An on-board intelligent computing and communication device and method

Abstract

The application relates to the technical field of unmanned aerial vehicle intelligent flight, in particular to an airborne intelligent computing and communication device and method, the device comprising a heat dissipation assembly, a core computing module, a carrier plate and an extension plate which are arranged in layers; the core computing module processes flight control data flow and video flow as an edge computing core; the carrier plate is integrated with a first communication module, an unmanned aerial vehicle interaction interface and a storage module; the extension plate is integrated with a second communication module; the two communication modules respectively communicate with the core computing module through virtual network ports to establish a heterogeneous double-link diagram transmission network. The method obtains data through the unmanned aerial vehicle interaction interface, identifies communication module hardware nodes and maps virtual network ports, and establishes a double-link communication network through a dialing program to perform data transmission. The application realizes integrated integration of high-computing-power edge computing and double-link diagram transmission, and solves the problem of electromagnetic interference through antenna isolation and cable shielding structure.

Description

Technical Field

[0001] This application relates to the technical field of intelligent flight of unmanned aerial vehicles (UAVs), and in particular to an airborne intelligent computing and communication device and method. Background Technology

[0002] With the rapid development of drone technology, drones have been widely used in various complex scenarios such as intelligent patrol, geological exploration, fire early warning, and security reconnaissance. In these applications, drones typically need to transmit high-definition images they capture to a ground control center in real time and receive complex flight control commands from the ground.

[0003] Currently, most existing UAV-borne image transmission devices use a single 4G network module to communicate with ground stations. As application scenarios become more complex, the demands on airborne computing power and image transmission quality are increasing, and existing technologies are gradually revealing significant shortcomings. On the one hand, a single network is highly susceptible to interference in the field or complex electromagnetic environments, resulting in low communication rates, high latency, and frequent image transmission lag or even aircraft disconnection when flying over long distances. On the other hand, ordinary lightweight image transmission devices lack edge computing capabilities, making it impossible to deploy high-performance computing models on the airborne end for image preprocessing or autonomous path planning.

[0004] More seriously, in order to improve communication bandwidth and computing power, current methods often attempt to attach more powerful communication transmission modules and computing motherboards to drones. However, the high-frequency electromagnetic radiation generated by high-power radio frequency communication modules during operation can cause crosstalk along communication cables or directly radiate into space to the drone's own sensitive electronic components (such as RTK real-time dynamic differential positioning modules). This can lead to difficulties in satellite acquisition and positioning drift, seriously threatening the drone's flight control safety. In addition, the addition of functional modules inevitably results in bulky equipment and cumbersome interfaces, making it difficult to meet the strict payload and aerodynamic shape limitations of small and medium-sized drones.

[0005] Therefore, the technical problem that needs to be solved is how to provide a solution that can meet the requirements of high bandwidth and high stability data transmission, provide powerful airborne computing power, and fundamentally solve the electromagnetic interference caused by complex integrated mount equipment to the native navigation system of UAVs. Summary of the Invention

[0006] To address the technical problems of existing UAV-borne image transmission devices, such as susceptibility to interference in single-network communication, lack of edge computing capabilities, and poor electromagnetic compatibility due to functional integration, the purpose of this application is to provide an airborne intelligent computing and communication device and method.

[0007] In a first aspect, the airborne intelligent computing and communication device provided in this application adopts the following technical solution: It includes a stacked heat dissipation assembly, a core computing module, a carrier board, and expansion boards; The heat dissipation component is located on one side of the core computing module and is used to actively dissipate heat from the core computing module. The core computing module, as the edge computing core, is communicatively connected to the carrier board and the expansion board, and is used to process UAV flight control data streams and video streams. The carrier board integrates a first communication module, a drone interaction interface, and a storage module; the drone interaction interface is used to physically connect with the drone to obtain power and data from the drone. The expansion board integrates a second communication module, and the expansion board is electrically connected to the carrier board. The first communication module and the second communication module communicate with the core computing module through virtual network ports to establish a heterogeneous dual-link image transmission network.

[0008] By adopting the above technical solution, the heat dissipation components, core computing module, carrier board, and expansion board are stacked, achieving miniaturization and lightweight design of the device, meeting the payload and size requirements of the UAV. The core computing module, as the edge computing core, can realize high-performance model deployment and data processing on the airborne end. The carrier board integrates a first communication module, and the expansion board integrates a second communication module. The two communication modules communicate with the core computing module through virtual network ports, establishing a heterogeneous dual-link image transmission network, which effectively improves the bandwidth, stability, and anti-interference capability of data transmission. When the signal of one link is unstable, communication can be guaranteed through the other link, fundamentally solving the problem of insufficient reliability of single network communication.

[0009] Optionally, the UAV interaction interface is a PSDK interface; the carrier board obtains power from the UAV through the PSDK interface, communicates with the UAV's flight control system through the asynchronous serial communication UART channel in the PSDK interface, and obtains the UAV's video stream through the virtual network port mapped by the USB channel.

[0010] By adopting the above technical solution, the PSDK interface is used to achieve multi-functional multiplexing of power supply, flight control communication and video stream acquisition. On the one hand, there is no need for additional power cables, and power is obtained directly from the UAV, which simplifies cable connection and device structure. On the other hand, low-latency communication with the flight control system is achieved through the UART channel, and high-bandwidth video stream transmission is achieved through the virtual network port mapped by the USB channel. The multi-channel characteristics of the PSDK interface are fully utilized, the number of peripheral interfaces is reduced, and it helps to make the whole machine lighter.

[0011] Optionally, the first communication module is a 5G communication module operating in the public network frequency band; the second communication module is a 1.4G communication module operating in the private network frequency band.

[0012] By adopting the above technical solutions, the 5G communication module operates in the public network frequency band, providing high-speed, low-latency data transmission capabilities, which can meet the high bandwidth requirements of high-definition video transmission; the 1.4G communication module operates in the private network frequency band. Private network base stations have limited coverage areas but high signal security and confidentiality, making them less susceptible to interception or interference from the outside world, making them more suitable for application scenarios such as public security and military units that have strict requirements for information security; the two form a heterogeneous dual link of public network and private network, which not only ensures the communication speed and coverage, but also takes into account the communication security and reliability.

[0013] Optionally, the device further includes an electromagnetic interference protection structure, the electromagnetic interference protection structure comprising: The antenna isolation structure includes a first communication module connected to a first antenna feed line and a second communication module connected to a second antenna feed line. The first antenna feed line and the second antenna feed line are physically separated and arranged within the housing of the device to avoid crossing or overlapping.

[0014] By adopting the above technical solution, the first antenna feed line and the second antenna feed line are physically separated and arranged inside the casing, avoiding the crossing and overlap of the two sets of antenna feed lines. This effectively reduces the mutual coupling interference and radiated crosstalk between the 5G communication module and the 1.4G communication module caused by the proximity or crossing of the antenna feed lines. By addressing the interference source, the intensity of the interference signal radiated from the antenna feed line is reduced, thereby reducing electromagnetic interference to the sensitive electronic components of the UAV and ensuring the normal operation of the UAV flight control system.

[0015] Optionally, the device further includes a cable shielding structure for the UAV interface, specifically including: The USB differential cable in the communication cable connected to the UAV interface is configured with a twisted pair structure. The communication cable is fitted with an anti-interference magnetic ring, and the communication cable is wound around the anti-interference magnetic ring. The outermost layers of the communication cable and the anti-interference magnetic ring are successively wrapped with a copper foil shielding layer and a wave-absorbing material layer.

[0016] By adopting the above technical solutions, the twisted-pair structure of the USB differential cable enables external interference signals to generate induced voltages of equal magnitude and opposite direction on the two conductors of the twisted pair, thus canceling each other out and effectively suppressing common-mode interference. The communication cable is fitted with an anti-interference magnetic ring and wound around it, increasing the impedance of high-frequency interference signals passing through the cable and converting the high-frequency interference energy into heat for dissipation. The outer copper foil shielding layer effectively reflects and absorbs external electromagnetic radiation, forming a Faraday cage effect, preventing external interference signals from entering the cable and simultaneously preventing signals inside the cable from radiating outwards. The outermost absorbing material layer further absorbs residual high-frequency electromagnetic wave energy and converts it into heat. Through the synergistic effect of these multi-layered protection structures, the path of electromagnetic interference generated by the communication module to sensitive components such as the UAV RTK module via the PSDK cable is effectively blocked, solving the problem of interference from airborne devices to the UAV's satellite positioning.

[0017] Optionally, the core computing module is a Jetson Orin NX core board; the carrier board also integrates at least one of the following debugging interfaces: a micro HDMI interface, a USB interface, and a gigabit Ethernet port; the storage module is a solid-state drive (SSD).

[0018] By adopting the above technical solutions, the Jetson Orin NX core board possesses powerful GPU computing capabilities and deep learning inference performance, enabling local deployment and operation of high-performance models and meeting the edge computing needs of airborne devices, such as image processing and target detection. The carrier board integrates debugging interfaces such as a miniature HDMI interface, a USB interface, and a gigabit Ethernet port, facilitating development, debugging, and system programming, thereby improving the maintainability and development efficiency of the device. Using solid-state drives (SSDs) as storage modules offers advantages over traditional hard disk drives (HDDs) such as faster read / write speeds, stronger vibration resistance, smaller size, and lighter weight, making it suitable for the vibration environment during UAV flight and meeting the high-speed storage requirements for system operation and data caching.

[0019] Secondly, this application provides an airborne intelligent image transmission method, which adopts the following technical solution: Includes the following steps: S1 establishes a connection with the drone through the drone interaction interface to obtain the drone's flight control data and video stream data; S2, the core computing module identifies the hardware physical node of the first communication module mounted on the carrier board, and the hardware physical node of the second communication module mounted on the expansion board; S3, at the software system layer, the first communication module and the second communication module are respectively mapped to the corresponding first virtual network port node and the second virtual network port node; S4, the first communication module and the second communication module are activated through the dialing program to enter the network registration state, a dual-link communication network is established, and the flight control data and the video stream data are sent to the ground terminal.

[0020] By adopting the above technical solution, and by sequentially completing the steps of hardware physical node identification, virtual network port node mapping, and dial-up activation, the standardized initialization of the two communication modules and the reliable establishment of the dual-link communication network are realized. This method performs the initialization process of the hardware layer and the software system layer in steps, ensuring the orderliness and reliability of the communication link establishment. This enables the device to transmit data to the ground end simultaneously through two heterogeneous network links, improving the bandwidth, stability, and redundancy of image transmission communication.

[0021] Optionally, the methods for obtaining data in step S1 may specifically include: The drone's flight control data and gimbal control data are transmitted and received via the asynchronous serial communication interface in the drone's interactive interface. The video stream data of the drone is obtained through a virtual network port mapped to the USB bus in the drone's interactive interface.

[0022] By adopting the above technical solutions, flight control data and gimbal control data are transmitted using an asynchronous serial communication interface. Leveraging its low latency and high reliability, the real-time performance of control commands is ensured. A virtual network port mapped from a USB bus is used to acquire video stream data. Utilizing the high bandwidth of the USB bus, the large data volume transmission requirements of high-definition video streams can be met. The two data types are transmitted through different channels, avoiding bandwidth contention and transmission delays caused by data mixing, thus improving the efficiency and reliability of data transmission.

[0023] Optionally, step S4 specifically includes: The first communication module registers and accesses the public network base station network by recognizing the hard SIM card; The second communication module registers and accesses the private network base station network by recognizing a hard SIM card or a soft SIM card; The core computing module communicates with the computer at the ground station via network image transmission through the public network base station network and the private network base station network according to the dialing script.

[0024] By adopting the above technical solution, the first communication module accesses the public network base station network through a hard SIM card, and realizes high-speed data transmission by utilizing the wide coverage of public 5G base stations; the second communication module supports two registration methods, hard SIM card or soft SIM card, which increases deployment flexibility, and realizes secure and reliable dedicated communication by accessing the private network base station network; the core computing module automatically completes network dialing and registration according to the dialing script, realizing the automated establishment of dual-link communication network; the dual-link heterogeneous communication of public network and private network ensures communication reliability and security in different network environments.

[0025] Optionally, during the data transmission in step S4, the core computing module performs local model inference processing on the video stream data from the UAV, and then pushes the processed high-definition video stream to the ground terminal through the first communication module or the second communication module.

[0026] By adopting the above technical solution, the core computing module performs local model inference processing on the video stream data before data transmission, enabling intelligent analysis functions such as target detection and image enhancement at the airborne end. This reduces the amount of redundant data transmitted to the ground end and lowers the bandwidth consumption. At the same time, the locally processed video stream contains intelligent analysis results, which can be directly used by the ground end after reception, reducing the computing pressure on the ground end and improving the response speed and work efficiency of the entire system.

[0027] In summary, this application includes at least one of the following beneficial technical effects: 1. By stacking the heat dissipation components, core computing module, carrier board and expansion board, the device is miniaturized and lightweight, meeting the payload and size requirements of small and medium-sized UAVs. At the same time, by integrating the first communication module and the second communication module, which operate in different frequency bands, on the carrier board and the expansion board respectively, a heterogeneous public network and private network dual-link image transmission network is established, which effectively improves the bandwidth, stability and redundancy of data transmission. 2. By physically dividing and arranging the antenna feeders of different communication modules through an antenna isolation structure, and by adopting a multi-layer protection structure for the PSDK communication cable, including USB differential twisted pair, anti-interference magnetic ring, copper foil shielding layer and wave-absorbing material layer, the electromagnetic interference problem of the airborne communication module to the UAV RTK satellite search and positioning system is effectively solved from both the source of interference and the conduction path, ensuring the safety of UAV flight control. 3. By adopting a high-performance computing module as the edge computing core, a high-performance computing model can be deployed on the airborne end for local inference processing, realizing intelligent analysis and preprocessing of video streams, reducing the occupation of communication bandwidth, and improving the response speed and intelligence level of the entire system. Attached Figure Description

[0028] Figure 1 This is a hardware block diagram of an airborne intelligent computing and communication device according to an embodiment of this application; Figure 2 This is a schematic diagram showing the external dimensions of an airborne intelligent computing and communication device according to an embodiment of this application; Figure 3 This is a schematic diagram of the board stack-up structure of the airborne intelligent computing and communication device according to an embodiment of this application; Figure 4 This is a flowchart illustrating the airborne intelligent image transmission method according to an embodiment of this application. Detailed Implementation

[0029] The following is in conjunction with the appendix Figure 1 - Appendix Figure 4 This application will be described in further detail below. Example

[0030] An airborne intelligent computing and communication device and method, referring to Figure 1 The device comprises stacked heat dissipation components, a core computing module, a carrier board, and expansion boards. From top to bottom, the entire device consists of stacked heat dissipation components, a core computing module, a carrier board, and expansion boards, with electrical connections and signal transmission between the layers achieved through board-to-board connectors or ribbon cables.

[0031] Reference Figure 2 In this embodiment, the overall dimensions of the device are 63.1mm (length) × 46.4mm (width) × 94.6mm (height), forming a compact cuboid structure. An external metal casing protects the internal circuit boards and modules. The sides and top of the casing are respectively provided with antenna interface holes for antenna feed lines and interface openings for peripheral interfaces.

[0032] Reference Figure 3 The heat dissipation component is located on one side above the core computing module. The heat dissipation component includes a cooling fan, which actively cools the core computing module. The cooling fan is fixed to the upper surface of the core computing module by screws or clips. When the core computing module is under high load, the cooling fan runs continuously, quickly dissipating the heat generated by the core computing module to the outside of the device, ensuring stable operation of the core computing module within a safe operating temperature range.

[0033] The core computing module serves as the edge computing core of the entire device. In this embodiment, the core computing module utilizes an NVIDIA Jetson Orin NX 8G core board. This core board integrates a high-performance GPU and a multi-core CPU, possessing powerful deep learning inference and parallel computing capabilities. It supports the local deployment of high-performance models and is suitable for edge computing tasks such as image recognition, object detection, and autonomous path planning on the airborne end. The core computing module communicates with the carrier board and expansion boards, receiving and processing data from various functional modules on the carrier board and expansion boards, including UAV flight control data streams and video streams.

[0034] The carrier board is located below the core computing module and uses a 10-layer PCB design. During the PCB design process, impedance matching and length equalization were implemented for high-speed signal lines such as PCIe, USB 2.0, USB 3.0, SDIO, and HDMI. Through-hole and blind / buried via processes were used, and multi-layer board stack-up design, copper pouring and copper thickness design, and PCB substrate selection were carried out to ensure the integrity and reliability of high-speed signal transmission.

[0035] The carrier board integrates the following functional modules and interfaces: The first communication module, in this embodiment, is a 5G communication module operating in the public network frequency band. The 5G communication module communicates with the core computing module via a USB 2.0 interface and is mapped as a virtual network interface node in the core computing module's operating system. The 5G communication module supports the insertion of a hard SIM card, registering and accessing the public 5G base station network through the hard SIM card to achieve high-speed, low-latency data transmission. The 5G communication module is connected to a first antenna feed line, which extends from the carrier board, passes through the antenna interface hole on the casing, and connects to the 5G antenna outside the casing.

[0036] In this embodiment, the drone interaction interface is a PSDK (Payload SDK) interface. The carrier board physically connects to the drone via the PSDK interface, obtaining a +24V / 4A power supply from the drone to power the entire device. Simultaneously, the PSDK interface integrates an asynchronous serial communication UART channel and a USB 2.0 channel: through the UART channel, the carrier board communicates bidirectionally with the drone's flight control system, enabling the transmission and reception of flight control commands and the control of the camera and gimbal; through the virtual network port mapped by the USB 2.0 channel, it acquires video stream data captured by the drone. This multi-channel multiplexing design of the PSDK interface achieves an integrated connection for power supply, flight control communication, and video stream acquisition, simplifying the cable connections between the device and the drone.

[0037] In this embodiment, the storage module is a 256GB solid-state drive (SSD). The SSD is connected to the core computing module at high speed via a PCIe interface and is used to store the operating system, applications, AI model data, and temporary runtime data. The SSD features fast read / write speeds, strong shock resistance, small size, and light weight, making it suitable for the vibration environment encountered during drone flight.

[0038] In addition, the carrier board also integrates the following debugging interfaces: The micro HDMI interface connects to the core computing module via the HDMI 2.0 protocol. It is used to output the graphical interface and video output of the core computing module during the debugging phase, which facilitates system debugging and status monitoring by developers.

[0039] The USB interface includes a USB 3.0 debug port (Type-C interface) and a USB 2.0 + Gigabit Ethernet + UART serial port combo programmable debug port (Type-C interface). The USB 3.0 debug port connects to the core computing module via the USB 3.0 / 2.0 protocol for high-speed data transfer and system debugging. The combo programmable debug port integrates the functions of USB 2.0, Gigabit Ethernet, and UART serial ports, allowing system programming, network debugging, and serial debugging to be performed through a single Type-C interface. This reduces the number of interfaces and contributes to the miniaturization of the device.

[0040] A gigabit Ethernet port is integrated into the three-in-one programming and debugging port. It enables gigabit Ethernet communication through a USB 2.0 interface converter and is used for wired network connection during the development and debugging phase.

[0041] The expansion board is located below the carrier board, and the expansion board and carrier board are electrically connected via board-to-board connectors or ribbon cables. The expansion board integrates a second communication module, which in this embodiment is a 1.4G communication module operating in the private network frequency band. The 1.4G communication module is electrically connected to the carrier board via a USB 2.0 interface, and then communicates with the core computing module, being mapped as another virtual network port node in the core computing module's operating system. The 1.4G communication module supports registration and access to the 1.4G private network base station network via either a hard SIM card or a soft SIM card, enabling secure communication in the private network environment. The power supply circuit for the 1.4G communication module on the expansion board is specially designed, and the USB 2.0 communication line is impedance matched and length-equalized to ensure the quality of the communication signal. The 1.4G communication module is connected to a second antenna feeder, which extends from the expansion board, passes through the antenna interface hole on the casing, and connects to the 1.4G antenna outside the casing.

[0042] A heterogeneous dual-link image transmission network is established by the first communication module (5G communication module) on the carrier board and the second communication module (1.4G communication module) on the expansion board communicating with the core computing module through their respective virtual network ports. The 5G link operates in the public network frequency band, providing high-speed, high-bandwidth data transmission capabilities; the 1.4G link operates in the private network frequency band, providing high-security and strong confidentiality communication capabilities. The two links can operate simultaneously or switch according to the actual network environment, ensuring communication reliability in various complex environments.

[0043] The device also includes an electromagnetic interference (EMI) shielding structure to address the electromagnetic interference generated by the airborne communication module on the UAV's RTK satellite positioning system. The EMI shielding structure comprises an antenna isolation structure and a cable shielding structure.

[0044] Regarding the antenna isolation structure: Inside the device's housing, the first antenna feed line connected to the first communication module and the second antenna feed line connected to the second communication module are physically separated and arranged. Specifically, the first and second antenna feed lines are placed in different areas inside the housing, maintaining sufficient physical spacing between them to avoid overlap. Simultaneously, the antenna feed lines are shortened or replaced with shorter ones to reduce the intensity of electromagnetic interference signals radiated from the antenna feed lines. Through these antenna isolation measures, the mutual coupling interference between the 5G and 1.4G communication modules is reduced at the source of interference, and the interference signals radiated from the antenna feed lines are lowered.

[0045] Regarding the cable shielding structure: The communication cable connecting to the UAV's interactive interface (PSDK interface) undergoes multi-layer shielding. First, the USB differential lines in the communication cable employ a twisted-pair structure, where the positive and negative conductors of the USB differential signal line are twisted together at a specific pitch. This allows the induced voltages generated by external interference signals on the two conductors to cancel each other out, effectively suppressing common-mode interference. Second, an anti-interference magnetic ring is fitted around the outside of the communication cable, with the cable wound multiple times around it. The anti-interference magnetic ring is a high-permeability ferrite ring. After the communication cable winds through the ring, high-frequency interference signals generate significant impedance, and the interference energy is converted into heat for dissipation, effectively filtering out high-frequency interference. Third, the outermost layers of the communication cable and the anti-interference magnetic ring are successively wrapped with a copper foil shielding layer and an absorbing material layer. The copper foil shielding layer uses highly conductive copper foil tightly wrapped to form a continuous electromagnetic shielding layer, effectively reflecting and absorbing external electromagnetic radiation, preventing interference signals from entering the cable and being conducted to the UAV, while also preventing signals inside the cable from radiating outwards. The absorbing material layer is wrapped around the outside of the copper foil shielding layer. It uses absorbing materials such as absorbing paper or absorbing tape to absorb the residual high-frequency electromagnetic wave energy that penetrates the copper foil shielding layer and convert it into heat for dissipation, thereby further enhancing the shielding effect.

[0046] By combining the antenna isolation structure and the cable shielding structure, the electromagnetic interference generated by the 5G and 1.4G communication modules is effectively blocked from interfering with the UAV's RTK module through two paths: antenna radiation and PSDK cable conduction. This solves the problem of UAV satellite acquisition difficulties and positioning drift caused by airborne devices, and ensures the safety of UAV flight control.

[0047] This device is compatible with multiple DJI models such as the M30, M300, and M350. To accommodate the load-bearing limitations and structural interference issues of different models, various degrees of weight reduction and structural optimization have been carried out in terms of PCB board design optimization, antenna selection, and material selection for the casing and mounting bracket, achieving a miniaturized and lightweight design of the entire device.

[0048] The implementation principle of this application embodiment is as follows: The device establishes a physical connection with the UAV through the PSDK interface, obtains power from the UAV, receives flight control data through the UART channel, and obtains video streams through the USB channel; the core computing module (Jetson Orin NX 8G core board) performs local AI model inference processing on the received video stream data to realize edge computing functions such as image recognition and target detection; the processed data is pushed to the industrial control computer or computer at the ground station via the 5G communication module on the carrier board through the public 5G base station network and / or the 1.4G communication module on the expansion board through the private 1.4G base station network in a dual-link heterogeneous image transmission mode, which is a high-definition video stream and flight control data; at the same time, the device effectively suppresses electromagnetic interference through the antenna isolation structure and cable shielding structure, ensuring the normal operation of the UAV navigation system. Example

[0049] An airborne intelligent computing and communication device and method, referring to Figure 4 This includes the following steps: Step S1: Establish a connection with the drone through the drone interaction interface to obtain the drone's flight control data and video stream data.

[0050] Specifically, after the device is powered on, the PSDK interface on the carrier board establishes a physical connection and protocol handshake with the drone. Through the asynchronous serial communication UART interface in the PSDK interface, a bidirectional communication channel is established with the drone's flight control system to send and receive flight control data (including aircraft attitude data, GPS positioning data, flight speed and altitude data, etc.) and gimbal control data (including gimbal rotation angle, camera zoom ratio, and other parameters). Simultaneously, through the USB 2.0 bus in the PSDK interface, a corresponding virtual network port is mapped in the operating system of the core computing module, and video stream data captured by the drone's camera is obtained through this virtual network port.

[0051] In step S2, the core computing module identifies the hardware physical node of the first communication module mounted on the carrier board and the hardware physical node of the second communication module mounted on the expansion board.

[0052] Specifically, after the core computing module (Jetson Orin NX core board) boots the operating system, it scans and enumerates the hardware physical node information of the 5G first communication module connected to the carrier board via the USB bus, including the device's VID (Vendor ID) and PID (Product ID), confirming that the 5G first communication module has been correctly connected and recognized by the system. Similarly, the core computing module scans and enumerates the hardware physical node information of the 1.4G second communication module connected to the expansion board via the USB bus, confirming that the 1.4G second communication module has been correctly connected and recognized by the system. The system records the hardware physical node paths of the two communication modules in the kernel log, providing the hardware foundation for subsequent software layer mapping.

[0053] Step S3: At the software system layer, the first communication module and the second communication module are mapped to the corresponding first virtual network port node and second virtual network port node, respectively.

[0054] Specifically, within the Linux operating system of the core computing module, after the 5G first communication module is recognized by the driver via the USB 2.0 interface, the system automatically creates a corresponding first virtual network interface node at the network device layer (e.g., mapped to network interface nodes such as usb0 or wwan0). Similarly, after the 1.4G second communication module is recognized by the driver via the USB 2.0 interface, the system automatically creates a corresponding second virtual network interface node (e.g., mapped to network interface nodes such as usb1 or wwan1). These two virtual network interface nodes exist as independent network interfaces in the operating system's network protocol stack, and their IP addresses, routing rules, and network policies can be configured separately.

[0055] Step S4: Activate the first communication module and the second communication module respectively through the dialing program to enter the network registration state, establish a dual-link communication network, and send the flight control data and the video stream data to the ground terminal.

[0056] Specifically, this step includes the following sub-steps: S41, the first communication module (5G communication module) sends a network registration request to the public 5G base station by recognizing the inserted hard SIM card. The core computing module, according to a preset dialing script, sends dialing commands to the 5G first communication module via AT commands or the QMI (Qualcomm MSM Interface) protocol. The 5G first communication module completes authentication and registration with the public 5G base station, establishing a data connection. After successful dialing, the first virtual network interface node obtains a public IP address, establishing the first communication link from the public 5G base station network to the ground station computer.

[0057] S42, the second communication module (1.4G communication module) sends a network registration request to the 1.4G private network base station by recognizing the inserted hard SIM card or the pre-installed soft SIM card. The core computing module sends a dialing command to the 1.4G second communication module according to a preset dialing script. The 1.4G second communication module completes authentication and registration with the 1.4G private network base station and establishes a data connection. After successful dialing, the second virtual network interface node obtains a private network IP address, establishing a second communication link via the 1.4G private network base station to the ground station computer.

[0058] 1.4G private network base stations are currently deployed in some cities. Their coverage is limited, but their signal security and confidentiality are high, and they are not easily intercepted or interfered with by the outside world. 1.4G private network communication is more suitable for use by systems such as public security or military units that have strict requirements for information security.

[0059] S43. After the dual-link communication network is established, the core computing module simultaneously or selectively transmits flight control data and video stream data to the ground terminal through the first communication link (5G public network link) and the second communication link (1.4G private network link). The core computing module can dynamically select the data transmission link based on the signal quality, bandwidth, and latency of the two links, or distribute different types of data to different links for transmission, thereby achieving load balancing and redundancy backup for data transmission.

[0060] During the data transmission in step S4, the core computing module can also perform local model inference processing on the video stream data from the UAV. Specifically, the core computing module (Jetson Orin NX 8G core board) utilizes its built-in GPU acceleration capabilities to run pre-deployed deep learning models (such as object detection models, image segmentation models, etc.) to perform real-time inference processing on the received video stream data, such as identifying specific targets in the video frame, performing image enhancement, or detecting abnormal events. The processed high-definition video stream and inference result data are pushed to the ground-based industrial control computer or computer through the first or second communication module for video display and control decision-making.

[0061] Through local model inference processing, the core computing module can complete a large amount of data preprocessing and intelligent analysis on the airborne end, transmitting only key data and processing results to the ground end, effectively reducing the occupation of communication bandwidth and improving the response speed and intelligence level of the entire system.

[0062] The implementation principle of this application embodiment is as follows: Through the sequential execution of steps S1 to S4, data communication with the UAV is first established through the PSDK interface to acquire flight control data and video streams; then, the identification of the two communication modules and virtual network port mapping are completed at the hardware and software layers respectively; next, the two communication links are automatically activated through the dialing script to establish a heterogeneous dual-link communication network of public and private networks; finally, after the core computing module performs local AI inference processing on the video stream, it pushes the high-definition video stream and flight control data to the ground terminal through the dual-link network. The entire method realizes a complete closed loop from data acquisition and local intelligent processing to dual-link high-reliability transmission, providing a high-bandwidth, high-stability, and high-security airborne intelligent image transmission solution for UAVs in complex application scenarios such as intelligent cruise and security reconnaissance.

[0063] The embodiments described in this specific implementation are preferred embodiments of this application and are not intended to limit the scope of protection of this application. Identical components are represented by the same reference numerals. Therefore, all equivalent changes made to the structure, shape, and principle of this application should be covered within the scope of protection of this application.

Claims

1. An airborne intelligent computing and communication device, characterized in that, It includes a stacked heat dissipation assembly, a core computing module, a carrier board, and expansion boards; The heat dissipation component is located on one side of the core computing module and is used to actively dissipate heat from the core computing module. The core computing module, as the edge computing core, is communicatively connected to the carrier board and the expansion board, and is used to process UAV flight control data streams and video streams. The carrier board integrates a first communication module, a drone interaction interface, and a storage module; the drone interaction interface is used to physically connect with the drone to obtain power and data from the drone. The expansion board integrates a second communication module, and the expansion board is electrically connected to the carrier board. The first communication module and the second communication module communicate with the core computing module through virtual network ports to establish a heterogeneous dual-link image transmission network.

2. The airborne intelligent computing and communication device according to claim 1, characterized in that, The UAV's interactive interface is a PSDK interface; the carrier board obtains power from the UAV through the PSDK interface, communicates with the UAV's flight control system through the asynchronous serial communication UART channel in the PSDK interface, and obtains the UAV's video stream through the virtual network port mapped by the USB channel.

3. The airborne intelligent computing and communication device according to claim 1, characterized in that, The first communication module is a 5G communication module operating in the public network frequency band; the second communication module is a 1.4G communication module operating in the private network frequency band.

4. The airborne intelligent computing and communication device according to claim 3, characterized in that, The device further includes an electromagnetic interference protection structure, which includes: The antenna isolation structure includes a first communication module connected to a first antenna feed line and a second communication module connected to a second antenna feed line. The first antenna feed line and the second antenna feed line are physically separated and arranged within the housing of the device to avoid crossing or overlapping.

5. An airborne intelligent computing and communication device according to claim 2, characterized in that, The device also includes a cable shielding structure for the UAV interaction interface, specifically including: The USB differential cable in the communication cable connected to the UAV interface is configured with a twisted pair structure. The communication cable is fitted with an anti-interference magnetic ring, and the communication cable is wound around the anti-interference magnetic ring. The outermost layers of the communication cable and the anti-interference magnetic ring are successively wrapped with a copper foil shielding layer and a wave-absorbing material layer.

6. The airborne intelligent computing and communication device according to claim 1, characterized in that, The core computing module is a Jetson Orin NX core board; the carrier board also integrates at least one of the following debugging interfaces: a micro HDMI interface, a USB interface, and a gigabit Ethernet port; the storage module is a solid-state drive (SSD).

7. An airborne intelligent image transmission method based on the device according to any one of claims 1-6, characterized in that, Includes the following steps: S1 establishes a connection with the drone through the drone interaction interface to obtain the drone's flight control data and video stream data; S2, the core computing module identifies the hardware physical node of the first communication module mounted on the carrier board, and the hardware physical node of the second communication module mounted on the expansion board; S3, at the software system layer, the first communication module and the second communication module are respectively mapped to the corresponding first virtual network port node and the second virtual network port node; S4, the first communication module and the second communication module are activated through the dialing program to enter the network registration state, a dual-link communication network is established, and the flight control data and the video stream data are sent to the ground terminal.

8. The airborne intelligent image transmission method according to claim 7, characterized in that, The specific methods for obtaining data in step S1 include: The drone's flight control data and gimbal control data are transmitted and received via the asynchronous serial communication interface in the drone's interactive interface. The video stream data of the drone is obtained through a virtual network port mapped to the USB bus in the drone's interactive interface.

9. The airborne intelligent image transmission method according to claim 7, characterized in that, Step S4 specifically includes: The first communication module registers and accesses the public network base station network by recognizing the hard SIM card; The second communication module registers and accesses the private network base station network by recognizing a hard SIM card or a soft SIM card; The core computing module communicates with the computer at the ground station via network image transmission through the public network base station network and the private network base station network according to the dialing script.

10. The airborne intelligent image transmission method according to claim 7, characterized in that, During the data transmission in step S4, the core computing module performs local model inference processing on the video stream data from the UAV, and then pushes the processed high-definition video stream to the ground terminal through the first communication module or the second communication module.

Images