A remote control communication system of a UAV supporting preset flight actions and cross-device authorization

Abstract

The application discloses a kind of unmanned plane remote control communication systems supporting preset flight action and cross-device authorization, it is related to unmanned plane communication and flight remote control technical field, including: mobile phone end APP built-in three-dimensional scene rendering, preset action management and device authorization module, support three-dimensional gesture control, multiple types preset flight action template, real scene recording conversion template, realize control authority cross-device distribution switching and multi-machine collaborative control;This system also has a number of innovative practical security functions: intelligent shortcut and related system functions, including the second communication channel;Intelligent agent system: the function image and information image of user setting information in service platform;Intelligent agent takeover or background management personnel manual takeover under abnormal condition.The application improves unmanned plane control ease of use and better user experience degree, operation reuse efficiency and system reliability, applicable to the application scene of remote control various unmanned planes.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) communication and flight remote control technology, and more specifically to a UAV remote control communication system that supports preset flight actions and cross-device authorization. Background Technology

[0002] Traditional drone remote control systems rely on dedicated remote controllers and point-to-point wireless links, which have inherent drawbacks such as inconvenience in carrying, limited communication distance, and weak functional expandability. With the widespread use of smartphones and public mobile communication networks, drone flight remote control technology based on mobile apps and public network servers has become feasible and practical. It breaks through the limitation of remote control distance by leveraging 4G / 5G networks and improves ease of use by relying on the interactive capabilities of mobile phones.

[0003] Based on this, the UAV remote control communication system of this invention proposes several key technical points and provides innovative technical solutions: First, multiple types of preset flight action templates. These include preset flight action templates with fixed parameters, preset flight action templates with adjustable parameters, and preset flight action templates with real-time manual parameter input, to match different application scenarios. Second, intelligent shortcuts; the service platform has a built-in intelligent agent system, as well as functional images and information images of user settings on the mobile phone, providing the functional basis for temporarily taking over the user's remote control of the UAV to continue flight or return in case of abnormalities, ensuring the safe flight of the UAV. Third, an innovative second communication channel. If the mobile APP is terminated or malfunctions, the user can directly call the service platform's service hotline to communicate with the intelligent agent system or issue commands to remotely control the UAV to continue flight or return. Thus, the risk resistance and safety system of the UAV remote control system are significantly enhanced.

[0004] Therefore, the invention proposes a UAV remote control communication system that supports preset flight actions and cross-device authorization, and overcomes the limitations of existing technologies. This invention represents a technological innovation and solution in the field of UAV remote control communication technology. Summary of the Invention

[0005] In view of this, the present invention provides a drone remote control communication system that supports preset flight maneuvers and cross-device authorization. Based on a WebSocket-based network relay communication architecture, it constructs a three-layer technical system comprising basic architecture, core functions, and deep optimization. It adds five major functions: 3D scene control, multiple types of preset flight maneuver templates, real-scene recording to template conversion, cross-device permission authorization, and multi-device / multi-drone collaboration. Furthermore, it proposes several technological innovations: multiple types of preset flight maneuver templates that can flexibly adapt to different application scenarios; practical and powerful intelligent shortcuts and a second communication channel, which are convenient, fast, safe, and reliable. This achieves a drone remote control communication system that is easy to operate, template-based in operation, collaborative in management and control, adaptive in transmission, and intelligent in control. To achieve the above objectives, the present invention adopts the following technical solution: A drone remote control communication system that supports preset flight maneuvers and cross-device authorization includes: Drone terminals, mobile terminals, and web servers; The drone is equipped with an embedded system including a wireless communication module that accesses the public network via a mobile communication network, and an optional configuration: a WebSocket client for the embedded system; The mobile app has a flight control device installed. This app integrates a WebSocket client and communicates with the server via the WebSocket protocol. The flight control app also integrates a 3D scene rendering module, a preset action management module, and a device authorization module. The 3D scene rendering module relies on the mobile phone hardware and software system to construct a 3D spatial scene mapping the spatial position of the drone within the flight control APP; the preset action management module is used to generate, edit, store and call preset flight action templates, which are a set of serialized flight control commands; the device authorization module works in conjunction with the server's authorization management unit to grant, revoke and switch drone control permissions between different mobile devices. The server is a WebSocket server deployed on the public network, used for authentication and session binding, and for bidirectional real-time data forwarding between the APP and the drone. The server is equipped with an authorization management unit and a template storage unit.

[0006] Optionally, the 3D scene rendering module supports one-click switching between front view, side view, top view and stereoscopic perspective view; it supports touch screen interactive operation, including single-finger swipe to control the horizontal displacement of the drone, two-finger zoom to adjust the view scale and flight altitude, two-finger rotation to adjust the drone's heading, long press and drag to draw continuous flight trajectory and automatically generate corresponding command sequences; the control operations in the 3D scene are converted into drone flight control commands through coordinate system mapping and sent to the drone terminal for execution via WebSocket link.

[0007] Optionally, the preset flight action templates managed by the preset action management module are divided into three types: Fixed parameter template: The action sequence and all running parameters are fixed, and the entire set of flight actions are executed directly when called; Adjustable parameter template: The core action framework is fixed, and key operating parameters can be customized by users. After the parameters are modified, their legality is automatically verified and a new instruction sequence is generated. Real-time parameter template: During the execution of flight maneuvers of the remote-controlled drone, the parameter values ​​of the flight maneuver template are manually entered to control the drone's flight.

[0008] Optionally, the preset action management module also includes a real-scene flight control recording unit, which is used to enable the operation recording function and record all remote control commands and UAV status data during the user's manual operation in real time; after recording, it supports editing operations such as trimming, parameter modification, and node addition and deletion of the command sequence, and saves it as a preset flight action template, turning the real-scene flight control operation into a reusable template.

[0009] Optionally, the device authorization module allows the master control mobile phone to authorize the control permissions of a specified drone to the flight control APP of another mobile phone; the authorization process generates an encrypted authorization token through the server authorization management unit, and the authorization scope includes the permission level, validity period and scope of executable operations; when the security mutual exclusion option is enabled, only one mobile phone terminal is allowed to have full control permissions for a drone at the same time, and the original control terminal permissions are automatically suspended when the new authorization takes effect.

[0010] Optionally, multiple drones can be managed and controlled simultaneously through a single mobile flight control app. Different drone device IDs can be matched through a drop-down list, supporting group management and batch command issuance. Multiple independent flight control app clones can be installed on the same mobile phone, and each flight control app clone can be matched and controlled separately for its corresponding drone.

[0011] Optionally, the same drone can be bound to multiple mobile devices, with a master control phone and at least one backup phone set up. The server monitors the connection status of the master control phone in real time, and automatically switches control to the backup phone when the master control phone malfunctions or the network signal is interrupted, thus seamlessly taking over.

[0012] Optionally, the server has a built-in intelligent agent system for performing functional mapping / information mapping of user settings and stored data in the user's mobile flight control APP; It also includes a second communication channel, through which users can bypass the mobile flight control APP and directly dial the server platform's service phone number / SMS to conduct two-way voice communication with the intelligent agent system, issue flight control commands through voice calls, and the intelligent agent system directly remotely controls the drone; If the intelligent agent system detects an anomaly in the drone's flight control communication, the intelligent agent system will call the user's mobile phone and interact with the user directly via voice, and the drone's flight control will continue. If communication between the server and the user is interrupted, the intelligent agent system will take over the flight control of the drone that is still in flight for the user. The intelligent agent system will control the drone's flight or return it to home based on the information image of the user on the platform.

[0013] Optionally, the server is also equipped with a cluster arbitration unit for airspace access management and task priority arbitration of multiple drone clusters; The cluster arbitration unit gathers the location, preset flight path and task priority data of all online drones in real time. When it detects that the future flight paths of two or more drones have overlapping airspace conflicts, it arbitrates according to the task priority level and equipment permission level, issues a hovering avoidance command to the low-priority drone, and restores the operation permission of the low-priority drone after the high-priority drone leaves the conflict airspace. When a communication or equipment malfunction is detected in a drone, the cluster arbitration unit can automatically transfer the unfinished task permissions and task snapshot to the highest priority standby drone in the same group for seamless task succession.

[0014] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a drone remote control communication system that supports preset flight actions and cross-device authorization, which has the following beneficial effects: This invention leverages the native computing power of mobile phones to achieve 3D spatial scene rendering and gesture interaction, upgrading 2D button control to 3D spatial control without additional hardware costs. This lowers the barrier to entry and improves trajectory control accuracy and spatial awareness. It constructs a system of three preset flight action templates: fixed, adjustable, and real-time, covering all scenario requirements from standardized to adaptive approaches. Combined with a one-click template conversion function for real-scene recording, it significantly reduces the barrier to creating customized job templates, enabling multiple reuses after a single flight and improving the efficiency and consistency of repetitive tasks. A standardized cross-device authorization mechanism supports hierarchical and time-limited permission distribution, adapting to the permission management needs of team operations. A multi-device backup mechanism enables seamless fault switching, improving system redundancy and reliability. Single-APP multi-device management and batch command issuance capabilities support cluster operation scenarios. Based on a link quality-aware dynamic command mode switching mechanism, it automatically matches the optimal command sending strategy and buffer configuration to address the inherent characteristics of public 4G / 5G link fluctuations. This achieves the best balance between efficiency and reliability in different network environments, solving the core pain point of public network remote control. By upgrading traditional time-driven template control to spatial location-driven control through a spatially triggered preset action mechanism, combined with 3D scene visualization configuration, precise control with actions executed at specific points is achieved. This also reduces reliance on command transmission and ensures operational accuracy even in weak network environments. Furthermore, a global cluster permission arbitration mechanism, leveraging the server's global situational awareness capabilities, automatically handles airspace conflicts and task handover, replacing manual arbitration and takeover. This significantly improves the operational efficiency and fault tolerance of multi-machine clusters, supporting large-scale cluster operation scenarios. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0016] Figure 1 This invention provides a structural framework diagram of a drone remote control communication system that supports preset flight actions and cross-device authorization. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] This invention discloses a drone remote control communication system that supports preset flight maneuvers and cross-device authorization, such as... Figure 1 As shown, it includes: drone terminal, mobile terminal, and network server; The drone is equipped with an embedded system including a wireless communication module that accesses the public network via a mobile communication network, and an optional configuration: a WebSocket client for the embedded system; The mobile app has a flight control device installed. This app integrates a WebSocket client and communicates with the server via the WebSocket protocol. The flight control app also integrates a 3D scene rendering module, a preset action management module, and a device authorization module. The 3D scene rendering module relies on the mobile phone hardware and software system to construct a 3D spatial scene mapping the spatial position of the drone within the flight control APP; the preset action management module is used to generate, edit, store and call preset flight action templates, which are a set of serialized flight control commands; the device authorization module works in conjunction with the server's authorization management unit to grant, revoke and switch drone control permissions between different mobile devices. The server is a WebSocket server deployed on the public network, used for authentication and session binding, and for bidirectional real-time data forwarding between the APP and the drone. The server is equipped with an authorization management unit and a template storage unit.

[0019] The server has a built-in intelligent agent system module, and it creates a functional image / information image of the user's mobile flight control APP user settings and stored data (including flight action templates).

[0020] Users can bypass the mobile app and directly call the server platform's service hotline to conduct two-way voice communication with the intelligent agent, issuing flight control commands via voice call, and the server directly remotely controls the drone.

[0021] In addition, if the server platform's intelligent system detects an anomaly in the drone's flight control communication, such as the mobile flight control APP being terminated or communication with the mobile flight control APP being abnormal or interrupted, the intelligent agent system will directly call the user's mobile phone to interact with the user via voice, and the drone's flight control will continue.

[0022] Furthermore, if an anomaly occurs—such as an interruption in communication between the server and the user—the server's intelligent agent system will take over the flight control of the user's drone, which is still in flight (entering intelligent agent working mode), also known as anomaly management. The intelligent agent system will control the drone's flight or return it to home based on the user's information on the platform. Once normal communication is restored at the user's end, the system will switch back from intelligent agent working mode to user mode (the default working mode).

[0023] At the same time, the intelligent agent system will also report abnormal information to the server platform administrator, who will then decide whether to manually take over (switch to manual takeover mode) based on the situation.

[0024] In a specific implementation, a drone remote control communication system that supports preset flight maneuvers and cross-device authorization includes the following system infrastructure and underlying communication mechanism: 1.1 Three-Tier Infrastructure The system consists of three layers: drone terminal, mobile terminal, and server. It establishes a full-duplex connection based on the WebSocket protocol and realizes bidirectional real-time data interaction through a public network server relay.

[0025] On the drone side: the core is an embedded system, integrating a mobile communication module and a main control chipset, running a WebSocket client program. The embedded system connects to the flight control system and various sensors via a serial port to achieve command reception, status feedback, and flight control. The mobile communication module supports 4G / 5G cellular network access and actively initiates a WebSocket connection request to the public network server after dialing up to the network.

[0026] Mobile terminal: Using smartphones or tablets as hardware carriers, a flight control APP is installed. The APP integrates a WebSocket client SDK. Users generate control commands through the APP interface, which are sent to the server via the WebSocket link, and at the same time receive status data transmitted back by the drone.

[0027] Server: Deployed on a cloud server with a public IP address, it is a standard WebSocket server. Its core functions include verifying the identities of both the drone and mobile devices, binding the sessions between the two ends, enabling bidirectional real-time data forwarding, and solving the problem of NAT traversal between the two ends.

[0028] 1.2 Basic Instruction Set

[0029] The system's basic instruction set includes multiple instruction types to adapt to different control scenarios. All instructions are transmitted via WebSocket links, and the preset flight action templates are ultimately parsed into this basic instruction sequence for execution.

[0030] Single command: A standard or basic command, such as unidirectional flight or single action adjustment.

[0031] One-way command group: The mobile app sends a series of multiple commands in succession. The drone receives the commands, stores them temporarily in the command buffer, and executes them sequentially. This is suitable for low-risk continuous action scenarios. One-way command groups support both factory-preset command group templates and user-defined preset command groups.

[0032] Two-way command group: The mobile app sends individual commands sequentially. After each command is executed, the drone sends a feedback signal to the mobile app to confirm the result before sending the next command, until the entire command group has been executed. This is suitable for scenarios with high reliability requirements. The two-way command group also supports both factory preset and user-defined modes.

[0033] Large command: This is a predefined command group identifier. After receiving it, the drone parses and executes the corresponding firmware pre-stored command group. The firmware pre-stored command group is a collection of multiple basic commands.

[0034] Large command customization extension: Allows users to define and pre-set command group identifiers, which are stored in the mobile APP or server and consist of multiple basic command sets; multiple large commands can be customized and pre-grouped into task-level commands.

[0035] Mission-level commands: A complete record of remote control commands executed in the field, stored in a specific mission area and named for reuse and re-execution. Mission-level commands support three execution methods: One method is to automatically send commands sequentially via a mobile app to remotely control the drone's flight; Secondly, the server receives the call request and automatically sends commands sequentially to remotely control the drone's flight. Third, all command sequences are pre-stored in the UAV command buffer at once and executed sequentially. Task-level commands support both safe mode and interactive confirmation mode: safe mode includes command integrity data verification and execution result feedback; in interactive confirmation mode, the UAV receives command feedback, executes it and then provides feedback again, and the mobile APP confirms before sending the next command, ensuring that the flight process is controllable.

[0036] 1.3 Basic Security and Data Mechanisms

[0037] The embedded system on the drone has a built-in safety control module and a flight information recording module to ensure flight safety and data traceability.

[0038] Safety control module: After takeoff, the flight path node recording function is automatically activated to record the UAV's positioning coordinates according to the preset plane distance interval or time interval, forming a flight path node record table; when the communication signal interruption is detected to exceed the first preset time threshold, the UAV is controlled to automatically enter the hovering state; if it is both in the communication interruption state and hovering exceeds the second preset time threshold, the UAV is controlled to automatically return to home according to the flight path node record table.

[0039] Flight Information Recording Module: Records all sensor data during flight in a timeline format and periodically uploads it to the server via WebSocket connection for cloud data backup.

[0040] In terms of communication security, the WSS protocol is used for encrypted transmission between the mobile app and the server, and between the drone and the server. The instructions issued by the mobile app are attached with digital signatures, and the drone executes them after verifying the validity of the signatures, thus preventing data leakage and instruction tampering.

[0041] 1.4 Basic Interaction Modes and Working Modes

[0042] The mobile app offers three basic interaction modes: a simple menu, a graphical interface, and voice commands, which users can switch between as needed. In the simple menu mode, commands are categorized into multi-level cascading menu items and can be sent with a single click. In the voice command mode, a built-in intelligent transliteration module translates speech into text and generates control commands by matching them with a preset command library. In the graphical interface mode, users can intuitively control the interface using directional keys, function keys, and other symbol keys, including eight directional keys such as east, west, south, and north, as well as function keys for ascending, descending, hovering, landing, automatic return to home, and preset flight missions.

[0043] The system has three basic operating modes: Direct access mode: The mobile app directly controls the drone via the server channel, with the server only acting as a data relay.

[0044] Agent mode: The server acts as a proxy to remotely control the drone and execute a preset sequence of commands.

[0045] Hybrid mode: Normally in pass-through mode, the server automatically takes over when the mobile signal is interrupted or the connection drops, and automatically switches back to pass-through mode when the signal is restored.

[0046] In a specific implementation, the 3D scene rendering control module specifically includes: This module relies on the mobile phone's hardware and software system to upgrade two-dimensional control to three-dimensional spatial control without increasing additional hardware costs, thereby improving the intuitiveness and accuracy of control.

[0047] 2.1 Coordinate System Mapping Mechanism

[0048] The core of 3D scene control is to establish an accurate mapping between screen touch operation and the drone's real spatial position, which is achieved through a three-level coordinate transformation: Geographic coordinate system to local coordinate system conversion: The latitude, longitude and altitude of the UAV in WGS84 are converted to the Northeastern Sky (ENU) local rectangular coordinate system with the takeoff point as the origin, so as to keep it consistent with the UAV flight control coordinate system and ensure the physical accuracy of command mapping.

[0049] Local coordinate system to rendering coordinate system: ENU coordinates are scaled and converted to the OpenGL rendering coordinate system to fit the mobile phone screen size; the scaling factor is dynamically adjusted according to the view scaling operation.

[0050] Rendering coordinate system to screen pixel coordinates: After transformation by the view matrix and projection matrix, it is finally mapped to the mobile phone screen pixel coordinates, completing the projection display from three-dimensional space to two-dimensional screen.

[0051] Through reverse coordinate transformation, users can click and drag to generate screen coordinates in a 3D scene, which can accurately calculate the corresponding real geographic coordinates, and then generate target waypoints and flight commands, achieving a WYSIWYG control effect.

[0052] 2.2 Scene Elements and Multi-View Switching

[0053] The 3D scene contains a variety of visualization elements to present flight status and environmental information in all aspects: a lightweight 3D model of the drone, with real-time synchronized position and attitude; satellite maps or elevation terrain base maps to construct a realistic environmental background; colored flight paths and waypoints to distinguish between already flown paths and waiting paths; and auxiliary information such as electronic fences, no-fly zones, obstacle markers, altitude scales, and compasses.

[0054] The system supports four types of views Figure 1 Key switching adapts to different control needs: Top view: A vertical view from directly above, equivalent to a two-dimensional map view, suitable for overall route planning and viewing the overall situation.

[0055] Front view / side view: Horizontal orthogonal perspective, which facilitates observation of height changes and vertical movements, and enables precise control of vertical operations.

[0056] Stereoscopic perspective view: A third-person following view with a sense of spatial depth. It is the default control view and supports free rotation of the observation angle.

[0057] 2.3 Touch Gesture Interaction Logic

[0058] The 3D scene fully supports multi-touch gestures, with different gestures corresponding to different control functions, and the operation conforms to the interaction habits of mobile phone users: Single-finger swipe: Swipe within the scene plane to generate a horizontal movement command. The swipe distance corresponds to the flight distance, and the swipe direction corresponds to the flight direction, achieving smooth directional control.

[0059] Two-finger pinch zoom: zoom in and out of the scene view. It can also be configured to adjust the flight altitude simultaneously. Swipe up with two fingers to raise the altitude and swipe down to lower it, taking into account both view adjustment and flight control.

[0060] Two-finger rotation: Rotate the scene observation view, and can also be configured to simultaneously adjust the drone's heading, achieving linked control of the view and heading.

[0061] Single-finger click: Click anywhere in the scene to generate a target waypoint command, and the drone will automatically fly to that coordinate point to achieve fixed-point flight.

[0062] Long press and drag: Long press and slide to draw a continuous curve. The system will automatically convert the curve into a smooth flight path command, supporting one-click execution of the drawn complete flight path, and quickly completing custom flight path planning.

[0063] All gesture-generated commands undergo parameter validity verification before being sent to the drone, preventing flight risks caused by misoperation. The 3D scene control mode is fully interoperable with the existing menu, button, and voice modes, and can be switched at any time, sharing the same underlying command system.

[0064] In a specific implementation, the preset flight maneuver template includes: This feature is key to improving the reusability and standardization of operations. 3.1 General Data Structure for Templates The preset flight maneuver templates are stored in structured JSON format. Core fields include template ID, template name, template type, applicable aircraft type, total duration, command node array, parameter configuration table, and signature verification value. The command node array is the core of the template; each node contains a node number, command type, command parameters, execution duration, trigger condition, and node feedback flag. Nodes are sorted by number to form a complete maneuver sequence. The parameter configuration table is used for adjustable parameter templates and real-time parameter templates, defining the name, value range, default value, and adjustment step size of adjustable parameters, as well as the adjustment rules and trigger thresholds for real-time parameters.

[0065] 3.2 Three types of preset flight maneuver templates

[0066] (1) Fixed parameter template

[0067] All command nodes and operating parameters in fixed-parameter templates are fixed values ​​and cannot be modified arbitrarily. They are typically created by the original manufacturer or professionals and are suitable for scenarios with a high degree of standardization. A typical application is drone formation performance, where the lighting changes, position movements, and attitude adjustments of each drone are all programmed into command nodes with fixed timing sequences to ensure precise synchronization of multiple drone movements. After the template is created, it is attached with an official digital signature. Users can only call and execute it, and cannot modify the content, ensuring consistency and security of the actions.

[0068] The execution process is as follows: user selects template → APP verifies template signature validity → sends template loading command to drone → after drone verification, loads the command buffer → after receiving execution command, executes node by node according to timeline → after all executions are completed, sends a completion signal back to the mobile phone.

[0069] (2) Adjustable parameter template

[0070] The adjustable parameter template retains a fixed action logic framework, while making key parameters such as flight altitude, flight speed, turning radius, action interval, and working width configurable. After the user adjusts the parameters according to the actual operation requirements, the system automatically recalculates and generates a complete instruction sequence.

[0071] Taking the rectangular route inspection template as an example, the template framework is: takeoff → rectangular cruise → return landing. Open parameters include the rectangle's length and width, flight altitude, cruise speed, and hovering photo point location. After the user inputs the parameters, the app's built-in trajectory planning algorithm automatically calculates the coordinates of each waypoint, flight duration, and turning angle, generating a complete sequence of command nodes. The system has a built-in parameter boundary verification mechanism for each aircraft model. When user-input parameters exceed the aircraft model's performance limits, it automatically prompts for correction, preventing the generation of illegal commands that could lead to flight risks.

[0072] (3) Real-time parameter template

[0073] During the execution of flight maneuvers by the remote-controlled drone, the parameter values ​​of the flight maneuver template are manually entered to control the drone's flight.

[0074] Furthermore, the real-time parameter template can also be set as an intelligent adaptive template, with a preset basic action framework and parameter adjustment rules. During execution, the operating parameters are dynamically adjusted based on the sensor data transmitted back by the UAV in real time, ensuring flight stability and operation quality in complex environments.

[0075] Taking the terrain-following plant protection template as an example, the basic action is a horizontal back-and-forth flight with preset base altitude and speed. The adjustment rules are: adjust the flight altitude in real time based on terrain height sensor data to maintain a constant relative altitude with the ground; increase the flight speed when facing headwinds and decrease the flight speed when facing tailwinds based on wind speed sensor data to maintain a stable ground speed. The execution logic is that the drone's flight control system collects sensor data in real time and dynamically corrects the current action parameters according to preset rules, without requiring intervention from the mobile phone; when key parameters are adjusted beyond the threshold, the system automatically reports to the mobile phone to ensure that the user can perceive the status changes.

[0076] 3.3 Template Execution Mode

[0077] The preset flight maneuver templates are compatible with the execution logic of the basic command system and support two execution modes, which users can choose according to the reliability requirements of the scenario: Fast execution mode: All template instructions are sent to the drone's instruction buffer at once, and the drone executes them automatically in sequence. No confirmation is required on the mobile phone during the execution process. The execution speed is fast and it is suitable for low-risk routine operations.

[0078] Interactive confirmation mode: Templates are sent out node by node. After the drone completes a node, it sends the execution result and status back to the mobile phone. The mobile phone confirms the result before sending out the next node. Critical nodes can be set to force manual confirmation to ensure that high-risk actions are controllable. This mode is suitable for precision operations and high-risk scenarios.

[0079] In a specific implementation, the real-scene flight control operation recording and template includes: This feature enables a one-click conversion from manual flight operations to reusable templates, significantly lowering the barrier to entry for creating customized templates.

[0080] The complete process of converting live-action recordings into templates consists of six stages: recording trigger, data acquisition, data preprocessing, template editing, template storage, and template retrieval.

[0081] 4.1 Recording Trigger and Data Acquisition

[0082] Users can trigger recording by clicking the "Start Recording" button in the flight control app. The system starts a recording thread in the background and simultaneously collects two types of data at a sampling frequency of 20Hz: Control command data: Records every command issued by the user, including command type, parameter value, and timestamp of issuance; supports command recording in all interaction modes, covering all scenarios of 3D scene gesture control, button control, and voice commands.

[0083] Drone status data: Synchronously record the drone's transmitted status data such as position, attitude, speed, battery level, signal strength, and sensor readings, and accurately align them with the command data according to the timestamp.

[0084] During the recording process, the app interface displays information such as recording duration, number of recorded commands, and storage space usage in real time. Users can pause, continue, or stop recording at any time.

[0085] 4.2 Data Preprocessing

[0086] The raw recording data has issues such as noise redundancy, trajectory jitter, and irregular timeline, requiring automated preprocessing to generate a high-quality initial template. Instruction redundancy removal: Filter out duplicate and invalid instructions and instructions that need to be quickly corrected in a short time, merge consecutive instructions of the same type, and reduce the number of instruction nodes.

[0087] Trajectory smoothing: The position trajectory is smoothed using a Bezier curve fitting algorithm to eliminate the jitter of manual control and generate a smooth flight path; at the same time, the speed and altitude change curves are smoothed to avoid sudden acceleration and sudden ascent and descent.

[0088] Timeline normalization: Convert the original relative timestamps into standardized node execution durations, adjust the overall timeline, and make the action rhythm uniform and controllable.

[0089] Outlier removal: Automatically identifies abnormal data points that deviate significantly from the normal trajectory and corrects them using interpolation algorithms to prevent them from being mistakenly entered into the template.

[0090] After preprocessing, the system automatically generates an initial template and provides a flight trajectory preview and an action duration preview.

[0091] 4.3 Template Editing and Storage

[0092] Users can make secondary fine-tuning adjustments to the initial template in the template editing interface: it supports timeline cropping to extract action segments within a specified time period; it supports node editing to insert, delete, and modify any instruction node and parameter; it supports parameter conversion to convert fixed parameter nodes into adjustable parameters and set their value range, thus upgrading a fixed template to an adjustable parameter template; and it supports node splitting to break down long actions into multiple sub-nodes to adapt to the segmented execution requirements of interactive confirmation modes.

[0093] After editing, users can name the template and set its category. They can choose to save it locally on their mobile phone or upload it to the cloud template storage unit on the server. Templates uploaded to the cloud can be set as private or shared with designated accounts to achieve sharing and reuse within the team.

[0094] In a specific implementation, the cross-device authorization and multi-device collaboration mechanism includes: 5.1 Cross-device authorization mechanism The authorization system is designed based on the RBAC (role-based access control) model and is implemented collaboratively by the mobile device authorization module and the server authorization management unit.

[0095] The permission levels are divided into four levels to adapt to different use cases: Administrator privileges: Possess full operational permissions, including the ability to modify device configurations, manage authorizations, and revoke permissions. Typically, this is the device owner's privileges.

[0096] Full control permissions: Allows all flight controls, template access, and parameter modification, but does not manage authorization or core device configurations.

[0097] Task execution permissions: Standardized tasks can only be executed by calling preset templates. Manual control and modification of template parameters are not allowed.

[0098] Observation permissions: Users can only view the drone's real-time status and transmitted images; they cannot issue any commands.

[0099] Complete authorization process: On the main control phone with administrator privileges, select the target drone in the APP authorization management interface, enter the authorized account, select the permission level, set the validity period and effective time, and submit the authorization request.

[0100] The server authorization management unit verifies the applicant's permissions. Once the verification is successful, it generates an encrypted JWT authorization token. The token is bound to the drone ID, the authorized account, the permission level, and the expiration information.

[0101] The server sends an authorization notification to the authorized account, and the authorized mobile phone can see the authorized drone in the device list after logging in.

[0102] When an authorized client initiates a connection, the server verifies the validity of the token. If the verification is successful, the server grants the corresponding permissions and establishes a session binding.

[0103] The system supports a secure mutual exclusion mechanism. When the single-end control option is enabled, only one online full control permission is allowed for the same drone at a time. When a new end with full control permission connects, the original control end is automatically downgraded to observation permission and receives a notification, preventing command conflicts caused by multiple ends issuing commands simultaneously. Permissions automatically expire upon authorization, and the server automatically revoks permissions.

[0104] 5.2 Seamless switching mechanism between primary and backup mobile phones

[0105] To address the pain points of frequent power outages and signal interruptions for mobile phones during outdoor operations, the system is designed with a seamless primary / backup phone switching mechanism to improve system redundancy and reliability. Users can configure one primary control phone and multiple backup phones for the drone under an administrator account, and set their priority. The server monitors the connection status of each online device in real time via WebSocket heartbeat packets.

[0106] The switching logic is as follows: Under normal circumstances, the master phone has full control permissions, while the backup phone is in observation permission or offline state.

[0107] If the server does not receive a heartbeat packet from the master control mobile phone for 3 consecutive seconds, it determines that the master control terminal is abnormal.

[0108] The server immediately sends a hovering command to the drone and simultaneously queries online backup mobile phones according to priority.

[0109] A takeover notification is pushed to the highest priority backup phone. Upon confirmation, its permissions are automatically upgraded to full control permissions, completing the permission switch.

[0110] Once the switch is complete, the drone is released from hover, and the backup phone can seamlessly continue operating the task.

[0111] After the master phone reconnects, it automatically reverts to observation privileges. The administrator needs to manually switch back to master control to avoid automatic privilege grabbing that could lead to operational chaos.

[0112] 5.3 Single App Multi-Device Management Mechanism

[0113] A single flight control app supports binding and managing multiple drones, adapting to swarm operation scenarios. Core functions include: Device list management: All bound drones are displayed in a list format, showing key information such as online status, battery level, and location. It supports management by group and by region.

[0114] Quickly switch controls: Click on any online drone to switch to the current control object, and the interface will update the drone's status data and 3D scene in real time.

[0115] Batch command issuance: Multiple drones can be selected to form a formation and the same command or the same preset template can be issued in batches. The server will forward the command to each drone to achieve synchronous execution.

[0116] Cluster 3D View: Displays the positions and flight paths of all selected drones simultaneously in a 3D scene, providing an intuitive understanding of the overall operational status of the cluster.

[0117] For multi-drone management scenarios that require physical isolation, it supports the installation of multiple independent flight control APP clones on the same mobile phone. Each clone logs into a different account and binds to a different drone, with complete data isolation and no interference between them.

[0118] In a specific implementation, the link quality adaptive instruction scheduling mechanism includes: Based on real-time link quality awareness, the adaptive switching technology for command transmission mode solves the core technical problem of balancing transmission efficiency and control reliability under dynamic fluctuations in public network links.

[0119] 6.1 Module Architecture and Core Principles

[0120] The inherent characteristics of public 4G / 5G cellular networks are that link quality dynamically changes with environment, distance, and base station load. Round-trip time (RTT) can fluctuate between tens of milliseconds and several seconds, and packet loss rate can also spike instantaneously. Traditional fixed-mode command transmission strategies have inherent flaws: the one-way continuous transmission mode suffers from a large number of command losses and out-of-order delivery when the network is poor, leading to flight control failure; the two-way confirmation mode wastes bandwidth and has low command transmission efficiency when the network is good, making smooth control impossible.

[0121] This mechanism is implemented through the collaboration of the link quality detection module on the mobile phone and the dynamic buffer scheduling unit on the drone. The core principle is to evaluate the link quality in real time and automatically match the optimal command sending mode and buffer configuration according to the quality level, so as to always maintain the best balance between efficiency and reliability in different network environments.

[0122] 6.2 Link Quality Quantitative Assessment Method

[0123] The link quality detection module uses WebSocket heartbeat packets as a basis and combines them with business data packets for multi-dimensional sampling to generate a comprehensive link quality score (QoS score, 0-100 points). The evaluation indicators include: Round-trip time (RTT): The mobile device sends a timestamped heartbeat packet to the server, and calculates the round-trip time after receiving the response packet. The sliding average of the last 10 times is taken as the current RTT; the smaller the RTT, the higher the score.

[0124] Packet loss rate: The packet loss rate is calculated by comparing the number of sent command packets to the number of received acknowledgment packets in the last 30 seconds; the lower the packet loss rate, the higher the score.

[0125] Latency jitter: The variance of consecutive RTTs, reflecting network stability; the lower the jitter, the higher the score.

[0126] The system sets two threshold levels as the basis for determining mode switching: First threshold (high-quality link): QoS score ≥ 80 points, corresponding to RTT ≤ 100ms, packet loss rate ≤ 1%, jitter ≤ 20ms; Second threshold (inferior link): QoS score ≤ 40 points, corresponding to RTT ≥ 300ms, packet loss rate ≥ 5%, jitter ≥ 80ms; The link between the two is a general link.

[0127] To avoid frequent mode switching caused by instantaneous network fluctuations, a dejitter mechanism is added: mode switching is only triggered when the link quality remains stable in the new level range for 3 seconds, thus preventing the ping-pong effect.

[0128] 6.3 Logic for Dynamic Switching of Command Sending Mode

[0129] The system automatically switches between three command transmission modes based on the link quality level. All switching processes are transparent to the user and require no manual intervention. High-quality link → One-way command group mode: When the network conditions are good, a one-way continuous command sending strategy is adopted. Commands are sent continuously without the need for confirmation each time, maximizing the efficiency of command sending and achieving a smooth control experience comparable to a local remote control; suitable for open areas and scenarios with good base station signals.

[0130] Normal Link → Large Command Preloading Mode: When network conditions are normal, continuous actions are packaged into large commands and sent to the drone's cache at once. The drone executes them locally in sequence, reducing the number of network interactions and mitigating the impact of occasional packet loss. This ensures a certain level of execution efficiency while improving resilience to fluctuations.

[0131] Poor link → Two-way command group mode: When the network condition is poor, it automatically degrades to the successive confirmation mode. After sending each command, it waits for the drone to respond with the execution result before sending the next command. Although the efficiency is reduced, it ensures that each command arrives reliably and avoids flight abnormalities caused by command loss. It is suitable for scenarios with signal obstruction and network congestion.

[0132] When switching modes, the system will first complete the currently executing instruction sequence and then smoothly switch to the new mode, without causing instruction interruption or sudden changes in flight.

[0133] 6.4 Dynamic Buffer Scheduling Mechanism

[0134] The command buffer on the drone side acts as a reservoir for command execution, and its water level threshold directly affects the resilience to fluctuations and control latency. The higher the water level, the stronger the resilience to network interruptions, but the greater the command buffer latency; the lower the water level, the smaller the control latency, but the weaker the resilience to fluctuations.

[0135] The dynamic buffer scheduling unit dynamically adjusts the high and low water levels of the buffer based on the link quality score: High-quality links: 5 low-water level links, 15 high-water level links, maintain shallow caching, reduce control latency, and improve control responsiveness; Typical links: 10 low-water level links, 30 high-water level links, medium buffer, balancing latency and resistance to fluctuations; Low-quality links: 20 low-water level links, 50 high-water level links, deep cache, prioritizing instruction continuity and resisting network fluctuations and short-term interruptions.

[0136] The buffer zone water level is adjusted gradually, with each adjustment not exceeding 20% ​​of the current water level, to avoid sudden changes in water level that could disrupt the execution rhythm.

[0137] 6.5 Technical Effects and Measured Data

[0138] Through comparative testing, under simulated network fluctuations, this mechanism reduces command loss rate by 87% and flight anomalies by 92% compared to the fixed one-way mode; compared to the fixed two-way mode, command issuance efficiency is improved by 210%. In actual field operation scenarios, it can effectively offset instantaneous network fluctuations caused by base station switching and terrain obstruction, significantly improving the stability and control feel of public network remote control. This is one of the core technological advantages that distinguishes this system from ordinary public network drone control solutions.

[0139] In a specific implementation, the space-triggered preset action mechanism includes: The spatially triggered flight motion control technology in a three-dimensional scene upgrades the traditional time-driven template to a spatial position-driven one, achieving high-precision, low-dependency intelligent operation control.

[0140] 7.1 Core Issues and Innovative Ideas

[0141] Traditional preset flight action templates are all time-series driven, that is, the actions are executed sequentially according to preset time points. The drawback is that the actual flight speed of the drone is affected by wind speed, wind direction and load, which will cause the action execution position to deviate from the expected position. It cannot meet the high-precision operation requirements of reaching the precise position to perform specific actions, such as taking pictures directly above the power pole in power inspection, taking pictures at the designated waypoint in surveying, and automatically turning on the nozzle when entering the field in plant protection.

[0142] This mechanism changes the triggering condition for preset actions from time arrival to spatial condition fulfillment. Users can visually configure the triggering area and triggering conditions in a 3D scene and bind the corresponding preset actions. The drone locally determines its own spatial state in real time and automatically triggers the action when the conditions are met, without the need for real-time instructions from the mobile phone or server. This improves the triggering accuracy and reduces the dependence on network links.

[0143] 7.2 Visual Configuration of Spatial Triggering Conditions

[0144] Leveraging the 3D scene rendering module, users can intuitively complete all trigger configurations on a 3D map without writing code or inputting coordinates. Trigger area drawing: Supports drawing trigger areas of various geometric shapes in a 3D scene, including circular areas, arbitrary polygonal areas, 3D cube areas, columnar areas, etc.; the drawing method is to click to add vertices and drag to adjust the shape, what you see is what you get.

[0145] Trigger type settings: Supports multiple trigger logics, including: Entry Trigger: Triggered when a drone enters the designated area from the outside; Departure Trigger: Triggered when the drone leaves the designated area from within; Altitude trigger: Triggered when the drone reaches a specified altitude range; Attitude trigger: Triggered when the drone's attitude reaches a specified angle; Composite triggering: Triggered by multiple conditions combined logic (AND, OR, NOT).

[0146] Action binding: Binds preset flight action templates to trigger conditions, and allows setting the number of triggers (single trigger, loop trigger), trigger delay, trigger prerequisites, etc.

[0147] The configured spatial triggering rules can be saved as part of a preset template or saved independently as a general rule library for reuse in different job templates.

[0148] 7.3 Local Decision Logic on the UAV End

[0149] The spatial determination unit runs on the embedded system of the UAV, operating independently of the communication link, ensuring normal triggering even in the event of a network outage. The core determination process is as follows: Data input: Real-time acquisition of GNSS positioning data (latitude, longitude, altitude), IMU attitude data (pitch, roll, heading), and barometric altitude data, followed by multi-source fusion filtering to improve spatial state accuracy.

[0150] Coordinate transformation: Converts the current geographic coordinates to local coordinates in the same coordinate system as the triggering area, unifying the calculation benchmark.

[0151] Spatial determination algorithm: Planar region determination: The ray method is used to determine whether the current position point is within the polygonal region; 3D region determination: Based on the height range, determine whether it is inside the cube; Boundary stabilization: Set a 2-meter boundary hysteresis zone to prevent frequent triggering caused by drone shaking at the boundary.

[0152] Action execution: After determining that the triggering conditions are met and the debouncing is successful, load the bound preset action instruction sequence from the local cache, insert it into the instruction buffer and execute it first; after execution, update the triggering status to avoid repeated triggering.

[0153] Since the decision-making logic is executed entirely locally on the drone, the triggering delay is only in milliseconds, which is much higher than the second-level delay of sending commands over the network, thus greatly improving the timing and positional accuracy of the action triggering.

[0154] 7.4 Typical Application Scenarios

[0155] Taking power line inspection as an example: Operators mark the cylindrical inspection area of ​​each tower in a 3D scene, bind a preset action template of hovering → gimbal tilt scan → take 3 photos → continue patrolling, generating an inspection template for the entire line. After takeoff, the drone flies along the route, automatically hovering and taking photos each time it reaches the trigger area of ​​a tower, then automatically continuing patrolling. Compared to traditional time-series templates, the spatial triggering mechanism reduces the photo position error from 5-10 meters to within 1 meter, and eliminates the need for precise manual control of waypoint timing, significantly improving operational efficiency and quality. Even if a network interruption occurs midway, the drone can still complete all triggering actions normally, preventing operational failure due to network outages.

[0156] In a specific implementation, the dynamic permission arbitration mechanism for multi-drone swarms includes: The cluster airspace permission arbitration and task handover technology based on global situational awareness solves the airspace conflict and fault tolerance problems in multi-machine cluster operations.

[0157] 8.1 Module Positioning and Core Value

[0158] As the scale of drone swarm operations expands, the simultaneous operation of multiple drones in the same airspace greatly increases the risk of overlapping flight paths and airspace conflicts. Furthermore, when a single drone malfunctions, manually assigning a backup drone to take over the task is inefficient and prone to errors. Traditional distributed multi-drone systems lack a global control node, relying solely on individual drones for obstacle avoidance and failing to provide global scheduling at the permission and task levels.

[0159] This system leverages the server's global data aggregation advantage and adds a cluster arbitration unit to uniformly arbitrate and dynamically schedule airspace permissions and task permissions within the cluster from a global perspective. This enables automatic conflict avoidance and automatic fault takeover, significantly improving the security and efficiency of cluster operations.

[0160] 8.2 Airspace Conflict Authority Arbitration Mechanism

[0161] (1) Conflict detection

[0162] The cluster arbitration unit aggregates three core data points from all online drones in real time: current location, preset future flight path, and flight speed and heading. Based on this data, a spatiotemporal two-dimensional conflict detection algorithm is employed. Spatial dimension: Generate a safe bounding box with a radius of 50 meters for each drone, and predict the flight trajectory bounding box within the next 30 seconds; Time dimension: Calculate the time difference between two drones arriving at the same spatial location; if it is less than 10 seconds, it is considered that there is a risk of conflict. If both conditions of spatial overlap and temporal proximity are met, it is determined to be a conflict pending arbitration.

[0163] (2) Priority determination

[0164] In the event of a conflict, the arbitration unit will comprehensively determine the priority of the two drones based on the following dimensions: the drone with higher priority will retain its original flight path and operational rights, while the drone with lower priority will receive an avoidance instruction: Task priority: The preset task level (e.g., emergency rescue > routine inspection > general surveying) has the highest weight; Permission levels: Permission levels for controlling users (Administrator > Operator > Temporary Authorization); Task progress: Prioritize tasks with high completion rates to avoid interruptions to tasks that have already incurred significant costs; Remaining battery power: Prioritize aircraft with low battery power to reduce flight time and lower the risk of crash.

[0165] Each dimension is assigned a weighting coefficient, and a weighted average score is calculated to obtain a comprehensive priority score. The higher the score, the higher the priority.

[0166] (3) Arbitration enforcement process

[0167] Upon detecting a conflict risk, the arbitration unit immediately calculates the priority ranking; Issue hovering avoidance commands to low-priority drones, and simultaneously push conflict warnings and estimated waiting times; Track the location of high-priority drones in real time, and issue a resumption command to low-priority drones after they have completely left the conflict airspace. The entire arbitration process is recorded in logs and uploaded to the auditing system, enabling full traceability of the conflict.

[0168] This mechanism upgrades the traditional single-machine passive obstacle avoidance to global permission scheduling, fundamentally avoiding the security risks of multiple machines encountering each other at close range, while ensuring the continuity of high-value tasks.

[0169] 8.3 Automatic transfer mechanism for abnormal task permissions

[0170] When a drone in the cluster experiences an anomaly such as communication interruption, low battery warning, or equipment failure and is unable to continue performing its mission, the cluster arbitration unit can automatically hand over the mission without manual intervention. Anomaly detection: Real-time detection of abnormal drone status through multiple methods such as heartbeat loss, battery threshold alarm, and fault feedback; Task snapshot generation: Upon detecting an anomaly, immediately capture data such as the drone's current task progress, unexecuted flight paths, and operational areas to generate a task snapshot; Backup drone matching: Among the backup drones in the same group, the optimal takeover drone is selected based on the principles of closest distance, sufficient power, and highest priority; Permissions and Task Transfer: Transfer the task execution permissions and task snapshot of the abnormal drone to the standby drone, and issue the instruction "Go to the takeover point → Continue to execute the task" to the standby drone; Status Update: Updates the cluster task status, pushes a handover notification to the administrator, and triggers automatic return or hovering in place for abnormal drones.

[0171] Tests have shown that the entire process from detecting an anomaly to the standby machine taking over the task can be completed within 5 seconds, which is much faster than the minute-level response time of manual operation. This can minimize task interruption time and ensure the continuous and stable operation of large-scale cluster operations.

[0172] 8.4 Cluster Management Effectiveness

[0173] In a cluster plant protection operation test involving 10 drones, the introduction of this mechanism reduced the number of manual interventions for airspace conflicts to zero and reduced task delays caused by conflicts by 100%. The task interruption time caused by a single drone malfunction was shortened from an average of 12 minutes to 8 seconds, and the overall operation efficiency of the cluster was improved by more than 35%, fully demonstrating the core value of the global authority arbitration mechanism.

[0174] In a specific embodiment, an agricultural plant protection platoon operation is used as the complete scenario to demonstrate the entire process of the system's operation, intuitively showcasing the synergistic value of each innovative function.

[0175] Scenario: A plant protection team has 5 plant protection drones and needs to carry out unified plant protection operations on 1,000 acres of farmland. The team consists of 1 supervisor and 3 operators. The terrain of the work area is slightly undulating, and there are local blind spots in the field's mobile phone signal.

[0176] Phase 1: Preliminary Preparation and Template Creation

[0177] The supervisor uses a mobile app in 3D scene mode to import farmland satellite maps, draws the boundaries of the work area in the 3D view, and sets parameters such as flight altitude of 3 meters, flight speed of 5 m / s, and spray width of 6 meters. The system automatically generates an adjustable parameter plant protection template for reciprocating flight paths.

[0178] The supervisor led the operators to the site, selected a typical field for manual test flights, and activated the live recording function to record the complete operation of manually navigating around obstacles and turning at the edge of the field. After recording, the data was edited and optimized to generate plant protection templates for complex fields, which were then uploaded to the cloud server and shared with all team accounts.

[0179] In the authorization management interface, the supervisor authorizes the task execution permissions of 5 drones to 3 operator accounts, with the authorization period from 8:00 to 18:00 on the same day; at the same time, a backup mobile phone is set up for each drone to ensure seamless takeover in case of signal interruption.

[0180] Phase 2: Formation Operation Execution

[0181] Three operators arrived at their respective work areas, each carrying a spare mobile phone; they opened the APP, logged into their accounts, and could see a list of authorized drones.

[0182] The operator selects the drone to be responsible for, calls the plant protection template for the corresponding plot, confirms the parameters, and starts the operation with one click; the drone takes off automatically and performs plant protection operations according to the preset route.

[0183] During the operation, the operator can view the drone's location, remaining drug quantity, and flight trajectory in real time through a 3D scene, intuitively grasping the progress of the operation; if temporary obstacles are encountered, the operation can be paused at any time and manually corrected.

[0184] When a drone flew into a signal blind spot, the connection to the main control mobile phone was interrupted. After the server detected the anomaly, it immediately triggered the backup mechanism, and the drone automatically hovered. At the same time, a takeover notification was pushed to the operator's backup mobile phone, and the backup mobile phone automatically took over control. After the drone flew out of the signal blind spot, the operation continued without interruption.

[0185] Phase 3: Assignment Completion and Review

[0186] After all drones have completed their operations, they automatically return to base, and the flight data and operation logs are automatically uploaded to the server.

[0187] Supervisors can view the operational trajectories, dosage, and duration of all drones in the backend and generate operational reports.

[0188] The optimized flight path from this operation will be saved as a new template for subsequent operations in the same area, gradually building up a standardized operation library.

[0189] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0190] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A remote control communication system for unmanned aerial vehicles (UAVs) that supports preset flight maneuvers and cross-device authorization, characterized in that, include: Drone terminals, mobile terminals, and web servers; The drone is equipped with an embedded system including a wireless communication module that accesses the public network via a mobile communication network, and an optional configuration: a WebSocket client for the embedded system; The mobile app has a flight control device installed. This app integrates a WebSocket client and communicates with the server via the WebSocket protocol. The flight control app also integrates a 3D scene rendering module, a preset action management module, and a device authorization module. The 3D scene rendering module relies on the mobile phone hardware and software system to construct a 3D spatial scene that maps the spatial position of the drone within the flight control APP. The preset action management module is used to generate, edit, store and call preset flight action templates, which are serialized flight control command sets; The device authorization module works in conjunction with the server's authorization management unit to grant, revoke, and switch drone control permissions between different mobile devices; The server is a WebSocket server deployed on the public network, used for authentication and session binding, and for bidirectional real-time data forwarding between the APP and the drone. The server is equipped with an authorization management unit and a template storage unit.

2. The UAV remote control communication system supporting preset flight actions and cross-device authorization according to claim 1, characterized in that, The 3D scene rendering module supports one-click switching between front view, side view, top view and stereoscopic perspective view; it supports touch screen interactive operation, including single-finger swipe to control the horizontal displacement of the drone, two-finger zoom to adjust the view scale and flight altitude, two-finger rotation to adjust the drone's heading, long press and drag to draw continuous flight trajectory and automatically generate corresponding command sequences; the control operations in the 3D scene are converted into drone flight control commands through coordinate system mapping and sent to the drone terminal for execution via WebSocket link.

3. The UAV remote control communication system supporting preset flight actions and cross-device authorization according to claim 1, characterized in that, The preset flight action templates managed by the preset action management module are divided into three types: Fixed parameter template: The action sequence and all running parameters are fixed, and the entire set of flight actions are executed directly when called; Adjustable parameter template: The core action framework is fixed, and key operating parameters can be customized by users. After the parameters are modified, their legality is automatically verified and a new instruction sequence is generated. Real-time parameter template: During the execution of flight maneuvers of the remote-controlled drone, the parameter values ​​of the flight maneuver template are manually entered to control the drone's flight.

4. A drone remote control communication system supporting preset flight actions and cross-device authorization according to claim 3, characterized in that, The preset action management module also includes a real-scene flight control recording unit, which is used to enable the operation recording function and record all remote control commands and drone status data during the user's manual operation process in real time. After recording, the system supports editing operations such as trimming, modifying parameters, and adding or deleting nodes in the instruction sequence, and saves it as a preset flight action template, transforming real-world flight control operations into reusable templates.

5. A drone remote control communication system supporting preset flight actions and cross-device authorization according to claim 1, characterized in that, The device authorization module allows the master control phone to authorize the control permissions of a specified drone to the flight control APP of another phone. The authorization process generates an encrypted authorization token through the server authorization management unit. The authorization scope includes the permission level, validity period and scope of executable operations. When the security mutual exclusion option is enabled, only one mobile terminal is allowed to have full control permissions for a drone at the same time. When the new authorization takes effect, the original control terminal's permissions are automatically suspended.

6. A drone remote control communication system supporting preset flight actions and cross-device authorization according to claim 1, characterized in that, include: Multiple drones can be managed and controlled simultaneously through a single mobile flight control app. Different drone device IDs can be matched through a drop-down list, and group control and batch command issuance are supported. Multiple independent flight control app clones can be installed on the same mobile phone, and each flight control app clone is matched and controlled by the corresponding drone.

7. A drone remote control communication system supporting preset flight actions and cross-device authorization according to claim 1, characterized in that, include: The same drone can be bound to multiple mobile devices, with a master control phone and at least one backup phone set. The server monitors the connection status of the master control phone in real time. When the master control phone malfunctions or the network signal is interrupted, it automatically switches control to the backup phone for seamless takeover.

8. A drone remote control communication system supporting preset flight actions and cross-device authorization according to claim 1, characterized in that, The server has a built-in intelligent agent system for performing functional mapping / information mapping of user settings and stored data from the user's mobile flight control APP; It also includes a second communication channel, through which users can bypass the mobile flight control APP and directly dial the server platform's service phone number / SMS to conduct two-way voice communication with the intelligent agent system, issue flight control commands through voice calls, and the intelligent agent system directly remotely controls the drone; If the intelligent agent system detects an anomaly in the drone's flight control communication, the intelligent agent system will call the user's mobile phone and interact with the user directly via voice, and the drone's flight control will continue. If communication between the server and the user is interrupted, the intelligent agent system will take over the flight control of the drone that is still in flight for the user. The intelligent agent system will control the drone's flight or return it to home based on the information image of the user on the platform.

9. A drone remote control communication system supporting preset flight actions and cross-device authorization according to claim 1, characterized in that, The server is also equipped with a cluster arbitration unit for airspace access management and task priority arbitration of multiple drone clusters. The cluster arbitration unit gathers the location, preset flight path and task priority data of all online drones in real time. When it detects that the future flight paths of two or more drones have overlapping airspace conflicts, it arbitrates according to the task priority level and equipment permission level, issues a hovering avoidance command to the low-priority drone, and restores the operation permission of the low-priority drone after the high-priority drone leaves the conflict airspace. When a communication or equipment malfunction is detected in a drone, the cluster arbitration unit can automatically transfer the unfinished task permissions and task snapshot to the highest priority standby drone in the same group for seamless task succession.

Images