] S130. At a fixed work station, the robotic arm 110, with the assistance of the visual positioning system 115, performs at least two of the following tasks in parallel on the drone: unloading, loading, and battery replacement.

[0136] S140, the carrying and transfer mechanism moves the completed UAV out of the integrated operation cabin 108 and transports it to the takeoff position 103;

[0137] S150, the drone took off from takeoff position 103.

[0138] The system's fully automated operation method, through a highly coordinated hardware system and intelligent control logic, enables the rapid turnaround of the drone from landing to takeoff, especially the parallel operation of the 110 robotic arms.

[0139] The complete process is as follows: Figure 3As shown, the UAV accurately lands on the physically separated landing position 102 based on guidance signals emitted from the drone's pod (such as RTK differential GPS or ultra-wideband positioning signals). Upon landing, the UAV's landing gear engages with the V-shaped or conical guide grooves on the platform surface, completing initial positioning through mechanical self-homing. The platform's built-in pressure sensors and photoelectric sensor array jointly determine that the UAV has landed stably, triggering the electromagnetic locking mechanism 116. The locking pin pops out and engages with the landing gear's dedicated locking hole, achieving a secure mechanical lock. Simultaneously, the retractable interface module in the center of the platform rises, and its data communication and power supply contacts connect to the UAV's underside interface, establishing a high-speed wired connection for quickly downloading flight mission commands, uploading logs, and reading real-time data across all dimensions, including battery charge status and health status.

[0140] After locking and docking are completed, the environmental isolation door 109 on the side of the integrated operation cabin 108 facing the docking platform automatically opens. The load-bearing transfer mechanism located inside the frame 101 then activates, and its horizontal movement unit 106 drives the moving platform along the precision linear guide rail 105 to directly below the landing position 102. Upon reaching the predetermined coordinates, the vertical lifting unit activates, and the load-bearing tray 107 on top rises smoothly. The conical positioning pins on the tray first insert into the corresponding guide holes of the UAV landing gear to achieve guidance and precise positioning; the tray continues to rise until it fully supports the UAV fuselage, and the electromagnetic locking module on it is energized to further attract and lock the landing gear, thereby completing the transfer of the UAV from the docking platform to the load-bearing transfer mechanism. Subsequently, the load-bearing transfer mechanism carries the locked UAV horizontally, passes through the opened door 109, enters the integrated operation cabin 108, and finally precisely docks at the fixed work position defined by ground sensors and mechanical limits. At this time, the environmental isolation door 109 closes, providing a closed environment for internal operations free from external interference.

[0141] At the fixed workstation, the core of the system—the parallel operation phase—immediately begins. This phase is coordinated by the local controller, with the robotic arm 110, the vision positioning system 115, the cargo buffer mechanism, and the battery storage and replacement mechanism working together. First, the global vision camera fixed on the top of the cabin performs a rapid 3D scan of the stationary UAV, identifying key features such as the cargo door 109 and the battery compartment interface, and sending the rough coordinates to the robotic arm controller. Subsequently, the robotic arm 110 begins its operation. Its parallelism is reflected in the optimized overlap of the robotic arm's motion paths and the collaborative scheduling of multiple tasks. For example, the end effector of the robotic arm first switches to a cargo gripper, and under the guidance of the local vision camera integrated at its end, moves to the UAV's cargo compartment to perform an unloading operation: opening the door 109, gripping the arriving cargo box, and transferring it to the fixed transfer platform next to the cargo buffer mechanism. Almost simultaneously, the shuttle system of the cargo buffer mechanism receives instructions and begins to store the cargo boxes on the transfer platform into designated locations on the automated storage and retrieval system, while simultaneously retrieving cargo boxes from another location and placing them on the transfer platform. The robotic arm can grab the cargo to be shipped and load it into the drone's cargo bay on its return path without waiting for the storage action to complete. Simultaneously, based on autonomous decisions made by the local controller using real-time drone battery data (such as a SOC below 20%), the robotic arm's end effector can seamlessly switch to battery plug-and-play mode via a tool changer or by utilizing its "integrated duplex" structure. Figure 4 As shown, under the real-time servo control of a local vision camera, the robotic arm precisely locates the drone's battery compartment, performs a complete battery swapping operation: removing the depleted battery, moving it to an empty charging drawer in the battery swapping mechanism, and then grabbing a fully charged battery and installing it back into the drone. Through optimized calculations by the trajectory planning module, discrete actions such as unloading, loading, and battery swapping are integrated into a continuous motion trajectory with the shortest total time and no spatial collisions. This results in a high degree of overlap in the time windows of multiple core processes, rather than traditional sequential execution.

[0142] After the parallel operations within the cabin are completed, the environmental isolation hatch 109 reopens. The carrying and transfer mechanism moves the entire UAV, with its cargo and battery replaced, from the fixed work area and transports it along a horizontal path to directly below the takeoff position 103. Subsequently, the carrying pallet 107 descends, smoothly releasing the UAV onto the surface of the takeoff position 103 platform. Figure 5 As shown, the electromagnetic locking mechanism 116 of takeoff position 103 briefly locks the UAV to maintain stability, while its interface module rises to perform final data verification and command injection before takeoff. After all takeoff conditions are confirmed to be correct, the locking mechanism releases, and the UAV takes off autonomously from takeoff position 103 to continue the next leg of its journey.

[0143] Throughout the process, the local controller continuously monitors the status of each device and maintains communication with the upper-level logistics scheduling system, providing real-time feedback on operation progress and system status, thereby ensuring that the modular node can be seamlessly integrated into a larger logistics network system.

[0144] This embodiment avoids the repeated positioning and start-up / stopping of the UAV itself by transferring the entire machine. By overlapping the time of multiple core processes through "parallel operation", the total time of a single transfer operation can be shortened from several minutes in the traditional serial mode to less than 60 seconds, which greatly improves the node throughput capacity.

[0145] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A modular unmanned aerial vehicle (UAV) parallel operation system, characterized in that, include: frame; The drone docking platform is mounted on the frame and includes physically separate drone landing positions and drone takeoff positions; A load-bearing and transfer mechanism is disposed inside the frame; An integrated work cabin is located inside the frame and includes a fixed work station, a cargo buffer mechanism, a battery storage and swapping mechanism, a robotic arm, and a vision positioning system. The carrying and transfer mechanism is used to transfer the UAV that has landed at the landing position to a fixed work station in the integrated work cabin; the robotic arm is used to perform at least two of the following operations on the UAV in parallel at the fixed work station: unloading, loading, and battery replacement; the visual positioning system is used to provide operation positioning for the robotic arm; and the carrying and transfer mechanism is also used to transfer the UAV to the takeoff position after the operation is completed.

2. The modular unmanned aerial vehicle (UAV) parallel operation system according to claim 1, characterized in that, The landing position and takeoff position are independent platforms arranged horizontally side by side. Both the landing position and takeoff position are equipped with: V-shaped or conical guide grooves formed on the surface of the docking platform; An electromagnetic locking mechanism built into the docking platform and corresponding to the guide groove; And a liftable interface module located in the central area of ​​the guide groove, the interface module being provided with data communication contacts and power supply contacts.

3. The modular unmanned aerial vehicle (UAV) parallel operation system according to claim 1, characterized in that, The load-bearing and transfer mechanism includes: A horizontal moving unit includes at least one linear guide rail fixed to the bottom of the integrated work cabin, a moving platform slidably connected to the guide rail, and a first driving unit that drives the moving platform to move along the guide rail. A vertical lifting unit, the base of which is fixedly mounted on the mobile platform; A carrying tray is fixedly installed on top of the vertical lifting unit; The carrying tray is driven to move up and down by the vertical lifting unit and moves horizontally with the mobile platform.

4. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 3, characterized in that, The top surface of the carrying tray is provided with a contour bracket, and the contour bracket is provided with a retractable positioning pin and an electromagnetic locking module; the first drive unit of the horizontal moving unit is a servo motor and a gear rack mechanism or synchronous belt mechanism connected thereto; the vertical lifting unit is a cross scissor lift or multiple sets of synchronous electric push rods.

5. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 1, characterized in that, The cargo buffering mechanism is an automated warehouse based on shuttle cars and telescopic forks. The automated warehouse includes multi-level racks, a shuttle car that moves horizontally on a track in front of the racks, a vertically lifting platform mounted on the shuttle car, and telescopic forks mounted on the platform that can reach into the racks to pick up and place standard cargo boxes. The cargo buffering mechanism also includes a fixed transfer platform for the robotic arm to transfer cargo boxes.

6. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 1, characterized in that, The battery storage and swapping mechanism includes a matrix-style multi-compartment drawer charging rack, each independent compartment of which is used to connect to and manage a standardized quick-swap battery module.

7. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 1, characterized in that, The robotic arm is a six-degree-of-freedom tandem joint robotic arm, and its end effector is equipped with a tool quick-change device, which is used to change the cargo gripper and battery insertion / removal tool.

8. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 1, characterized in that, The end effector of the robotic arm is an integrated duplex structure, which integrates an electric parallel gripper for picking up goods and a battery operating head for inserting and removing batteries.

9. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 7, characterized in that, The battery insertion and removal tool includes a guide pin, an electrical connector, and a rotating jaw locking mechanism; the rotating jaw locking mechanism is controlled by a micro servo motor through a worm gear-driven annular cam.

10. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 1, characterized in that, The integrated work cabin is also equipped with a visual positioning system, which includes a global vision camera fixedly installed on the top of the cabin and a local vision camera integrated into the end of the robotic arm.

11. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 10, characterized in that, The visual positioning system is configured to: perform global scanning and coarse positioning of the fixed work station using the global vision camera, and perform local fine positioning and real-time visual servo control when the robotic arm approaches the target using the local vision camera.

12. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 1, characterized in that, It also includes the local controller; The local controller is communicatively connected to the load-bearing transfer mechanism, the robotic arm, the cargo buffer mechanism, and the battery storage and swapping mechanism, and is configured to: Receive mission instructions and real-time status data of the UAV; Based on the task instructions and the real-time status data of the UAV, a sequence of operation instructions is generated to control the robotic arm to perform operations.

13. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 12, characterized in that, The local controller is also configured to: autonomously decide whether to include a battery replacement operation in the sequence of operation instructions based on the real-time battery data of the UAV and a preset safety threshold; and / or to perform real-time compensation on the motion trajectory of the robotic arm based on feedback from the visual positioning system.

14. A modular unmanned aerial vehicle (UAV) parallel operation system according to claim 1, characterized in that, The integrated work cabin is also equipped with an openable and closable environmental isolation door; the system also includes a trajectory planning module that is communicatively connected to the load-bearing transfer mechanism, the working robotic arm and the vision positioning system. The trajectory planning module is configured to plan a collision-free motion trajectory for the working robotic arm with the goal of minimizing the total operation time, so that the unloading, loading and battery replacement operations at least partially overlap in time.

15. A method for operating a modular unmanned aerial vehicle (UAV) parallel operation system according to any one of claims 1-14, characterized in that, Includes the following steps: The drone landed at the designated landing site; The carrying and transfer mechanism transfers the entire UAV to a fixed workstation within the integrated work cabin; At the fixed work station, the robotic arm, with the assistance of the vision positioning system, performs at least two of the following tasks in parallel on the drone: unloading, loading, and battery replacement; The carrying and transfer mechanism moves the completed UAV out of the integrated operation cabin and transports it to the takeoff position; The drone takes off from the launch position.

Images