A multi-mode satellite communication-Ascend computing power integrated module and data processing method

Abstract

This invention discloses a multi-mode satellite communication and Ascend computing power integrated module and data processing method. The module includes: an Ascend NPU chip, which runs an AI-based scheduling model to predict and evaluate link latency, bandwidth, and power consumption for drones in real time, while dynamically selecting the optimal communication mode based on path planning and obstacle avoidance algorithms; a Tiantong satellite communication module, which transmits the drone's status log; a Beidou positioning and short message module, which transmits the drone's location, status, and emergency messages; a 4G / 5G module, which transmits real-time control commands and data; an IMU module, which combines Beidou GNSS for integrated navigation and positioning; and an antenna for transmitting and receiving wireless signals. This invention integrates multiple communication modes and combines them with protocol scheduling optimization based on Ascend computing power to provide low-altitude unmanned aerial vehicles with wide-area coverage and precise control communication capabilities.

Description

Technical Field

[0001] The present invention relates to the field of multimode communication technology, and in particular to a multimode satellite communication-Ascend computing power integrated module and a data processing method. Background Art

[0002] In low-altitude economic scenarios, there are problems such as obstruction by urban buildings and unstable signals in remote areas, which make the single communication mode extremely susceptible to shielding and interference, and there is also the problem of insufficient coverage, which seriously restricts the flight safety and efficiency of drones in application scenarios such as urban logistics, air traffic management, and emergency response.

[0003] Multimode communication, by integrating satellite, 4G / 5G, and other technologies, enables seamless connectivity across the entire region, redundant backup, and real-time data transmission. This ensures the precise positioning and collaborative operation capabilities of low-altitude equipment, and is an essential foundation for the economic, large-scale, and intelligent development of low-altitude communications. However, existing multimode communication modules typically integrate Tiantong, Beidou, and 4G / 5G in pairs, which also cannot meet the needs of low-altitude communications. They suffer from transmission limitations for heterogeneous data (data from different sources with varying structures, formats, or semantics. In low-altitude scenarios, different sensor types generate data in different formats, including video, images, and environmental data) and the inability to rationally allocate multimode communications.

[0004] Therefore, the existing technology still needs to be improved and developed. Summary of the Invention

[0005] The main purpose of this invention is to provide a multi-mode satellite communication-Ascend computing power integrated module and data processing method, aiming to solve the problems in the existing technology that multi-mode communication cannot meet the needs of low-altitude communication, has shortcomings in heterogeneous data transmission, and multi-mode communication cannot be reasonably allocated.

[0006] To achieve the above objectives, the present invention provides a multi-mode satellite communication-Ascend computing power integrated module, which includes:

[0007] The Ascend NPU chip is used to run AI-based scheduling models, making real-time predictions and evaluating link latency, bandwidth, and power consumption for drones. It also dynamically selects the optimal communication mode among Tiantong satellite communications, Beidou-3 short message communications, and 4G / 5G communications based on path planning and obstacle avoidance algorithms.

[0008] Tiantong satellite communication module, used to transmit the drone's status log;

[0009] Beidou positioning and short message module, used to transmit drone location, status and emergency messages;

[0010] 4G / 5G module for transmitting real-time control instructions, high-definition images and high-definition video streams;

[0011] The IMU module is used for integrated navigation and positioning in conjunction with Beidou GNSS, providing autonomous navigation capabilities when GNSS signals are interrupted, and correcting GNSS multipath errors;

[0012] Antenna, used to transmit and receive wireless signals.

[0013] Optionally, the multi-mode satellite communication-Ascend computing power integrated module, wherein the antenna includes: Tiantong external antenna, Beidou external antenna, 4G / 5G external antenna and WIFI / Bluetooth antenna.

[0014] Optionally, the multi-mode satellite communication-Ascend computing power integrated module further includes: a communication interface, which is used to communicate with an external device.

[0015] Optionally, the multi-mode satellite communication-Ascend computing power integrated module, wherein the communication interface includes: RS-232, RS-485, WIFI and Bluetooth.

[0016] In addition, to achieve the above objectives, the present invention also provides a data processing method based on the multi-mode satellite communication-Ascend computing power integrated module, the data processing method comprising:

[0017] Acquire external data and input the external data into the multi-mode satellite communication-Ascend computing power integrated module, wherein the external data includes the real-time status of the UAV, environmental information, and flight control command information;

[0018] The Ascend NPU chip runs a scheduling model based on AI algorithms to make real-time predictions and evaluate link latency, bandwidth, and power consumption for drones. It also dynamically selects the optimal communication mode among Tiantong satellite communications, Beidou third-generation short message communications, and 4G / 5G communications based on path planning and obstacle avoidance algorithms.

[0019] Optionally, the data processing method, wherein the Ascend NPU chip runs a scheduling model based on an AI algorithm, performs real-time prediction and evaluation of link latency, bandwidth, and power consumption for the UAV, and dynamically selects the optimal communication mode among Tiantong satellite communication, Beidou third-generation short message communication, and 4G / 5G communication based on the path planning algorithm and obstacle avoidance algorithm, specifically includes:

[0020] The Ascend NPU chip aligns path coordinates, obstacle distribution, and real-time communication channel status into a unified spatiotemporal coordinate system.

[0021] The Ascend NPU chip uses CNN to analyze the shape and density of obstacles on the path within a preset time in the future and predict signal attenuation areas;

[0022] The Ascend NPU chip uses a reinforcement learning algorithm to integrate path risks, latency prediction, and energy consumption constraints to generate the optimal communication mode in Tiantong satellite communication, Beidou third-generation short message communication, and 4G / 5G communication, and achieves end-to-end decision-making through Ascend parallel computing;

[0023] The Ascend NPU chip dynamically allocates resources based on the NPU computing power.

[0024] Optionally, in the data processing method, the path planning algorithm is used to generate a flight trajectory, and the obstacle avoidance algorithm is used to output a real-time obstacle heat map.

[0025] Optionally, the data processing method, wherein generating the optimal communication mode in Tiantong satellite communication, Beidou third-generation short message communication and 4G / 5G communication, specifically includes:

[0026] Prioritize 4G / 5G communication, BeiDou-3 short message communication, and Tiantong satellite communication in descending order in advance;

[0027] Monitor 4G / 5G signal strength and latency. When the latency is greater than the preset latency threshold or the RSRP is less than the preset signal receiving power threshold, activate BeiDou III short message communication.

[0028] When the Beidou short message channel information is congested or in a preset blocking environment, and 4G / 5G communication has not returned to normal, switch to Tiantong satellite communication.

[0029] Optionally, the data processing method, wherein the Ascend NPU chip runs a scheduling model based on an AI algorithm, performs real-time prediction and evaluation of link latency, bandwidth, and power consumption for the UAV, and dynamically selects the optimal communication mode among Tiantong satellite communication, Beidou third-generation short message communication, and 4G / 5G communication based on a path planning algorithm and an obstacle avoidance algorithm, further includes:

[0030] Redundant communication links are defined according to the flight mission, and the redundant communication links are kept in hot standby state. Channels are applied for regularly and transmission contexts are preloaded.

[0031] Optionally, the data processing method further includes defining redundant communication links according to flight missions, maintaining the redundant communication links in a hot standby state, periodically applying for channels, and preloading transmission contexts, and then further includes:

[0032] Through the optimal communication mode, commands and messages are sent to mobile phone / PAD mobile terminal devices, drones and data management centers.

[0033] In the present invention, the multi-mode satellite communication-Ascend computing power integrated module includes: Ascend NPU chip, which is used to run a scheduling model based on AI algorithm, perform real-time prediction of drones and evaluate link delay, bandwidth and power consumption, and dynamically select the optimal communication mode among Tiantong satellite communication, Beidou third-generation short message communication and 4G / 5G communication based on path planning algorithm and obstacle avoidance algorithm; Tiantong satellite communication module, which is used to transmit the status log of the drone; Beidou positioning and short message module, which is used to transmit the location, status and emergency messages of the drone; 4G / 5G module, which is used to transmit real-time control instructions, high-definition pictures and high-definition video streams; IMU module, which is used to combine Beidou GNSS for combined navigation and positioning, provide autonomous navigation capability when the GNSS signal is interrupted, and correct GNSS multipath error; antenna, which is used to transmit and receive wireless signals. This invention integrates Tiantong satellite communication, Beidou third-generation short message communication and 4G / 5G communication, and combines it with protocol scheduling optimization based on Ascend computing power to achieve communication function integration, lightweight protocol, and a highly integrated general computing module, providing low-altitude unmanned aerial vehicles with wide-area coverage and precise control communication functions. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a structural diagram of the multi-mode satellite communication-Ascend computing power integrated module of the present invention;

[0035] Figure 2 This is a flowchart of a preferred embodiment of the data processing method based on the multi-mode satellite communication-Ascend computing power integrated module of the present invention;

[0036] Figure 3 This is a schematic diagram of the data processing implementation process of the present invention based on the multi-mode satellite communication-Ascend computing power integrated module. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solutions and advantages of the present invention more clear and distinct, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0038] The Tiantong satellite system is a satellite mobile communication system that mainly provides communication services such as voice, text messages, and data. It has dedicated satellite phones or terminals that support voice calls, short messages, and data transmission. Its application scenarios include emergency rescue, marine fisheries, geological exploration, and other scenarios without ground network coverage. It is autonomous and controllable, highly secure, and can make up for ground communication blind spots, making it suitable for special industry needs.

[0039] The Beidou system is a global satellite navigation system (similar to GPS), but its unique feature is its short message communication capability. It can send short text or location information via satellite. The communication method is two-way short messages, without relying on ground networks. Navigation and communication are integrated, with low power consumption and high reliability, making it suitable for extreme environments.

[0040] 4G (LTE) and 5G are mainstream terrestrial mobile communication technologies. Based on cellular networks, they provide high-speed data transmission and wide-area coverage, high bandwidth, low latency, and support the connection of massive devices.

[0041] Existing multi-mode communication modules typically integrate Tiantong, Beidou, and 4G / 5G in pairs, each with its own advantages and disadvantages. Currently, no single module can simultaneously address the shortcomings of existing communications and integrate the advantages of all three. The disadvantages of pairwise integration are shown in Table 1:

[0042] Table 1: Comparison of the functions and disadvantages of two-way communication module combinations

[0043]

[0044] The multi-mode satellite communication-Ascend computing power integrated module described in the preferred embodiment of the present invention is as follows: Figure 1 As shown, the multi-mode satellite communication-Ascend computing power integrated module includes:

[0045] The Ascend NPU chip is used to run AI-based scheduling models, making real-time predictions and evaluating link latency, bandwidth, and power consumption for drones. It also dynamically selects the optimal communication mode among Tiantong satellite communications, Beidou-3 short message communications, and 4G / 5G communications based on path planning and obstacle avoidance algorithms.

[0046] Tiantong satellite communication module, used to transmit the drone's status log;

[0047] Beidou positioning and short message module, used to transmit drone location, status and emergency messages;

[0048] 4G / 5G module for transmitting real-time control instructions, high-definition images and high-definition video streams;

[0049] The IMU module (Inertial Measurement Unit, a sensor device used to measure an object's acceleration, angular velocity, and attitude) is used for integrated navigation and positioning with Beidou GNSS, providing autonomous navigation capabilities when GNSS signals are interrupted and correcting GNSS multipath errors.

[0050] Antenna, used to transmit and receive wireless signals.

[0051] Among them, the Ascend NPU chip refers to Huawei's Ascend series AI processors (chips designed to provide powerful computing power support for artificial intelligence applications). The Ascend NPU chip runs edge computing, AI algorithms, etc., based on multi-modal heterogeneous data fusion and multi-model collaborative prediction such as CNN-LSTM model and Ascend NPU hardware acceleration, real-time prediction of link latency, bandwidth, and power consumption, achieving second-level evaluation of latency, bandwidth, and power consumption, and dynamically optimizing communication link selection.

[0052] The Ascend NPU chip dynamically selects the optimal communication mode based on path planning and obstacle avoidance algorithms, specifically including:

[0053] (1) Data fusion: aligning path coordinates, obstacle distribution, and real-time communication channel status (5G / Beidou / Tiantong signal strength) into a unified spatiotemporal coordinate system;

[0054] (2) Signal obstruction / attenuation prediction: Use CNN to analyze the shape and density of obstacles on the path within a preset time in the future (e.g., the next 5-10 seconds) to predict signal attenuation areas (e.g., 5G attenuation caused by building obstruction, Beidou attenuation caused by entering a pipe gallery);

[0055] (3) Communication decision-making: Using reinforcement learning algorithms (Q-Learning, a classic model-free reinforcement learning algorithm, belongs to the temporal difference (TD) learning method. It optimizes the decision-making strategy of intelligent agents (such as drones and robots) by continuously updating the action value function (Q function), and ultimately learns to take the best action in a specific environment to maximize the cumulative reward) to integrate path risk, delay prediction and energy consumption constraints, generate the optimal communication mode in Tiantong satellite communication, Beidou third-generation short message communication and 4G / 5G communication (for example, switch to Tiantong + Beidou redundancy before entering the 5G obstruction area), and achieve end-to-end decision-making through Ascend parallel computing;

[0056] (4) Protocol tuning: Dynamically allocate resources based on the NPU computing power to ensure the transmission of high-priority data such as obstacle avoidance instructions.

[0057] Among them, the Tiantong satellite communication module is used to transmit the drone status log (for example, to the user end, that is, the end that needs data and image frequency) to ensure basic communication needs.

[0058] Among them, the Beidou positioning and short message module is used to transmit the drone's location, status and emergency messages (for example, to the user end, that is, the end that needs data and image frequency).

[0059] Among them, the 4G / 5G module is used to transmit real-time control instructions, high-definition pictures and high-definition video streams (for example, to the user end, that is, the end that needs data and images).

[0060] Among them, the IMU module is used to combine with Beidou GNSS for combined navigation and positioning. The IMU module provides autonomous navigation capability when the GNSS signal is interrupted, and at the same time corrects the GNSS multipath error to achieve continuous and high-precision combined positioning, thereby improving the robustness of the drone's navigation system in complex environments.

[0061] Among them, the antennas include: Tiantong external antenna, Beidou external antenna, 4G / 5G external antenna, WIFI / Bluetooth antenna, which are used to transmit and receive wireless signals.

[0062] Among them, the communication protocol has one-way redundancy, that is, in Tiantong satellite communication, Beidou third-generation short message communication and 4G / 5G communication, two communication protocols are always guaranteed to be activated to prevent switching errors.

[0063] The multi-mode satellite communication-Ascend computing power integrated module also includes: a communication interface, which is used to communicate with external devices. The communication interfaces include: RS-232, RS-485, WIFI and Bluetooth. RS-232 and RS-485 are used to connect to sensors and other devices. Mobile phones / PAD mobile terminal devices (Portable Android Device or Portable Data Terminal, a portable smart terminal that integrates data acquisition, processing and communication functions) can send and receive data via Bluetooth or WIFI.

[0064] Furthermore, based on the multi-mode satellite communication-Ascend computing power integrated module, the present invention also provides a data processing method. The data processing method described in the preferred embodiment of the present invention is as follows: Figure 2 and Figure 3 As shown, the data processing method includes:

[0065] Step S10: Acquire external data and input the external data into the multi-mode satellite communication-Ascend computing power integrated module, wherein the external data includes the real-time status of the UAV, environmental information, and flight control command information;

[0066] In step S20, the Ascend NPU chip runs a scheduling model based on an AI algorithm to make real-time predictions and evaluate link latency, bandwidth, and power consumption for the drone. At the same time, based on the path planning algorithm and obstacle avoidance algorithm, it dynamically selects the optimal communication mode among Tiantong satellite communication, Beidou third-generation short message communication, and 4G / 5G communication.

[0067] Specifically, the step S20 includes:

[0068] Step S21: The Ascend NPU chip aligns the path coordinates, obstacle distribution, and real-time communication channel status (5G / Beidou / Tiantong signal strength) into a unified space-time coordinate system.

[0069] In step S22, the Ascend NPU chip uses CNN to analyze the shape and density of obstacles on the path within a preset time in the future (e.g., the next 5-10 seconds) and predict signal attenuation areas (e.g., 5G attenuation caused by building obstructions, Beidou attenuation caused by entering a pipe gallery).

[0070] Step S23: The Ascend NPU chip uses a reinforcement learning algorithm (Q-Learning) to integrate path risks, latency prediction, and energy consumption constraints to generate the optimal communication mode in Tiantong satellite communication, Beidou third-generation short message communication, and 4G / 5G communication (such as switching to Tiantong + Beidou redundancy before entering a 5G obstruction area), and achieve end-to-end decision-making through Ascend parallel computing;

[0071] Step S24: The Ascend NPU chip dynamically allocates resources based on the NPU computing power to ensure the transmission of high-priority data such as obstacle avoidance instructions.

[0072] Among them, the path planning algorithm (UAV path planning algorithm is the core technology to ensure that UAVs complete flight missions efficiently and safely, involving key links such as environmental perception, dynamic obstacle avoidance, and optimal path generation, such as genetic algorithms and ant colony algorithms) is used to generate flight trajectories, and the obstacle avoidance algorithm (UAV obstacle avoidance algorithm is the core technology to ensure the safe flight of UAVs in complex environments, mainly through sensors to perceive obstacles and adjust the flight path in real time, such as end-to-end obstacle avoidance and semantic segmentation assistance) is used to output a real-time obstacle heat map (real-time obstacle heat map is a tool that uses dynamic visualization technology to intuitively display the distribution and threat level of obstacles in the environment. It is widely used in autonomous driving, UAV navigation, robot obstacle avoidance and security monitoring and other fields).

[0073] Generating the optimal communication mode in Tiantong satellite communication, Beidou third-generation short message communication and 4G / 5G communication specifically includes:

[0074] Prioritize 4G / 5G communication, BeiDou-3 short message communication, and Tiantong satellite communication in descending order in advance;

[0075] Monitor 4G / 5G signal strength and latency. When the latency is greater than the preset latency threshold (e.g., latency > 20ms) or RSRP (reference signal received power) is less than the preset signal received power threshold (e.g., RSRP < -110dBm), that is, when the signal is weak or the latency increases suddenly, activate BeiDou-3 short message communication.

[0076] When the Beidou short message channel is congested or in a pre-set obstruction environment (such as underground or in a severely obstructed environment), and 4G / 5G communication has not yet returned to normal, it switches to Tiantong satellite communication to ensure basic communication. Depending on the recovery of 4G / 5G and Beidou, it automatically switches back to the highest priority mode.

[0077] External data (such as the status of the drone itself, environmental information, etc. obtained by the sensors on the drone, as well as instructions from the management center) is input into the multi-mode satellite communication-Ascend computing power integrated module, including: 1) real-time status, such as the drone's posture status, image frequency, etc.; 2) environmental information, such as obstacles, no-fly zones, etc.; 3) flight control command information, such as ascent, descent, return, etc.; Based on the Ascend NPU, it supports intelligent communication channel prediction and scheduling and flight control collaboration, realizes the coupling optimization of flight control commands and communication commands, combines module input and historical link data status indicators, evaluates and predicts the status of the three communication links of 4G / 5G, Beidou and Tiantong, and operates at the same time. Flight control algorithms such as path planning and obstacle avoidance are combined with flight status parameters to dynamically define link priorities based on flight instructions. At the same time, the Ascend NPU is used to extract features from large amounts of flight control and environmental inputs, such as images, using neural network algorithms to compress the data volume and meet the lightweight protocol requirements in low-bandwidth conditions. (This refers to optimizing data transmission methods and reducing redundant information in low-bandwidth environments to minimize data volume and ensure efficient transmission of critical information. This includes: using Ascend NPU hardware to accelerate the real-time operation of lightweight neural networks to extract features and compress image or environmental data; and using the NPU to analyze link quality in real time and dynamically adjust the compression rate.)

[0078] Furthermore, redundant communication links are defined according to the flight mission. At the same time, the redundant communication links are kept in hot standby mode, channels are regularly requested, and transmission contexts are preloaded to reduce switching delays. As shown in Table 2:

[0079] Table 2: Allocation relationship between primary links and redundant links

[0080]

[0081] Furthermore, commands, messages, etc. are ultimately sent to mobile phone / PAD mobile terminal devices, drones, and data management centers through the optimal communication mode.

[0082] like Figure 3As shown, external data (drone status, environmental graphics, flight control instructions, etc.) is obtained as a hardware board (i.e., the multi-mode satellite communication-Ascend computing power integrated module of the present invention). With the support of Ascend NPU computing power, the intelligent prediction and scheduling engine predicts and evaluates the link status and dynamically selects the optimal communication mode. The intelligent flight control collaboration engine uses a lightweight protocol to meet dynamic protocol requirements based on the flight control algorithm. Finally, through the optimal communication mode, instructions, messages, etc. are sent to mobile phone / PAD mobile terminal devices, drones, and data management centers.

[0083] The technical effects that the present invention can bring are as follows:

[0084] (1) Realize triple-mode redundant communication and full area coverage.

[0085] (2) Through the computing power of Ascend NPU, dynamic scheduling of three modes is achieved to solve the problems of insufficient regional coverage and unstable signals in single / dual communication modes.

[0086] (3) In combination with flight control algorithm requirements such as path planning, Ascend unifies scheduling and jointly optimizes protocol scheduling to achieve real-time resource allocation and reduce communication energy consumption.

[0087] (4) The priority of each module in the three-mode communication is clearly defined to improve scheduling efficiency.

[0088] In summary, the present invention provides a multi-mode satellite communication-Ascend computing power integrated module and data processing method, wherein the module includes: an Ascend NPU chip, which is used to run a scheduling model based on an AI algorithm, perform real-time prediction of the drone and evaluate the link delay, bandwidth and power consumption, and dynamically select the optimal communication mode among Tiantong satellite communication, Beidou third-generation short message communication and 4G / 5G communication according to the path planning algorithm and obstacle avoidance algorithm; a Tiantong satellite communication module, which is used to transmit the status log of the drone; a Beidou positioning and short message module, which is used to transmit the location, status and emergency messages of the drone; a 4G / 5G module, which is used to transmit real-time control instructions, high-definition pictures and high-definition video streams; an IMU module, which is used to perform combined navigation and positioning in combination with Beidou GNSS, provide autonomous navigation capabilities when the GNSS signal is interrupted, and correct GNSS multipath errors; an antenna, which is used to transmit and receive wireless signals. This invention integrates Tiantong satellite communication, Beidou third-generation short message communication and 4G / 5G communication, and combines it with protocol scheduling optimization based on Ascend computing power to achieve communication function integration, lightweight protocol, and a highly integrated general computing module, providing low-altitude unmanned aerial vehicles with wide-area coverage and precise control communication functions.

[0089] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or terminal comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or terminal comprising the element.

[0090] Of course, those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware (such as a processor, controller, etc.) through a computer program. The program can be stored in a computer-readable storage medium that can be read by a computer. When executed, the program can include the processes in the above-described method embodiments. The computer-readable storage medium can be a memory, a magnetic disk, an optical disk, etc.

[0091] It should be understood that the application of the present invention is not limited to the above examples. For those skilled in the art, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A multi-mode satellite communication and Ascend computing power integrated module, characterized by: The multi-mode satellite communication-Ascend computing power integrated module includes: The Ascend NPU chip is used to run AI-based scheduling models, making real-time predictions and evaluating link latency, bandwidth, and power consumption for drones. It also dynamically selects the optimal communication mode among Tiantong satellite communications, Beidou-3 short message communications, and 4G / 5G communications based on path planning and obstacle avoidance algorithms. Tiantong satellite communication module, used to transmit the drone's status log; Beidou positioning and short message module, used to transmit drone location, status and emergency messages; 4G / 5G module for transmitting real-time control instructions, high-definition images and high-definition video streams; The IMU module is used for integrated navigation and positioning in conjunction with Beidou GNSS, providing autonomous navigation capabilities when GNSS signals are interrupted, and correcting GNSS multipath errors; Antenna, used to transmit and receive wireless signals; The Ascend NPU chip aligns path coordinates, obstacle distribution, and real-time communication channel status into a unified space-time coordinate system; the Ascend NPU chip uses CNN to analyze the shape and density of obstacles on the path within a preset time in the future, and predicts the signal attenuation area; the Ascend NPU chip uses a reinforcement learning algorithm to integrate path risk, delay prediction, and energy consumption constraints, and pre-prioritizes 4G / 5G communication, Beidou-3 short message communication, and Tiantong satellite communication in descending order; monitors 4G / 5G signal strength and delay, and detects when the delay is greater than the preset delay threshold or the RSRP is less than the preset signal receiving power threshold. When the Beidou short message channel is congested or in a preset obstruction environment, and the 4G / 5G communication has not returned to normal, it switches to Tiantong satellite communication, defines redundant communication links according to the flight mission, and keeps the redundant communication links in hot backup mode. It applies for channels regularly, preloads the transmission context, and sends instructions and messages to mobile phones / PAD mobile terminal devices, drones, and data management centers through the optimal communication mode, and realizes end-to-end decision-making through Ascend parallel computing. The Ascend NPU chip dynamically allocates resources according to the NPU computing power to ensure the transmission of high-priority data such as obstacle avoidance instructions.

2. The multi-mode satellite communication-Ascend computing power integrated module according to claim 1, characterized in that: The antennas include: Tiantong external antenna, Beidou external antenna, 4G / 5G external antenna and WIFI / Bluetooth antenna.

3. The multi-mode satellite communication-Ascend computing power integrated module according to claim 1, characterized in that: The multi-mode satellite communication-Ascend computing power integrated module also includes: a communication interface, which is used to communicate with external devices.

4. The multi-mode satellite communication-Ascend computing power integrated module according to claim 3, characterized in that: The communication interfaces include: RS-232, RS-485, WIFI and Bluetooth.

5. A data processing method based on the multi-mode satellite communication-Ascend computing power integrated module according to any one of claims 1 to 4, characterized in that: The data processing method includes: Acquire external data and input the external data into the multi-mode satellite communication-Ascend computing power integrated module, wherein the external data includes the real-time status of the UAV, environmental information, and flight control command information; The Ascend NPU chip runs a scheduling model based on AI algorithms to make real-time predictions and evaluate link latency, bandwidth, and power consumption for drones. It also dynamically selects the optimal communication mode among Tiantong satellite communications, Beidou third-generation short message communications, and 4G / 5G communications based on path planning and obstacle avoidance algorithms.

6. The data processing method according to claim 5, characterized in that: The Ascend NPU chip runs an AI-based scheduling model to predict and evaluate the drone's link latency, bandwidth, and power consumption in real time. Based on path planning and obstacle avoidance algorithms, it dynamically selects the optimal communication mode among Tiantong satellite communications, Beidou-3 short message communications, and 4G / 5G communications. Specifically, the chip: The Ascend NPU chip aligns path coordinates, obstacle distribution, and real-time communication channel status into a unified spatiotemporal coordinate system. The Ascend NPU chip uses CNN to analyze the shape and density of obstacles on the path within a preset time in the future and predict signal attenuation areas; The Ascend NPU chip uses a reinforcement learning algorithm to integrate path risks, latency prediction, and energy consumption constraints to generate the optimal communication mode in Tiantong satellite communication, Beidou third-generation short message communication, and 4G / 5G communication, and achieves end-to-end decision-making through Ascend parallel computing; The Ascend NPU chip dynamically allocates resources based on the NPU computing power.

7. The data processing method according to claim 5, characterized in that: The path planning algorithm is used to generate a flight trajectory, and the obstacle avoidance algorithm is used to output a real-time obstacle heat map.

8. The data processing method according to claim 6, characterized in that: Generating the optimal communication mode in Tiantong satellite communication, Beidou third-generation short message communication and 4G / 5G communication specifically includes: Prioritize 4G / 5G communication, BeiDou-3 short message communication, and Tiantong satellite communication in descending order in advance; Monitor 4G / 5G signal strength and latency. When the latency is greater than the preset latency threshold or the RSRP is less than the preset signal receiving power threshold, activate BeiDou III short message communication. When the Beidou short message channel information is congested or in a preset blocking environment, and 4G / 5G communication has not returned to normal, switch to Tiantong satellite communication.

9. The data processing method according to claim 5, characterized in that: The Ascend NPU chip runs an AI-based scheduling model to predict and evaluate the drone's link latency, bandwidth, and power consumption in real time. Based on path planning and obstacle avoidance algorithms, it dynamically selects the optimal communication mode among Tiantong satellite communications, Beidou-3 short message communications, and 4G / 5G communications. It also includes: Redundant communication links are defined according to the flight mission, and the redundant communication links are kept in hot standby state. Channels are applied for regularly and transmission contexts are preloaded.

10. The data processing method according to claim 9, characterized in that: The method further includes defining redundant communication links according to flight missions, maintaining the redundant communication links in a hot standby state, periodically applying for channels, and preloading transmission contexts, and then: Through the optimal communication mode, commands and messages are sent to mobile phone / PAD mobile terminal devices, drones and data management centers.

Images