Unmanned aerial vehicle docking method and system, electronic device

Abstract

The application provides a UAV docking method and system, and an electronic device, and relates to the technical field of UAV docking. Based on the motion state of a target UAV and the pose state of a docking device, an initial guiding path from the current position of the target UAV to the target position of the docking device is generated. Based on the maintained global space-time occupation map, conflict prediction is performed on the initial guiding path, and a cooperative guiding instruction is generated. The conflict includes the intersection of the path of the target UAV with other moving entities on the initial guiding path. The motion of the target UAV is controlled according to the cooperative guiding instruction until the power receiving device of the target UAV and the docking device complete docking. The problem of decision oscillation caused by real-time micro-variation of the state of multiple devices in the bottleneck area is solved. Uncertain distributed conflicts are resolved into certain static scheduling, and the certainty and overall throughput efficiency of system operation in a high-frequency and high-density ground transfer scene are ensured.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and more specifically, to a UAV docking method, system, and electronic device. Background Technology

[0002] With the rapid development of smart logistics and high-end unmanned equipment, fixed-wing UAVs, with their significant advantages of long range, high speed, and large payload, have shown enormous application potential in regionalized, high-frequency freight scenarios. To support their large-scale, commercial operation, achieving a high degree of automation and efficiency in ground support processes (including transfer, docking, and takeoff preparation) has become a key technological bottleneck that the industry urgently needs to overcome. Currently, a series of breakthroughs have been achieved in related technological fields: for example, electromagnetic catapult technology, due to its advantages of smooth acceleration, high efficiency, and good repeatability, is considered a cutting-edge direction for achieving short-distance or zero-length takeoff of UAVs; at the same time, extensive research has been conducted on automatic guidance, high-precision path planning, and conflict avoidance algorithms for unmanned vehicles (such as automated guided vehicles, AGVs).

[0003] However, current technological development exhibits a significant characteristic of "single-point optimization and disconnected processes." While individual modules such as electromagnetic catapults, automated guided vehicles, and path planning algorithms are relatively mature, integrating them into a unified, intelligent ground support system capable of adapting to high-frequency, high-density operations still faces severe challenges. When current technological solutions are applied to actual operational scenarios, the following core shortcomings are exposed: Existing decentralized and independent path planning and conflict resolution methods are primarily designed for open or low-density environments and are ill-suited for scenarios with limited space, high path coupling, and dense traffic, such as ground transfer areas (e.g., fixed taxiways, intersections, docking areas). When multiple automated guided vehicles (AGVs) dynamically converge near bottleneck areas like intersections and docking waiting areas, conflict reduction strategies based on distributed decision-making according to instantaneous states (e.g., distance, priority) are highly susceptible to repeated priority reversals due to sensor noise, communication delays, or minor real-time fluctuations in state. This can induce decision-making oscillations among multiple vehicles, causing the vehicle group to become hesitant, frequently start and stop, or even deadlocked in local areas, severely disrupting the overall system's operational determinism and process continuity. Decision oscillations not only directly reduce the throughput efficiency of ground transfers, failing to meet the cycle time requirements of high-frequency freight transport, but the resulting uncertainty and delays can also propagate upstream, disrupting the precise timing of subsequent high-precision docking and catapult takeoff, significantly reducing the reliability and efficiency of the entire support process.

[0004] Therefore, the industry urgently needs a solution that can deeply integrate automated transfer, multi-vehicle collaboration, high-precision docking, and catapult launch process control from a system-level perspective. This solution must overcome the inherent limitations of existing decentralized decision-making in intensive ground operations. By introducing a collaborative control mechanism with a global perspective and deterministic decision-making capabilities, it can fundamentally eliminate decision oscillations and ensure smooth, efficient, and reliable execution of the entire process from drone unloading, ground transfer, docking, to catapult launch. This will build a solid critical infrastructure for the large-scale operation of ton-class fixed-wing cargo drones. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a drone docking method, system, and electronic device to improve the above-mentioned problems existing in the prior art.

[0006] In a first aspect, embodiments of this application provide a method for docking unmanned aerial vehicles (UAVs). The method includes: generating an initial guidance path from the current position of the target UAV to the target position of the docking device based on the motion state of the target UAV and the pose state of the docking device; predicting conflicts on the initial guidance path based on a maintained global spatiotemporal occupancy map and generating a cooperative guidance command; wherein the conflict includes the target UAV having paths intersecting with those of other moving entities on the initial guidance path; and controlling the movement of the target UAV according to the cooperative guidance command until the power receiving device of the target UAV and the docking device complete docking.

[0007] In the above implementation process, by integrating the real-time motion parameters of the target UAV with the attitude feedback of the docking device, a spatiotemporal consistency check is introduced in the initial guidance path generation stage; the dynamic obstacle information in the global spatiotemporal occupancy map is projected onto the path, and the potential intersection area is simulated forward, thereby outputting a collaborative guidance command that includes speed adjustment and heading correction; under the drive of the command, the target UAV gradually reduces the attitude deviation until the spatial constraint relationship between the power receiving device and the docking device meets the docking judgment conditions, and the physical connection is achieved.

[0008] Optionally, generating an initial guidance path from the current position of the target UAV to the target position of the docking device based on the motion state of the target UAV and the pose state of the docking device includes: calculating the original heading deviation and the original distance deviation between the current pose of the target UAV and the target pose of the docking device; compensating for the original heading deviation and the original distance deviation based on environmental interference parameters to obtain compensated heading deviation and compensated distance deviation; and generating an initial guidance path from the current position of the target UAV to the target position of the docking device based on the compensated heading deviation and the compensated distance deviation; wherein the initial guidance path is tangent to the heading of the current position of the target UAV at the starting point and consistent with the alignment direction of the target position of the docking device at the ending point.

[0009] In the above implementation process, the original heading and distance deviations between the target UAV and the docking device are first calculated using the real-time pose of the target UAV and the target pose of the docking device as inputs. Then, wind field and magnetic field drift environmental interference parameters are introduced to perform vector compensation on the original deviations, resulting in corrected heading and distance deviations. Based on this, a smooth curve path is constructed with the compensated deviations as boundary conditions. The starting heading is tangent to the current velocity direction of the UAV, and the ending heading is consistent with the alignment direction of the docking device. This path serves as the initial guidance path. The above steps can reduce the frequency of repeated recalculations caused by external disturbances, reduce the onboard computing load, and improve the timeliness of subsequent conflict prediction and command updates.

[0010] Optionally, the maintenance-based global spatiotemporal occupancy map, which performs conflict prediction on the initial guidance path and generates cooperative guidance instructions, includes: discretizing the initial guidance path into multiple predicted pose points distributed in a time series; generating a corresponding spatial occupancy envelope for each predicted pose point based on the physical contour of the target UAV; and mapping each spatial occupancy envelope to the global spatiotemporal occupancy map according to its corresponding timestamp to form the predicted spatiotemporal pipeline; wherein the global spatiotemporal occupancy map is stored in the form of a spatiotemporal voxel grid, and the conflict is determined by the existence of voxel overlap in the voxel grid.

[0011] In the above implementation process, the initial guidance path is first discretized into a series of predicted pose points with equal time steps, and a spatial occupancy envelope that varies with attitude is generated for each point based on the geometric shape of the target UAV. Then, each envelope is written into the corresponding voxel layer of the global spatiotemporal occupancy map according to its timestamp, forming a continuous predicted spatiotemporal pipeline. When the pipeline overlaps with any moving entity voxel recorded in the map, a conflict marker is triggered, and a cooperative guidance command for velocity scaling or heading offset is output accordingly, so that the UAV can continue to approach the docking device while maintaining a safe distance.

[0012] Optionally, the process of predicting conflicts on the initial guidance path based on the maintained global spatiotemporal occupancy map and generating collaborative guidance instructions further includes: defining a conflict resolution window if a conflict is predicted; freezing the autonomous path replanning function of all obstructions involved in the conflict within the conflict resolution window and prohibiting new path planning from occupying the spatiotemporal region corresponding to the window; and determining the obstruction with the highest priority and planning yield paths and waiting instructions for other obstructions to form a passage sequence and timetable, thereby generating collaborative guidance instructions; wherein the highest priority is calculated based on the task urgency factor, load factor, and estimated detour cost factor of the obstruction.

[0013] In the above implementation process, firstly, multi-dimensional conflict detection is performed on the initial guidance path based on the maintained global spatiotemporal occupancy map, and potential collision risks are identified through spatiotemporal trajectory cross-analysis. When a conflict is predicted, the system dynamically delineates a conflict resolution window period, which is a reserved time interval before the conflict occurs, used to coordinate the passage order of multiple agents. During this window period, the system takes mandatory control measures: on the one hand, it freezes the autonomous path replanning function of all obstructing agents involved in the conflict to prevent temporary path changes from causing coordination failure; on the other hand, it prohibits any new path planning application from occupying the spatiotemporal area corresponding to this window period, ensuring the exclusivity of the conflict resolution space. Subsequently, the system initiates a priority arbitration mechanism, comprehensively calculating the task urgency factor of each obstructing agent, which reflects the urgency of the task deadline, the load factor, which characterizes the importance or danger level of the load and the estimated detour cost factor, and which quantifies the resource consumption of alternative paths. A weighted comprehensive priority score is obtained to determine the passage subject with the highest priority. For obstructions with lower priority, the system automatically plans a yielding path, such as an alternative route, or generates a waiting instruction, specifying the waiting area and duration, ultimately forming an orderly passage sequence and a precise timetable. Based on this, collaborative guidance instructions are generated and issued for execution, thereby enabling safe and efficient collaborative passage of multiple agents under shared spatiotemporal resources.

[0014] Optionally, the method of performing conflict prediction on the initial guidance path based on the maintained global spatiotemporal occupancy map and generating cooperative guidance instructions further includes: if no conflict is predicted, using the initial guidance path as the cooperative guidance instruction.

[0015] In the above implementation process, the dynamic changes in the global spatiotemporal occupancy map are continuously monitored, and real-time conflict prediction analysis is performed on the initial guidance path. When the prediction results show that there is no conflict risk in the current path within the planned spatiotemporal range, it indicates that the path is exclusive and feasible in the spatiotemporal dimension, and there is no need to initiate complex conflict resolution and cooperative scheduling mechanisms. At this time, the system directly confirms the initial guidance path as the final execution plan, encapsulates it into a cooperative guidance instruction, and issues it to the corresponding intelligent agent for execution. This processing method avoids unnecessary consumption of computing resources and scheduling delays, and achieves efficient and lightweight processing of path planning while ensuring passage safety, ensuring rapid response and smooth passage in conflict-free scenarios.

[0016] Optionally, controlling the movement of the target UAV according to the cooperative guidance command until the target UAV's power receiving device and the docking device complete docking includes: when the cooperative guidance command does not include time constraints, generating a first motion control command based on the path information in the cooperative guidance command and the real-time pose of the target UAV through spatial path tracking calculation; when the cooperative guidance command includes time constraints, generating a second motion control command based on the path information in the cooperative guidance command, the time constraints, and the real-time pose of the target UAV through spatiotemporal joint tracking calculation; wherein the time constraints include the target arrival timestamp and target velocity corresponding to each path point in the path point sequence.

[0017] In the above implementation process, the content characteristics of the collaborative guidance command are first analyzed to determine whether it contains time constraints, and then a differentiated motion control strategy is adopted. When the collaborative guidance command only contains spatial path information without time constraints, the system enters the spatial path tracking mode: based on the path information in the command and the real-time pose of the target UAV, the lateral and longitudinal control quantities are calculated to generate the first motion control command, enabling the UAV to accurately track the preset spatial trajectory, focusing on ensuring path geometric accuracy and attitude stability. When the collaborative guidance command contains time constraints, the system switches to the spatiotemporal joint tracking mode: in addition to path information, time constraints are introduced and calculated through the spatiotemporal joint tracking algorithm. While ensuring spatial path accuracy, the timing and speed of the UAV arriving at each path point are strictly controlled to generate the second motion control command, enabling the UAV to accurately arrive at the docking position within the specified time window, meeting the timing synchronization requirements of multi-UAV collaborative docking. The two control modes adaptively switch according to mission requirements, ultimately driving the target UAV to move until its power receiving device and docking device complete physical docking.

[0018] Optionally, the step of generating the second motion control command through spatiotemporal joint tracking calculation includes: calculating the spatial deviation between the current position of the target UAV and the target waypoint; calculating the time tracking error between the current time and the target arrival timestamp corresponding to the target waypoint; and calculating the fused speed and heading control quantities based on the spatial deviation and the time tracking error to form the second motion control command.

[0019] In the above implementation process, the real-time pose information of the target UAV is first acquired, and the spatial deviation between its current position and the target path point is calculated, including multi-dimensional spatial metrics such as lateral position deviation, longitudinal position deviation, and altitude deviation, to quantify the degree of geometric deviation of the UAV from the desired spatial trajectory. Simultaneously, the system reads the current system time and calculates the time tracking error between the current system time and the target arrival timestamp corresponding to the target path point. This error reflects the UAV's leading or lagging state relative to the cooperative scheduling plan in the time dimension. Furthermore, based on the aforementioned spatial deviation and time tracking error, the system constructs a fusion control law: when the time tracking error is negative, i.e., the UAV is ahead of the plan, the speed control input is appropriately reduced to delay the arrival time; when the time tracking error is positive, i.e., the UAV is lagging behind the plan, the speed control input is moderately increased to catch up; at the same time, the heading control input is corrected in real time according to the spatial deviation to ensure geometric path tracking accuracy. By weighted and fused feedback information from both spatiotemporal dimensions, the system calculates comprehensive speed and heading control quantities to form a second motion control command. This enables joint optimization of spatial trajectory accuracy and time synchronization accuracy, ensuring that the UAV arrives at the target path point precisely at the specified time and speed, thus meeting the stringent requirements of spatiotemporal collaborative docking.

[0020] Optionally, controlling the target UAV's movement according to the cooperative guidance command until the target UAV's power receiving device and the docking device complete docking includes: confirming docking completion when the following conditions are met simultaneously: the locking force feedback value from the docking device reaches a preset threshold; the relative position deviation between the power receiving device and the docking device obtained by visual measurement is less than a preset tolerance; receiving a lock confirmation signal from the docking device; and in response to the docking completion, sending a start command to the electromagnetic catapult to trigger the catapult process and deleting the path occupancy information associated with the target UAV from the global spatiotemporal occupancy map.

[0021] In the above implementation process, a multi-source information fusion approach is used to jointly determine the docking completion status, ensuring the reliability and safety of the docking operation. Specifically, the system monitors and verifies the following three conditions in real time: First, the locking force feedback value from the docking device reaches a preset threshold, indicating that the mechanical locking mechanism has generated sufficient physical clamping force, and a stable mechanical connection is formed between the receiving device and the docking device; Second, the relative positional deviation between the receiving device and the docking device, including lateral, longitudinal, and angular deviations, measured by the airborne vision sensor, is less than a preset tolerance range, confirming that the two have achieved high-precision alignment in geometric position; Third, the system receives a locking confirmation signal from the control unit of the docking device, indicating that the electrical control system of the docking device has completed the self-check of the locking action and confirmed successful locking. The system determines that the docking completion status is established only when all three conditions are met simultaneously. Subsequently, the following process is triggered: a start command is sent to the electromagnetic catapult device to drive it into the pre-ejection energy storage state or to execute the ejection release process; at the same time, the global spatiotemporal occupancy map is updated, all path occupancy information associated with the target UAV is deleted, and the spatiotemporal resources originally occupied by it are released for other UAVs or intelligent agents to re-plan and use, thereby realizing the dynamic recycling and efficient reuse of spatiotemporal resources.

[0022] Secondly, embodiments of this application also provide a UAV docking system, characterized in that the UAV docking system includes: a status acquisition module, a path generation module, a cooperative arbitration module, a tracking control module, and a trigger closed-loop module; the status acquisition module is used to acquire the motion state of the target UAV and the pose state of the docking device in real time; the path generation module is used to generate an initial guidance path from the current position of the target UAV to the target position of the docking device based on the motion state of the target UAV and the pose state of the docking device; the cooperative arbitration module is used to predict conflicts on the initial guidance path based on a maintained global spatiotemporal occupancy map and generate cooperative guidance commands; the tracking control module is used to control the movement of the target UAV according to the cooperative guidance commands until the power receiving device of the target UAV and the docking device complete docking.

[0023] In the above implementation process, the UAV docking system collects the motion state of the target UAV and the pose state of the docking device in real time through the state acquisition module, providing accurate initial data for subsequent path planning; the path generation module generates an initial guidance path from the current position to the target position based on the above state information, using optimal control or geometric guidance algorithms to ensure the feasibility and smoothness of the path; the cooperative arbitration module introduces a global spatiotemporal occupancy map to perform dynamic conflict prediction on the initial guidance path, identify potential spatiotemporal conflicts with other UAVs or obstacles, and generate cooperative guidance commands to achieve multi-UAV coordinated collision avoidance; after receiving the cooperative guidance commands, the tracking control module drives the target UAV to move along the optimized path through a closed-loop control algorithm; finally, when the target UAV's power receiving device and the docking device reach the preset pose tolerance range, the system determines that the docking is complete, improving the accuracy, safety and multi-UAV cooperative efficiency of autonomous UAV docking.

[0024] Thirdly, embodiments of this application also provide an electronic device, which includes a memory and a processor. The memory stores program instructions, and when the processor reads and runs the program instructions, it executes the steps in any of the above implementation methods.

[0025] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer program instructions, which, when read and executed by a processor, perform the steps in any of the above implementations. Attached Figure Description

[0026] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a first schematic diagram of the drone docking method provided in the embodiments of this application; Figure 2 A schematic diagram of the unmanned aerial vehicle docking system provided in the embodiments of this application; Figure 3 This is a second schematic diagram of the drone docking method provided in the embodiments of this application; Figure 4 A third schematic diagram of the UAV docking method provided in the embodiments of this application; Figure 5 A fourth schematic diagram of the UAV docking method provided in the embodiments of this application; Figure 6The fifth schematic diagram of the UAV docking method provided in the embodiments of this application; Figure 7 The sixth schematic diagram is a UAV docking method provided in the embodiments of this application; Figure 8 This is a block diagram of an electronic device provided in an embodiment of this application.

[0028] Icons: 001-Status acquisition module; 002-Path generation module; 003-Cooperative arbitration module; 004-Tracking control module; 005-Trigger closed-loop module; 100-Electronic device; 111-Memory; 112-Memory controller; 113-Processor; 114-Peripheral interface; 115-Input / output unit; 116-Display unit. Detailed Implementation

[0029] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of the embodiments of this application.

[0030] In a first aspect, embodiments of this application provide a drone docking method applied to a server, which can be an electronic device with logical computing functions such as a personal computer (PC), tablet computer, smartphone, or personal digital assistant (PDA).

[0031] Please see Figure 1 , Figure 1 This is a first schematic diagram of the drone docking method provided in an embodiment of this application.

[0032] The UAV docking method includes: generating an initial guidance path from the current position of the target UAV to the target position of the docking device based on the motion state of the target UAV and the pose state of the docking device; predicting conflicts on the initial guidance path based on a maintained global spatiotemporal occupancy map and generating cooperative guidance commands; wherein, the conflicts include the target UAV having paths intersecting with other moving entities on the initial guidance path; controlling the movement of the target UAV according to the cooperative guidance commands until the target UAV's power receiving device and the docking device complete docking.

[0033] In the above implementation process, high-frequency position, velocity, attitude, and angular velocity information of the target UAV can be acquired in real time through airborne GNSS, IMU, visual odometry, or UWB positioning systems. The docking device provides its position, attitude, and motion status in real time through its own sensors or an external positioning system. A global spatiotemporal occupancy map is constructed and maintained through lidar, visual sensors, or multi-sensor fusion. This global spatiotemporal occupancy map not only includes static obstacles but also dynamically tracks and predicts the future trajectories of other moving entities such as UAVs, vehicles, and personnel. Based on the current state of the target UAV and the target state of the docking device, and considering UAV dynamic constraints, a smooth and feasible initial spatial path is generated. At the end or near the end of the path, attitude adjustment commands need to be added to the planning to ensure that the orientation of the UAV's power receiving device matches the interface orientation of the docking device. The initial path is compared with the global spatiotemporal occupancy map to detect potential conflict points on the path. This includes spatial intersections with the predicted trajectories of moving entities, as well as temporal overlaps. If a conflict is detected, a cooperative mechanism is activated.

[0034] In some embodiments of this application, the strategy may include assigning passage priorities to conflicting parties based on task urgency or rules. Local adjustments may be made to the initial path, such as changing the altitude layer, temporarily detouring, or hovering and waiting in a safe location. The speed of the drone or the conflicting party may be fine-tuned to achieve staggered passage in time and space. The cooperative strategy is translated into specific guidance commands, including three-dimensional position setpoints, velocity vectors, heading angles, and specific maneuver commands. The UAV flight control system drives the aircraft to accurately track the path based on the received cooperative guidance commands. In the final stage of approaching the docking device, a higher-precision relative navigation mode may be switched to perform position and attitude fine-tuning to ensure physical alignment between the powered device and the docking interface. The UAV is controlled to complete the docking at a safe contact speed. Successful docking is confirmed in real time through contact sensors, current detection, or visual feedback on the airborne or docking device. If docking fails, a retreat strategy is triggered to retry or return to a safe position. Throughout the guidance and docking process, the UAV status, environmental changes, and system health are continuously monitored. Multi-layered safety boundaries are established. If a positional deviation exceeding a preset threshold, communication delay, sensor malfunction, or sudden obstacle intrusion occurs, docking will be immediately terminated, and an emergency abort procedure will be executed, such as hovering, obstacle avoidance, or returning along a safe path.

[0035] Secondly, embodiments of this application also provide a drone docking system; please refer to [link to relevant documentation]. Figure 2 , Figure 2 A schematic diagram of a drone docking system provided in an embodiment of this application.

[0036] The UAV docking system includes: a status acquisition module 001, a path generation module 002, a cooperative arbitration module 003, a tracking control module 004, and a trigger closed-loop module 005. The status acquisition module 001 is used to acquire the motion status of the target UAV and the pose status of the docking device in real time. The path generation module 002 is used to generate an initial guidance path from the current position of the target UAV to the target position of the docking device based on the motion status of the target UAV and the pose status of the docking device. The cooperative arbitration module 003 is used to predict conflicts on the initial guidance path based on the maintained global spatiotemporal occupancy map and generate cooperative guidance commands. The tracking control module 004 is used to control the movement of the target UAV according to the cooperative guidance commands until the target UAV's power receiving device and the docking device complete docking.

[0037] In the above implementation process, the state acquisition module 001 is used as the data source to inject the real-time pose into the path generation module 002 to generate an initial guidance path that conforms to the dynamic constraints; the collaborative arbitration module 003 then calls the global spatiotemporal occupancy map to complete conflict prediction and priority sorting, and outputs executable collaborative guidance commands; the tracking control module 004 switches between spatial or spatiotemporal joint tracking modes according to the command type to continuously converge the pose error; each module iterates periodically until the docking completion conditions of the multi-source criteria are met, realizing a closed loop of the entire link from perception, planning, arbitration to control.

[0038] The drone docking method will be described in detail below in conjunction with the drone docking system. Please refer to [link / reference]. Figure 3 , Figure 3 This is a second schematic diagram of the drone docking method provided in an embodiment of this application.

[0039] Optionally, based on the motion state of the target UAV and the pose state of the docking device, an initial guidance path is generated from the current position of the target UAV to the target position of the docking device, including: calculating the original heading deviation and the original distance deviation between the current pose of the target UAV and the target pose of the docking device; compensating for the original heading deviation and the original distance deviation based on environmental interference parameters to obtain the compensated heading deviation and the compensated distance deviation; and generating the initial guidance path from the current position of the target UAV to the target position of the docking device based on the compensated heading deviation and the compensated distance deviation; wherein the initial guidance path is tangent to the heading of the current position of the target UAV at the starting point and consistent with the alignment direction of the target position of the docking device at the ending point.

[0040] In the above implementation process, the original heading and distance deviations between the target UAV and the docking device are calculated, forming the most basic geometric input for path planning. Given the uncertainties in the actual flight environment, such as wind disturbance, sensor noise, and positioning drift, the system does not directly use the original deviations. Instead, environmental interference parameters are introduced and corrected using an embedded compensation algorithm to obtain the compensated heading and distance deviations. Based on the accurate compensated deviations, the path planner is invoked. Its core constraint is that at the starting point (the UAV's current position), the tangent direction of the generated path must be consistent with the UAV's current heading, thus achieving a smooth, abrupt trajectory start. At the endpoint or docking target point, not only do the positions coincide, but the direction of arrival must also be completely consistent with the alignment direction required by the docking device interface. By constructing and maintaining a unified global spatiotemporal occupancy map, the planned paths of all ground mobile devices are mapped as spatiotemporal pipelines with a time dimension, enabling accurate and forward-looking prediction of potential path conflicts. The core improvement lies in the fact that when a conflict is predicted, the system does not engage in real-time, dynamic, and iterative negotiations. Instead, it proactively defines a conflict resolution window and freezes the local dynamic replanning functions of the relevant equipment during this window. A central arbitration unit then makes a one-time, unchangeable decision regarding the passage sequence and timetable based on pre-defined deterministic rules. The technical effect of this approach is that it fundamentally eliminates the decision-making oscillation problem caused by real-time micro-changes in the state of multiple devices in bottleneck areas. It transforms uncertain distributed conflicts into deterministic static scheduling, ensuring the determinism of system operation and overall throughput efficiency in high-frequency, high-density ground transportation scenarios.

[0041] In one embodiment of this application, a data processing module receives data collected by the state acquisition module 001 and calculates the heading deviation and distance deviation between the current pose of the target UAV and the target pose of the docking device; a guidance path generation module 002 generates an initial guidance path from the current position of the target UAV to the target position of the docking device based on the heading deviation and distance deviation; the input of the data processing module also includes wind speed and direction data and ground light intensity data from the environmental perception module, which constitute environmental interference parameters; the data processing module uses a formula and The calculated original heading deviation Deviation from original distance Pre-compensation is performed, among which For wind speed vectors, Light intensity, and The pre-calibrated compensation coefficient, The compensated heading deviation is obtained for the preset control period. and distance deviation The guidance path generation module 002 is based on and The initial guidance path is generated using a Bézier curve generation algorithm with curvature continuity constraints. This path is tangent to the current heading of the target UAV at its starting point and is aligned with the target alignment direction of the docking device at its ending point.

[0042] Optionally, please refer to Figure 4 , Figure 4 This is a third schematic diagram of the drone docking method provided in the embodiments of this application.

[0043] Based on the maintained global spatiotemporal occupancy map, conflict prediction is performed on the initial guidance path, and cooperative guidance instructions are generated. This includes: discretizing the initial guidance path into multiple predicted pose points distributed in a time series; generating a corresponding spatial occupancy envelope for each predicted pose point based on the physical contour of the target UAV; and mapping each spatial occupancy envelope to the global spatiotemporal occupancy map according to its corresponding timestamp to form a predicted spatiotemporal pipeline. The global spatiotemporal occupancy map is stored in the form of a spatiotemporal voxel grid, and conflict is determined by the existence of voxel overlap in the voxel grid.

[0044] In the above implementation process, firstly, the smooth initial guidance path generated in the previous step is discretely sampled according to the time step to obtain a series of predicted pose points with timestamps. Based on the physical contour of the UAV (including the envelope of the airframe size and moving parts such as rotors), a corresponding spatial occupancy envelope is generated for each future pose point. Then, these spatial envelopes are mapped sequentially onto the global map in chronological order. The maintained global spatiotemporal occupancy map adopts a spatiotemporal voxel grid data structure. The core operation for conflict detection is to compare the predicted spatiotemporal pipeline constructed above with this global grid. The conflict criterion is defined as follows: at some same future moment, the voxels occupied by the UAV pipeline overlap with the voxels pre-occupied by other entities (dynamic or static).

[0045] In one embodiment of this application, the collaborative arbitration module 003 accesses a global spatiotemporal occupancy map it maintains, which records the planned dynamic paths of other ground support equipment and their spatiotemporal occupancy information; it compares the initial guidance path with the global spatiotemporal occupancy map to predict whether path intersection conflicts will occur in the future; if a conflict is predicted, a conflict resolution window covering the conflict's spatiotemporal point is defined, and during this window period, the dynamic path replanning function of all mobile entities involved in the conflict is frozen; the global spatiotemporal occupancy map is stored in the form of a spatiotemporal voxel grid; the collaborative arbitration module 003 discretizes the initial guidance path into a series of time steps. The system generates a 3D spatial occupancy envelope for each predicted pose point based on the physical contour of the target UAV. The collaborative arbitration module 003 maps the spatial occupancy envelope of each predicted pose point to the spatiotemporal voxel grid according to its corresponding timestamp, forming a predicted spatiotemporal pipeline for the target UAV. The collaborative arbitration module 003 performs voxel-level conflict detection between this predicted spatiotemporal pipeline and the spatiotemporal pipelines of other devices already existing in the map. If voxel overlap exists, it is determined that there is a path intersection conflict in the prediction, and the start time of the conflict resolution window is... Subtract the safety margin from the earliest timestamp of voxel overlap. End time Add a safety margin to the timestamp of the latest voxel overlap. and execution time margin .

[0046] Optionally, please refer to Figure 5 , Figure 5 This is a fourth schematic diagram of the drone docking method provided in the embodiments of this application.

[0047] Based on the maintained global spatiotemporal occupancy map, conflict prediction is performed on the initial guidance path, and collaborative guidance instructions are generated. This also includes: defining a conflict resolution window if conflict is predicted; freezing the autonomous path replanning function of all obstructions involved in the conflict within the conflict resolution window and prohibiting new path planning from occupying the spatiotemporal region corresponding to that window; and identifying the obstruction with the highest priority, planning yield paths and waiting instructions for other obstructions, forming a passage sequence and timetable, and generating collaborative guidance instructions; wherein the highest priority is calculated based on the obstruction's task urgency factor, load factor, and estimated detour cost factor.

[0048] In the above implementation process, the time period and spatial area of ​​the conflict are first precisely analyzed to define a conflict resolution window. This window clarifies the decision-making and execution time boundaries for resolving the conflict. During the window, temporary control is implemented in the conflict area, temporarily freezing the autonomous path replanning functions of all obstructions (other drones, mobile robots, or aerial vehicles) involved in the conflict, putting them into a controlled collaborative mode and subject to central scheduling. Any new path planning requests are prohibited from occupying the corresponding spatiotemporal voxels within this window, effectively reserving clean airspace resources for the conflict resolution. For each obstruction, its task urgency factors (such as battery power, task deadline, payload factors, carrying critical equipment, and estimated detour costs, such as additional energy consumption and time loss) are comprehensively evaluated, and a priority score is calculated through weighted calculation. The obstruction with the highest score is granted the highest priority within the current conflict window and has the right to proceed according to the original plan or the optimal path. Based on the above arbitration results, the system does not simply make lower-priority obstructions wait, but proactively plans efficient and safe yielding paths and provides precise waiting instructions, such as specifying avoidance points and hovering durations. Ultimately, a clear sequence of actions and a precise timetable are established that are known to all parties to the conflict.

[0049] In one embodiment of this application, the collaborative arbitration module 003 sends a path replanning freeze command to the tracking control module 004 corresponding to all mobile entities involved in the conflict; the path replanning freeze command includes an identifier of the conflict resolution window, and Upon receiving the instruction, the tracking control module 004 stops executing the dynamic local path replanning algorithm based on real-time sensor data during the window period, and instead fully follows the instruction from the collaborative arbitration module 003. At the same time, the collaborative arbitration module 003 marks the spatiotemporal region covered by the conflict resolution window as an arbitration locked state in the global spatiotemporal occupancy map, prohibiting new path planning from directly occupying the region.

[0050] The preset deterministic arbitration rule is a static priority rule. The collaborative arbitration module 003 stores a mapping table between device types and fixed priorities. When deterministic arbitration is required, the collaborative arbitration module 003 queries this mapping table to obtain the device types of all mobile entities involved in the conflict, and determines their respective fixed priority values ​​accordingly. Collaborative Arbitration Module 003 compares the various entities'... , will the highest The entity corresponding to the value is determined as the highest priority entity; the collaborative arbitration module 003 maintains the original predicted spatiotemporal pipeline for the highest priority entity or makes minimal optimization adjustments, calculates the precise waiting position for other entities before the conflict point, and generates the time from the current moment to... The deceleration and stopping instruction, and from The instructions to restart and proceed along the adjusted path are then executed, thus forming a deterministic passage sequence and a precise timetable. The preset deterministic arbitration rule is a dynamic weighted priority rule; the collaborative arbitration module 003 calculates a dynamic priority weight for each moving entity involved in the conflict. The calculation formula is: ,in This is a urgency factor based on the countdown to the task deadline. The load factor is set based on the priority of the transported goods. This is the estimated detour distance factor required to yield the right-of-way. The weighting coefficients are preset; the collaborative arbitration module 003 compares the values ​​of each entity. , will the highest The entity corresponding to the value is determined as the highest priority entity; the collaborative arbitration module 003, with the constraint of ensuring the smooth path of the highest priority entity, solves the path adjustment and time scheduling scheme that satisfies the collision-free condition and minimizes the total system cost for other entities, generating a deterministic passage sequence and a precise timetable.

[0051] In one embodiment of this application, the collaborative guidance instruction generated by the collaborative arbitration module 003 includes at least a path point sequence and a target arrival timestamp corresponding to each path point. and the target speed at each waypoint. The collaborative arbitration module 003 will process the path point sequence, and The command is encapsulated into a data packet and sent to the tracking and control module 004 corresponding to the target UAV; simultaneously, the collaborative arbitration module 003, based on the path point sequence and... The precise spatiotemporal occupancy pipeline of the target UAV during the conflict resolution window is recalculated, and this new pipeline is used to cover or update the original predicted spatiotemporal pipeline in the global spatiotemporal occupancy map to ensure that the map information is consistent with the arbitration result.

[0052] Optionally, please refer to Figure 6 , Figure 6 This is a fifth schematic diagram of the drone docking method provided in the embodiments of this application.

[0053] Based on the maintained global spatiotemporal occupancy map, conflict prediction is performed on the initial guidance path, and cooperative guidance instructions are generated. It also includes: if no conflict is predicted, the initial guidance path is used as the cooperative guidance instruction.

[0054] In the above implementation process, the initial guidance path is discretized and combined with the UAV's physical contour to construct its predicted spatiotemporal pipeline in the global spatiotemporal occupancy map. This pipeline is compared with the spatiotemporal voxels already occupied by other entities in the map to perform conflict detection. If the detection shows that the predicted spatiotemporal pipeline does not overlap with any other entity's occupied voxels, it is determined that there is no conflict. At this time, the initial guidance path is directly adopted and issued to the target UAV as the final determined cooperative guidance command, ensuring that the system can perform the docking mission with the lowest latency and highest efficiency when the airspace is clear. If voxel overlap is detected, it is determined that there is a conflict. Then, a complete cooperative decision-making procedure is initiated, defining the conflict resolution window period and clarifying the spatiotemporal range of the conflict; freezing the autonomous replanning of relevant obstructing entities and locking the conflict area; calculating priorities based on factors such as task urgency, payload, and detour cost, determining the passage sequence, and planning yielding paths and waiting instructions for low-priority obstructing entities to form a cooperative scheme including a timetable. Finally, the adjusted path formulated for the UAV and the yielding instructions formulated for other entities are packaged together to generate a complete set of cooperative guidance instructions.

[0055] Optionally, controlling the target UAV's movement according to the cooperative guidance command until the target UAV's power receiving device and docking device complete docking includes: when the cooperative guidance command does not contain time constraints, generating a first motion control command based on the path information in the cooperative guidance command and the real-time pose of the target UAV through spatial path tracking calculation; when the cooperative guidance command contains time constraints, generating a second motion control command based on the path information in the cooperative guidance command, the time constraints, and the real-time pose of the target UAV through spatiotemporal joint tracking calculation; wherein, the time constraints include the target arrival timestamp and target velocity corresponding to each path point in the path point sequence.

[0056] In the above implementation process, after receiving the cooperative guidance command generated in the previous step, the UAV's flight control system does not simply execute path point tracking. Instead, it first parses the constraint type of the command and initiates different control laws accordingly to achieve precise matching with the upper-level cooperative decision-making intent. Parsing the cooperative guidance command determines whether it contains precise time constraints. These constraints are typically given in the form of target arrival timestamps and target speeds for each point in the path point sequence, directly reflecting the results of multi-aircraft cooperative scheduling. If the command does not contain time constraints, it indicates that the current airspace is unobstructed, or that the cooperation only requires spatial avoidance without strict timing requirements. In this case, the core objective is to accurately track the spatial path. Based on the path information in the command and the UAV's real-time pose, a first motion control command is generated with position and heading as control objectives, guiding the UAV smoothly to the docking point.

[0057] In one embodiment of this application, the tracking control module 004 is internally divided into two processing branches. When the input cooperative guidance command does not contain a timing control command, the first branch is used to output a first motion control command, causing the UAV to fly along a given route. This branch uses a pure spatial path tracker, whose input is the path point sequence in the cooperative guidance command and the real-time pose of the target UAV fed back by the state acquisition module 001. By calculating the lateral deviation and heading deviation, the desired front wheel angle and speed control amount are output to form the underlying motion control command.

[0058] Optionally, a second motion control command is generated through spatiotemporal joint tracking calculation, including: calculating the spatial deviation between the current position of the target UAV and the target waypoint; calculating the time tracking error between the current time and the target arrival timestamp corresponding to the target waypoint; and calculating the fused speed and heading control quantities based on the spatial deviation and time tracking error to form the second motion control command.

[0059] In the above implementation process, if the command includes time constraints, it indicates that the UAV is operating within a precise timing schedule for multi-UAV collaboration. At this point, the controller's objective is upgraded to simultaneously satisfy both spatial path and time node requirements. Based on path information, time constraints, and real-time pose, advanced control algorithms capable of explicitly handling time variables, such as time-varying model predictive control, generate a second motion control command. This second motion control command not only controls the UAV to reach the designated position but also strictly controls its arrival time and instantaneous speed, ensuring the entire swarm operates collaboratively like precise gears.

[0060] In one embodiment of this application, when the input cooperative guidance command includes timing control commands, a second branch is followed, and a second motion control command is output to perform closed-loop adjustment of the UAV speed, ensuring that it passes through a specified path point at a specified time. This branch employs a spatiotemporal joint tracker, whose input includes, in addition to the path point sequence and real-time pose, the target arrival timestamp from the cooperative guidance command. and target speed The tracker additionally calculates the time tracking error. ,in The current time is used, and a time-velocity coupled controller is used to... Converted to reference speed Adjustment amount The final output integrates the desired front wheel steering angle and speed control quantity, which combines spatial correction and time synchronization, to form the underlying motion control command. The motion execution module is a differential drive or Ackerman steering chassis and its servo controller, which receives the underlying motion control command, calculates it into specific control signals for the left and right wheels or steering servos, and executes them.

[0061] Optionally, please refer to Figure 7, Figure 7 This is a sixth schematic diagram of the drone docking method provided in the embodiments of this application.

[0062] Controlling the target UAV's movement according to the collaborative guidance command until the target UAV's power receiving device and docking device complete docking includes: confirming docking completion when the following conditions are met simultaneously: the locking force feedback value from the docking device reaches a preset threshold; the relative position deviation between the power receiving device and the docking device obtained by visual measurement is less than a preset tolerance; receiving a lock confirmation signal from the docking device; in response to docking completion, sending a start command to the electromagnetic catapult to trigger the catapult process, and deleting the path occupancy information associated with the target UAV from the global spatiotemporal occupancy map.

[0063] In the above implementation process, the system will only confirm the docking completion if all of the following conditions are met simultaneously: the locking force (or contact pressure) feedback value from the docking device reaches the preset physical locking threshold, proving that a reliable mechanical connection has been formed; the relative position and attitude deviation between the interface of the receiving device and the docking device is determined to be less than the preset millimeter tolerance through a high-precision visual measurement system, proving that the electrical interface has been precisely aligned; and a locking completion or connection establishment digital signal actively sent by the docking device is received, completing the handshake at the communication protocol level. Once docking completion is confirmed, the system immediately executes two key state transition operations to achieve task handover and resource release. In response to the docking completion signal, the control system automatically sends a start command to the electromagnetic catapult device (or other launch / recovery mechanism), thereby triggering the subsequent fixing, charging, or catapult redeployment process. Simultaneously, all future path occupancy information associated with the target UAV is deleted from the global spatiotemporal occupancy map. This operation is crucial because it means that the drone has transformed from a mobile entity in transit to a fixed docking entity, and its previously reserved space and time resources (especially the airspace occupied by the planned path) are immediately released for use by other drones in the system, thereby improving the overall airspace utilization efficiency.

[0064] In one embodiment of this application, the preset docking completion standard must simultaneously meet three conditions: Condition 1, the locking force feedback value from the embedded force sensor on the docking device. Greater than or equal to the preset rated locking force threshold Condition two: The relative positional deviation norm between the power receiving device and the docking device is calculated by the state acquisition module 001 through visual measurement. Less than the preset allowable positional tolerance Condition 3: The trigger closed-loop module 005 receives a digital signal lock confirmation pulse from the docking device controller. The trigger closed-loop module 005 determines to generate a valid docking completion signal only when the above three conditions are verified to be true within the same control cycle. In response to the valid signal, the trigger closed-loop module 005 first sends a "catapult start" trigger command to the main controller of the electromagnetic catapult device, and then deletes all spatiotemporal occupancy pipeline records associated with the identifier of the target UAV from the data structure of the global spatiotemporal occupancy map, and marks the resources under the identifier as reassignable.

[0065] In a specific embodiment of this application, after the target UAV-01 lands on helipad A, its automated guided vehicle (AGV) is activated. The status acquisition module 001, by fusing visual and ultra-wideband data, determines the initial pose of UAV-01 as coordinates 0 meters, 0 meters, heading 0 degrees, and velocity zero. Simultaneously, another recovered UAV-02 reports its location at coordinates 150 meters, 30 meters, planning to proceed to helipad B. The collaborative arbitration module 003 converts this information into a spatiotemporal occupancy area and updates it to the global spatiotemporal occupancy map.

[0066] After receiving the UAV-01's pose and the target pose coordinates of the electromagnetic catapult docking device (100m and 200m), the path generation module 002 calculates the initial heading and distance deviations. The data processing module incorporates current wind speed (1.5 m / s) and light intensity (80,000 lux) data provided by the environmental perception module, and uses pre-calibration coefficients to pre-compensate for the deviations. The compensation formula is as follows: and The time period is 0.1 seconds. Based on the compensated deviation, a smooth initial guiding path from the starting point to the ending point is generated using a cubic Bézier curve algorithm.

[0067] After receiving the initial guidance path from UAV-01, the collaborative arbitration module 003 initiates the collaborative arbitration process. The module discretizes the path and predicts its spatiotemporal pipeline, comparing it with the existing predicted pipeline of UAV-02 in the global map. Calculations reveal that the two paths spatially overlap near the taxiway intersection R at approximately 18 seconds, indicating a path intersection conflict.

[0068] Accordingly, the collaborative arbitration module 003 defines a conflict resolution window with a time range of 16 to 25 seconds, and marks the spatiotemporal region of intersection R as an arbitration locked state during this window period, while notifying relevant mobile entities to freeze the dynamic path replanning function.

[0069] Subsequently, the collaborative arbitration module 003 performs deterministic arbitration based on the dynamic weight priority rule. The calculation formula for this rule is as follows: Among the key factors, mission urgency, payload priority, and detour distance were considered. Calculations showed that UAV-01, due to its urgent mission and carrying medical supplies, received higher weight and was granted priority passage. Based on this, the collaborative arbitration module 003 performed static planning: UAV-01 maintained its original plan to pass through the intersection at a speed of 2.5 meters per second; UAV-02 was replanned to stop and wait at a designated waiting point before the intersection until the end of the passage window. Finally, a collaborative guidance instruction containing timestamp constraints was generated for UAV-01, and this was used to update the global spatiotemporal occupancy map, ensuring that the map state was consistent with the arbitration result.

[0070] Upon receiving a coordinated guidance command with timing constraints, the tracking control module 004 of UAV-01 activates the spatiotemporal joint tracker. Approaching intersection R, the tracker calculates spatial deviation and temporal tracking error in real time. For example, at 17.5 seconds, the actual position of the UAV lags behind the command requirement by 0.5 seconds. The spatiotemporal joint tracker, through a proportional control algorithm, converts the time error into a speed adjustment, outputting a low-level motion control command accelerating to 2.6 meters per second, while simultaneously combining the steering correction output from the spatial path tracker. The tracking control module 004 executes these commands, driving UAV-01 to pass through the center area of ​​the intersection precisely at 18.0 seconds. Simultaneously, UAV-02 strictly adheres to the command, remaining stationary at the waiting point, achieving conflict-free coordinated operation.

[0071] UAV-01 eventually moves to the electromagnetic catapult position, where its power receiving device and docking device complete physical connection. The docking completion signal is generated after three conditions are simultaneously met: the locking force reaches a threshold of 1250 Newtons, the visual positioning deviation is less than 0.02 meters, and a mechanical lock confirmation signal is received. The tracking control module 004 responds to this signal, triggering the electromagnetic catapult process and removing UAV-01 from the global spatiotemporal occupancy map, releasing all its occupied logical resources. At this point, the system completes a fully automated closed-loop operation from automatic guidance, conflict arbitration, precise docking, to catapult launch. Throughout the process, the introduction of a conflict resolution window and deterministic arbitration effectively avoids decision oscillation problems in multi-entity path planning, ensuring system determinism and operational efficiency under high-frequency operations.

[0072] Optionally, please refer to Figure 8 , Figure 8 This is a block diagram illustrating an electronic device according to an embodiment of this application. The electronic device 100 may include a memory 111, a memory controller 112, a processor 113, a peripheral interface 114, an input / output unit 115, and a display unit 116. Those skilled in the art will understand that... Figure 8 The structure shown is for illustrative purposes only and does not limit the structure of the electronic device 100. For example, the electronic device 100 may also include components that are more... Figure 8The more or fewer components shown, or having the same Figure 8 The different configurations shown.

[0073] The aforementioned memory 111, memory controller 112, processor 113, peripheral interface 114, input / output unit 115, and display unit 116 are electrically connected directly or indirectly to each other to achieve data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines. The aforementioned processor 113 is used to execute executable modules stored in the memory.

[0074] The memory 111 can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory 111 stores programs, and the processor 113 executes these programs upon receiving execution instructions. The methods executed by the electronic device 100 as defined in any embodiment of this application can be applied to the processor 113, or implemented by the processor 113.

[0075] The aforementioned processor 113 may be an integrated circuit chip with signal processing capabilities. The processor 113 may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a digital signal processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor.

[0076] The peripheral interface 114 described above couples various input / output devices to the processor 113 and the memory 111. In some embodiments, the peripheral interface 114, the processor 113, and the memory controller 112 can be implemented on a single chip. In other instances, they can be implemented on separate chips.

[0077] The input / output unit 115 described above is used to provide user input data. The input / output unit 115 may be, but is not limited to, a mouse and keyboard, etc.

[0078] The aforementioned display unit 116 provides an interactive interface (e.g., a user interface) between the electronic device 100 and the user, or displays image data for the user's reference. In this embodiment, the display unit can be a liquid crystal display (LCD) or a touch display. If it is a touch display, it can be a capacitive touchscreen or a resistive touchscreen that supports single-point and multi-point touch operations. Supporting single-point and multi-point touch operations means that the touch display can sense touch operations generated simultaneously from one or more locations on the touch display and pass the sensed touch operations to the processor for calculation and processing.

[0079] This application also provides a computer-readable storage medium storing computer program instructions, which are read and executed by a processor to perform steps in the UAV docking method.

[0080] In summary, this application provides a UAV docking method, system, and electronic device, relating to the field of UAV docking technology. Based on the motion state of the target UAV and the pose state of the docking device, an initial guidance path is generated from the current position of the target UAV to the target position of the docking device. Based on a maintained global spatiotemporal occupancy map, conflict prediction is performed on the initial guidance path, and cooperative guidance commands are generated. Conflicts include path intersections between the target UAV and other moving entities on the initial guidance path. The target UAV's movement is controlled according to the cooperative guidance commands until the target UAV's power receiving device and the docking device complete docking. This solves the decision oscillation problem caused by real-time micro-changes in the state of multiple devices in bottleneck areas, transforming uncertain distributed conflicts into deterministic static scheduling, ensuring the determinism of system operation and overall throughput efficiency in high-frequency, high-density ground transportation scenarios.

[0081] In the several embodiments provided in this application, it should be understood that the disclosed device can also be implemented in other ways. The device embodiments described above are merely illustrative; for example, the block diagrams in the accompanying drawings illustrate the possible architecture, functions, and operations of the device according to various embodiments of this application. In this regard, each block in the block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram, and combinations of block diagrams, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0082] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0083] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0084] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application. It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0085] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

[0086] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A method for docking unmanned aerial vehicles (UAVs), characterized in that, The method includes: Based on the motion state of the target UAV and the pose state of the docking device, an initial guidance path is generated from the current position of the target UAV to the target position of the docking device; Based on the maintained global spatiotemporal occupancy map, conflict prediction is performed on the initial guidance path, and cooperative guidance instructions are generated; wherein, the conflict includes the target UAV having path intersections with other mobile entities on the initial guidance path; The target UAV is controlled to move according to the cooperative guidance command until the target UAV's power receiving device and the docking device complete docking.

2. The method according to claim 1, characterized in that, The generation of an initial guidance path from the current position of the target UAV to the target position of the docking device based on the motion state of the target UAV and the pose state of the docking device includes: Calculate the original heading deviation and original distance deviation between the current pose of the target UAV and the target pose of the docking device; The original heading deviation and the original distance deviation are compensated based on environmental interference parameters to obtain the compensated heading deviation and the compensated distance deviation. Based on the compensated heading deviation and the compensated distance deviation, an initial guidance path is generated from the current position of the target UAV to the target position of the docking device; The initial guidance path is tangent to the heading of the target UAV at its starting point and aligned with the target position of the docking device at its ending point.

3. The method according to claim 1, characterized in that, The maintained global spatiotemporal occupancy map performs conflict prediction on the initial guidance path and generates cooperative guidance instructions, including: The initial guidance path is discretized into multiple predicted pose points distributed according to a time series. Based on the physical contour of the target UAV, a corresponding spatial occupancy envelope is generated for each of the predicted pose points; Each spatial occupancy envelope is mapped to the global spatiotemporal occupancy map according to its corresponding timestamp, forming a predictive spatiotemporal pipeline; The global spatiotemporal occupancy map is stored in the form of a spatiotemporal voxel grid, and the conflict is determined by the existence of voxel overlap in the voxel grid.

4. The method according to claim 3, characterized in that, The maintained global spatiotemporal occupancy map, which performs conflict prediction on the initial guidance path and generates cooperative guidance instructions, also includes: In the event of a predicted conflict, a conflict resolution window period should be defined. During the conflict resolution window, the autonomous path replanning functions of all obstructing entities involved in the conflict are frozen, and new path planning is prohibited from occupying the spatiotemporal region corresponding to this window; and Identify the obstacle with the highest priority, plan yielding paths and waiting instructions for other obstacles, form a passage sequence and timetable, and generate coordinated guidance instructions; The highest priority is calculated based on the task urgency factor, load factor, and estimated detour cost factor of the obstruction.

5. The method according to claim 3, characterized in that, The maintained global spatiotemporal occupancy map, which performs conflict prediction on the initial guidance path and generates cooperative guidance instructions, also includes: If no conflict is predicted, the initial guidance path is used as the cooperative guidance instruction.

6. The method according to claim 1, characterized in that, The step of controlling the target UAV to move according to the cooperative guidance command until the target UAV's power receiving device and the docking device complete docking includes: If the cooperative guidance command does not contain time constraints, a first motion control command is generated based on the path information in the cooperative guidance command and the real-time pose of the target UAV through spatial path tracking calculation. When the collaborative guidance command includes time constraints, a second motion control command is generated based on the path information, time constraints, and real-time pose of the target UAV in the collaborative guidance command through spatiotemporal joint tracking calculation. The time constraints include the target arrival timestamp and target speed corresponding to each path point in the path point sequence.

7. The method according to claim 6, characterized in that, The generation of the second motion control command through spatiotemporal joint tracking calculation includes: Calculate the spatial deviation between the current position of the target UAV and the target waypoint; Calculate the time tracking error between the current time and the target arrival timestamp corresponding to the target path point; Based on the spatial deviation and the time tracking error, the fused speed and heading control quantities are calculated to form the second motion control command.

8. The method according to claim 1, characterized in that, The step of controlling the target UAV to move according to the cooperative guidance command until the target UAV's power receiving device and the docking device complete docking includes: The docking is confirmed to be complete when the following conditions are met simultaneously: The locking force feedback value from the docking device reaches a preset threshold; The relative positional deviation between the power receiving device and the docking device, obtained by visual measurement, is less than a preset tolerance. Received a lock confirmation signal from the docking device; In response to the completion of the docking, a start command is sent to the electromagnetic catapult to trigger the catapult process, and the path occupancy information associated with the target UAV is deleted from the global spatiotemporal occupancy map.

9. A drone docking system, characterized in that, The UAV docking system includes: a status acquisition module, a path generation module, a collaborative arbitration module, a tracking and control module, and a triggering closed-loop module; The state acquisition module is used to acquire the motion state of the target UAV and the pose state of the docking device in real time. The path generation module is used to generate an initial guidance path from the current position of the target UAV to the target position of the docking device based on the motion state of the target UAV and the pose state of the docking device. The collaborative arbitration module is used to predict conflicts in the initial guidance path based on the maintained global spatiotemporal occupancy map and generate collaborative guidance instructions. The tracking and control module is used to control the movement of the target UAV according to the cooperative guidance command until the power receiving device of the target UAV completes docking with the docking device.

10. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing program instructions, and the processor executing the steps of the method according to any one of claims 1-8 when running the program instructions.

Images