Robot dynamic networking system based on 3GPP

Abstract

The invention discloses a 3GPP (3rd Generation Partnership Project)-based robot dynamic networking system, which is characterized in that an intelligent robot dynamic networking mechanism provided by a 3GPP system can dynamically establish a communication group according to task requirements and spatial distribution, so that cooperative work among robots is ensured, wide-area communication and cross-regional cooperation can be supported, and the system is high in reliability and high in reliability. The communication range and structure can be flexibly adjusted according to the change of the task to ensure that the task execution is not limited by communication; a protocol and media type interoperability mechanism provided by the 3GPP system can realize intercommunication between robots of different platforms and devices, and can effectively solve the interoperability problem between the robots; the 3GPP system can ensure that robots from different manufacturers can realize seamless communication and cooperatively complete tasks through protocol conversion, media type conversion and data format standardization. The method is widely applied to the technical field of robot control.

Description

Technical Field

[0001] The present invention relates to the field of robot control technology, and in particular to a 3GPP-based robot dynamic networking system. Background Art

[0002] With the continuous advancement of industrial automation, intelligent robots are gradually replacing traditional human labor in industries such as construction and manufacturing, taking on complex, dangerous, and high-intensity tasks. Unlike traditional robots, intelligent robots possess autonomous decision-making and environmental perception capabilities, allowing them to flexibly adjust their working methods based on actual needs. In scenarios where these intelligent robots collaborate, they not only need to complete complex tasks but also rely on efficient communication systems to exchange information in real time.

[0003] In areas such as industrial production, transportation, and rescue operations, a single robot's limited capabilities and range of motion make it inefficient for the task. Consequently, multiple robots are often required to collaborate to perform the same task. However, these multiple intelligent robots often come from different manufacturers or platforms, using different protocols, media types, and technical standards. This heterogeneity poses significant challenges to inter-robot communication and collaboration. Summary of the Invention

[0004] In response to the technical problem of networking when multiple intelligent robots work together, the purpose of the present invention is to provide a robot dynamic networking system based on 3GPP.

[0005] In one aspect, an embodiment of the present invention includes a 3GPP-based robot dynamic networking system, wherein the 3GPP-based robot dynamic networking system includes the following steps:

[0006] Terminal layer; the terminal layer includes a plurality of terminal units corresponding to robots;

[0007] Access and transport layer; the access and transport layer is used to control the dynamic connection topology between each terminal unit and the core network layer;

[0008] Core network layer; the core network layer is used to provide 3GPP protocol access and dynamic resource allocation to each terminal unit.

[0009] Furthermore, the terminal unit includes:

[0010] Multimodal collaboration module; the multimodal collaboration module includes heterogeneous sensors, dynamic resource sensing units, inter-device direct communication units and action execution units; the heterogeneous sensors are used to collect sensor data, the dynamic resource sensing unit is used to sense its own communication resource data, and the inter-device direct communication unit is used to perform inter-device direct communication.

[0011] Protocol and media adapter; the protocol and media adapter is used to encapsulate the sensor data and the communication resource data into a first ROS message and send it to the outside, and receive a second ROS message from the outside to parse and obtain an action instruction;

[0012] The action execution unit is used to execute the action instruction.

[0013] Furthermore, the access and transport layer includes:

[0014] Heterogeneous wireless access module; the heterogeneous wireless access module is used to establish a heterogeneous wireless connection with the terminal unit and to send and receive the first ROS message and the second ROS message;

[0015] Dynamic topology management module; the dynamic topology management module is used to perform QoE monitoring according to the first ROS message, perform reinforcement learning to predict link quality according to the QoE monitoring result, and switch the connection path of the terminal unit according to the link quality prediction result.

[0016] Furthermore, the switching of the connection path of the terminal unit according to the link quality prediction result includes:

[0017] Determining a next hop of a first terminal unit according to a link quality prediction result; the first terminal unit is any terminal unit;

[0018] When the next hop of the first terminal unit is the heterogeneous wireless access module, triggering the first terminal unit to connect to the heterogeneous wireless access module;

[0019] When the next hop of the first terminal unit is a second terminal unit, a direct device connection between the first terminal unit and the second terminal unit is triggered; the second terminal unit is a terminal unit other than the first terminal unit.

[0020] Furthermore, the core network layer includes:

[0021] A network function virtualization module; the network function virtualization module is used to run a network slice orchestrator to dynamically allocate slice resources to each terminal unit;

[0022] Protocol / media interoperability gateway; the protocol / media interoperability gateway is used to run the protocol state machine model and perform communication protocol conversion corresponding to the first ROS message and / or the second ROS message.

[0023] Furthermore, the 3GPP-based robot dynamic networking system also includes:

[0024] Application and service layer: The application and service layer is used to provide media application services to the terminal unit.

[0025] Furthermore, the application and service layer includes:

[0026] A task scheduling platform; the task scheduling platform is used to open network capabilities to the first ROS message and cooperate with the terminal unit to perform the robot task according to the first ROS message;

[0027] Security and privacy module; the security and privacy module is used to run the 3GPP enhanced security mechanism to perform data traceability and integrity verification on the first ROS message.

[0028] Furthermore, the application and service layer is connected to the core network layer via a 3GPP API, and the core network layer is connected to the access and transport layer via an NGAP / Xn / N2 interface.

[0029] Furthermore, the terminal layer includes a plurality of robots, and any of the robots includes one or more terminal units;

[0030] Each of the terminal units is combined into one or more terminal unit groups, and any of the terminal unit groups includes one or more of the terminal units;

[0031] For any of the terminal unit groups, one of the terminal units in the terminal unit group communicates with the outside world of the terminal unit group, while the other terminal units in the terminal unit group keep their communications with the outside world hidden, and direct device-to-device communication is carried out between the terminal units in the terminal unit group.

[0032] Furthermore, each of the terminal units is combined into one or more terminal unit groups, including:

[0033] When working in the first networking mode, all the terminal units belonging to the same robot are combined into the same terminal unit group, and the terminal units belonging to different robots are combined into different terminal unit groups;

[0034] When operating in the second networking mode, determining a combination range, combining all the terminal units belonging to the combination range into a first terminal unit group, and for any terminal unit not belonging to the combination range, combining the terminal unit alone or with terminal units belonging to the same robot into a corresponding second terminal unit group;

[0035] When working in the third networking mode, each of the terminal units is determined as a corresponding terminal unit group.

[0036] On the other hand, an embodiment of the present invention also includes a computer device, including a memory and a processor, the memory is used to store at least one program, and the processor is used to load at least one program to execute the 3GPP-based robot dynamic networking system in the embodiment.

[0037] On the other hand, an embodiment of the present invention also includes a computer-readable storage medium, which stores a program executable by a processor. When the program executable by the processor is executed by the processor, it is used to execute the 3GPP-based robot dynamic networking system in the embodiment.

[0038] The beneficial effects of the present invention are as follows: the 3GPP-based robot dynamic networking system in the embodiment, the intelligent robot dynamic networking mechanism provided by the 3GPP system can dynamically establish a communication group according to task requirements and spatial distribution, ensure the collaborative work between robots, support wide area communication and cross-regional collaboration, and can flexibly adjust the communication range and structure according to changes in tasks to ensure that task execution is not restricted by communication; the protocol and media type interoperability mechanism provided by the 3GPP system can realize intercommunication between robots of different platforms and devices, and can effectively solve the interoperability problem between robots; the 3GPP system can ensure that robots from different manufacturers can achieve seamless communication and collaboratively complete tasks through protocol conversion, media type conversion and data format standardization. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of the structure of a 3GPP-based robot dynamic networking system in an embodiment;

[0040] Figure 2 Schematic diagram of interfaces between various layers of a 3GPP-based robot dynamic networking system in an embodiment;

[0041] Figure 3 Schematic diagram of the robot and terminal unit in the embodiment;

[0042] Figure 4 Schematic diagram of a dynamic networking mechanism in an embodiment;

[0043] Figure 5 Schematic diagram of the protocol / media interoperability mechanism in an embodiment;

[0044] Figure 6 Schematic diagram of the first networking mode in the embodiment;

[0045] Figure 7 Schematic diagram of the second networking mode in the embodiment;

[0046] Figure 8 Schematic diagram of the third networking mode in the embodiment. DETAILED DESCRIPTION

[0047] Explanation of terms:

[0048] Multimodal robot collaboration module: supports collaborative work between different types of robots (such as AGVs, drones, and robotic arms), and is equipped with multiple sensors (LiDAR, cameras, IMUs, etc.) to achieve environmental perception and navigation.

[0049] Network Function Virtualization (NFV) and Edge Computing: Distributed UPF is deployed on MEC edge nodes to provide local protocol conversion services (such as ROS to HTTP / 2) and accelerate media stream processing. The network slice orchestrator dynamically allocates network resources based on application requirements.

[0050] D2D Communication Module: This module is particularly suitable for low-latency applications such as SLAM data transmission. It supports direct device-to-device (D2D) communication technology.

[0051] Wireless Access Point (AP): A wireless access point is a network device used to connect wireless devices (such as smart robots, mobile phones, etc.) to a wired network or the Internet. It acts as a bridge between wireless devices and wired networks.

[0052] Wireless Local Area Network (WLAN): A local area network built using wireless communication technologies (such as Wi-Fi) that allows devices to communicate wirelessly within a certain range.

[0053] Dynamic wireless network: A wireless network that can adjust its topology in real time based on device distribution, task requirements, or network conditions.

[0054] Bluetooth Low Energy (BLE): BLE, Bluetooth Low Energy

[0055] Bluetooth Low Energy (BLE) is a technology designed to significantly reduce the power consumption of Bluetooth devices, making it suitable for battery-powered devices that require long-term operation. BLE allows devices to maintain a connection while consuming minimal energy, making it ideal for applications such as sensor data transmission and status updates, where small amounts of data are exchanged.

[0056] Bluetooth SIG (Special Interest Group): is an international industry association responsible for managing and promoting Bluetooth technology standards.

[0057] AGV (Automated Guided Vehicle): A mobile robot that can navigate autonomously along a predetermined path or through environmental perception, commonly used for material handling tasks in logistics and manufacturing.

[0058] Unmanned aerial vehicle (UAV): An aircraft that can fly without a human pilot on board and can be used in a variety of application scenarios such as aerial photography, monitoring, and delivery.

[0059] ROS2.0 message encapsulation: Robot Operating System (ROS) 2.0 is a flexible framework for writing robot software. Its message encapsulation mechanism defines how to pass information between nodes, supports complex data structures, and ensures efficient and reliable communication.

[0060] JSON / Protobuf data conversion: JSON (JavaScript Object Notation) is a lightweight data exchange format that's easy for humans to read and write. Protocol Buffers (Protocol Buffers), a binary serialization format developed by Google, is more compact and efficient. Conversion between the two allows data to be seamlessly shared across different systems or applications.

[0061] URLLC (Ultra-Reliable Low Latency Communication): A key feature of 5G technology designed to provide extremely low latency and very high reliability, suitable for time-sensitive tasks such as autonomous vehicle control or remote surgery.

[0062] eMBB (enhanced mobile broadband): Another core feature of 5G technology, it focuses on significantly improving the data rate and capacity of mobile networks to support bandwidth-intensive applications such as high-definition video streaming and virtual reality.

[0063] Open interface NEF: Network Exposure Function (defined in 3GPP TS23.501): An open interface for 5G core network capabilities that allows external AI applications (such as task allocation engines) to access network status (such as current slice bandwidth utilization) through standardized APIs, enabling cross-domain collaborative optimization.

[0064] 5GC Core Network: 5G Core Network, including standardized control plane and user plane functional components (such as AMF, SMF, UPF), customized for this patent to extend the protocol interoperability capabilities.

[0065] SMF: Session Management Function (3GPP TS23.501): Session management function that allocates specific slices for robot communication sessions (such as URLLC slices bound to ROS control flows) and triggers the activation of the protocol conversion engine in the UPF.

[0066] AMF / NSSF: AMF: Access and Mobility Management Function (AMF), which handles robot access authentication and location updates. NSSF: Network Slice Selection Function, which selects the optimal network slice instance based on terminal requirements.

[0067] 5G NR-U base station: A base station based on the 5G NR-U (New Radio-Unlicensed) standard, operating in unlicensed frequency bands (such as 5 GHz and 6 GHz), supports coexistence with Wi-Fi 7 (dynamic spectrum sharing), and provides wide-area coverage and high reliability.

[0068] Wi-Fi 7AP: An IEEE 802.11be-compliant access point that supports Multi-Link Deployment (MLD) and 4096-QAM modulation, providing high-throughput (>40Gbps) short-range transmission for robots and backhauling large files (such as point cloud maps).

[0069] Mesh D2D: A self-organizing mesh network consisting of direct device-to-device connections, supporting multi-hop relays, providing redundant paths for robots in base station coverage blind spots (such as factory basements), and dynamically maintaining topology based on the AODV protocol.

[0070] NGAP / Xn / N2 Interface: NGAP: Control plane interface between NG-RAN and AMF (TS 38.413), used for base station registration and bearer management. Xn: Interface between base stations, supporting handover and interference coordination (TS 38.420). N2: Interface between AMF and base stations (TS 38.413), used to transmit session control information.

[0071] L1 / L2 wireless interface: L1: physical layer interface (such as OFDM waveform, modulation and coding scheme).

[0072] Currently, when building a network for intelligent robots, you can consider either Wi-Fi or Bluetooth. However, these networking technologies each have the following disadvantages:

[0073] (1) Disadvantages of Wi-Fi-based intelligent robot networking solutions

[0074] 1. WLAN signals are susceptible to interference from other wireless devices, such as microwave ovens, other WLAN networks, and Bluetooth devices. This interference can lead to decreased communication quality, packet loss, and increased latency. Furthermore, in intelligent robotic systems, components must communicate data in real time and reliably. WLAN interference can cause delays or loss of robot control commands, impacting the robot's real-time response and precise control, making dynamic networking between intelligent robots highly unstable.

[0075] 2. WLAN communications may be subject to the risk of data leakage and unauthorized access, requiring additional security measures to protect communication security. Data leakage and unauthorized access risks can affect the trust relationship between different devices and the security of data exchange. This can lead to impaired protocol interoperability, as devices may be unable to securely exchange data or perform necessary authentication and authorization steps.

[0076] (2) Disadvantages of Bluetooth-based intelligent robot networking solutions

[0077] 1. Bluetooth technology has a relatively short transmission range, typically around 10 meters (although this range can be extended by increasing transmission power and optimizing receiver sensitivity, it still has limitations). In intelligent robotic systems that require wide-range communication, the limited transmission range of Bluetooth technology may result in poor communication between robots or require the deployment of more Bluetooth devices to cover the entire work area. This increases system complexity and cost.

[0078] 2. In intelligent robotic systems, if the robot needs to transmit large amounts of data (such as high-definition images, video streams, or extensive sensor data), the bandwidth limitations of Bluetooth technology may result in slow or unstable data transmission. This can affect the robot's real-time performance and decision-making capabilities. Furthermore, bandwidth limitations may affect protocol interoperability and media compatibility between different devices. For example, certain media types may require higher bandwidth to transmit high-quality content, and Bluetooth technology may not be able to meet these requirements. This can lead to communication barriers between devices or necessitate the use of alternative communication technologies to replace Bluetooth.

[0079] Therefore, there are some limitations in the application of networking technologies such as Bluetooth and WiFi in intelligent robot collaboration scenarios.

[0080] In this embodiment, a robot dynamic networking system based on 3GPP is provided. 3GPP is not limited to traditional open interconnection model specifications, it also involves the architectural design of the entire mobile communication system, including but not limited to RAN (radio access network), EPC (evolved packet core), 5GC (5G core network) and other components. It provides a complete set of standardized solutions from underlying hardware to high-level applications for smart devices (such as smart robots), ensuring that products from different manufacturers can work together in the same network environment, and continue to improve and expand as technology develops. The dynamic networking mechanism, protocol and media type interoperability mechanism, and the ability of multimodal information transmission to work in conjunction with AI models provided by the 3GPP system can better meet the communication and collaboration needs in intelligent robot collaboration scenarios.

[0081] Reference Figure 1 , based on 3GPP, the robot dynamic networking system includes the terminal layer, access and transport layer, core network layer and application and service layer. Figure 2 The application and service layer is connected to the core network layer through the 3GPP API, the core network layer is connected to the access and transport layer through the NGAP / Xn / N2 interface, and the access and transport layer is connected to the terminal layer through the L1 / L2 interface.

[0082] The terminal layer includes a plurality of terminal units corresponding to the robots. Figure 3 The terminal layer includes multiple robots, such as Robot 1, Robot 2, and Robot 3. Each robot can be a single device, such as an AGV (automated guided vehicle), UAV (unmanned aerial vehicle), robotic arm, humanoid robot, or bionic robot. Each robot is equipped with multiple terminal units, or one robot serves as a terminal unit. For example, Robot 1 includes multiple terminal units, such as Terminal Unit 1, Terminal Unit 2, and Terminal Unit 3, while Robot 2 as a whole serves as Terminal Unit 4. When a robot is equipped with multiple terminal units, each controllable component on the robot can be regarded as a terminal unit. For example, if Robot 1 is a humanoid robot, Terminal Unit 1 on Robot 1 can be the head of the humanoid robot, Terminal Unit 2 can be the hand of the humanoid robot, and Terminal Unit 3 can be the foot of the humanoid robot.

[0083] In this embodiment, refer to Figure 1Each terminal unit is equipped with a multimodal collaboration module and a protocol and media adapter. Let's take one of the terminal units as an example for illustration. The multimodal collaboration module includes components such as heterogeneous sensors, dynamic resource perception units, inter-device direct communication units, and action execution units. Among them, heterogeneous sensors include LiDAR (laser radar), cameras, IMU (Inertial Measurement Unit) and other sensors with different principles, which can be used to collect sensor data such as environmental point clouds, optical images and their own positions; the dynamic resource perception unit is used to perceive its own communication resource data, specifically including the terminal unit's own power, RTT (Round Trip Time), packet loss rate and jitter data; the inter-device direct communication unit can be used to perform inter-device direct communication (D2D, a communication protocol in the 3GPP system), for example, establishing inter-device direct communication with the inter-device direct communication units in other terminal units to transmit various data including the first ROS message and the second ROS message; the protocol and media adapter encapsulates the sensor data and communication resource data collected by the multimodal collaboration module into a first ROS message and sends it to the outside, and receives the second ROS message from the outside to parse and obtain action instructions. That is, for any terminal unit, the first ROS message is the data sent by the terminal unit to the outside world. The first ROS message may be received by other terminal units or the access and transport layer. The second ROS message is the data received by the terminal unit from the outside world. The second ROS message may come from other terminal units or the access and transport layer. Both the first and second ROS messages are data encapsulated according to the ROS message standard (for example, ROS 2.0).

[0084] In this embodiment, the protocol and media adapter can also be compatible with 3GPP communication protocols (such as HTTP / 3, TCP / IP) and non-3GPP communication protocols (such as MQTT, CoAP), dynamically switch according to business scenarios, and support ROS2.0 message encapsulation, H.265 video stream and JSON / Protobuf data conversion to meet low latency (URLLC) and high bandwidth (eMBB) requirements.

[0085] In this embodiment, the action execution unit can specifically be a component such as a motor, a hydraulic press, or a laser transmitter that can be controlled to perform actions such as rotation, extension, and signal transmission. When the second ROS message includes an action instruction (such as controlling the movement direction or speed), the action execution unit executes the action instruction.

[0086] In this embodiment, refer to Figure 1, the access and transport layer includes a heterogeneous wireless access module and a dynamic topology management module. Among them, the heterogeneous wireless access module includes access devices such as 5G NR-U base stations and Wifi7AP, which can establish heterogeneous wireless connections with terminal units, and receive and send the first ROS message and the second ROS message to each terminal unit respectively. Specifically, the heterogeneous wireless access module can send the second ROS message that needs to be sent to a terminal unit so that the terminal unit can receive the second ROS message; if a terminal unit sends a first ROS message, the heterogeneous wireless access module can receive the first ROS message and send it to the core network layer for further processing.

[0087] In this embodiment, the dynamic topology management module is used to monitor QoE based on the first ROS message sent by each terminal unit, perform reinforcement learning to predict link quality based on the QoE monitoring results, and switch the connection path of the terminal unit based on the link quality prediction results. Specifically, the dynamic topology management module can run the AI-Driven routing optimizer (reinforcement learning model) and the QoE monitoring module (real-time measurement of RTT / packet loss rate / jitter). The algorithm flow of the dynamic topology management module for dynamic topology management can be summarized as follows:

[0088] IF (wireless access layer link quality degrades) THEN

[0089] 1. QoE monitoring detects a threshold violation (such as RTT > 50ms or packet loss rate ≥ 5%)

[0090] 2. AI route optimizer triggers handover decision (see algorithm below)

[0091] 3. Dynamic topology manager updates the connection path

[0092] ELSE

[0093] Maintain the current primary path (5G NR-U or Wi-Fi 7 dual connection)

[0094] END IF

[0095] Among them, the reinforcement learning routing decision algorithm is adjusted according to the following contents:

[0096] Network status: available bandwidth, interference level (based on CSI reports), base station load

[0097] Terminal status: remaining battery power of the robot, D2D neighbor list (ProSe protocol discovery results)

[0098] QoE indicators: RTT, packet loss rate, and jitter.

[0099] In this embodiment, for any two terminal units among the plurality of terminal units, namely the first terminal unit (specifically, Figure 1 、 Figure 2 or Figure 3 Terminal unit 1 in the example) and the second terminal unit (specifically, Figure 1 or Figure 3 In the terminal unit 2), the dynamic topology management module may perform the following steps when switching the connection path of the terminal unit according to the link quality prediction result:

[0100] S1 determines the next hop of the first terminal unit (terminal unit 1) based on the link quality prediction results;

[0101] S2. When the next hop of the first terminal unit (terminal unit 1) is a heterogeneous wireless access module, the first terminal unit (terminal unit 1) is triggered to connect to the heterogeneous wireless access module;

[0102] S3. When the next hop of the first terminal unit (terminal unit 1) is the second terminal unit (terminal unit 2), a direct device connection is triggered between the first terminal unit (terminal unit 1) and the second terminal unit (terminal unit 2).

[0103] By executing steps S1-S3, the dynamic topology management module can dynamically control the topological connection relationship between each terminal unit and between the terminal unit and the heterogeneous wireless access module. Specifically, a terminal unit can choose to establish and maintain a connection with the heterogeneous wireless access module, or can choose to establish a direct device-to-device connection with another terminal unit, and ultimately connect to the heterogeneous wireless access module through another terminal unit (or more terminal units), thereby optimizing the communication quality between the terminal unit and the access and transport layers.

[0104] By setting up a dynamic topology management module and a heterogeneous wireless access module, the access and transport layer can control the dynamic connection topology between each terminal unit and the core network layer.

[0105] In this embodiment, refer to Figure 1 The core network layer includes a network function virtualization module and a protocol / media interoperability gateway. Among them, the network function virtualization module runs the network slice orchestrator to dynamically allocate slice resources to each terminal unit. Specifically, the network function virtualization module uses distributed UPF deployed on the MEC edge node for edge computing, and allocates corresponding types of slices according to the type of terminal unit. For example, the network function virtualization module allocates high-reliability slices to some types of terminal units (such as terminal units belonging to AGVs) and allocates large-bandwidth slices to other types of terminal units (such as terminal units belonging to inspection drones), so that different types of terminal units are allocated to appropriate slices.

[0106] In this embodiment, the first ROS message sent by the terminal layer and the second ROS message received by the terminal layer are non-3GPP protocol data, while the core network layer and the application and service layer process 3GPP protocol data. Therefore, the protocol / media interoperability gateway can run the protocol state machine model as a cross-domain protocol conversion engine to perform communication protocol conversion corresponding to the first ROS message and / or the second ROS message. Specifically, the protocol / media interoperability gateway can convert the first ROS message sent by the terminal unit in the terminal layer into 3GPP protocol data for processing by the core network layer and the application and service layer; and can convert the 3GPP protocol data processed by the core network layer and the application and service layer into a second ROS message, and send the second ROS message to the corresponding terminal unit.

[0107] By setting up a network function virtualization module and a protocol / media interoperability gateway, the core network layer can provide 3GPP protocol access and dynamic resource allocation to each terminal unit.

[0108] In this embodiment, refer to Figure 1 and Figure 2 , the application and service layer includes a task scheduling platform and a security and privacy module. Among them, the security and privacy module is used to run the 3GPP enhanced security mechanism to perform data traceability and integrity verification on the first ROS message. For example, the security and privacy module integrates the secondary authentication of 5G AKA+ robot device fingerprint and SEPP (Security Edge Protection Proxies) to ensure cross-domain data transmission; the security and privacy module can also use the media watermark injection function to embed a steganographic watermark into the video stream and other data sent to the terminal unit, so that the data returned by the legally connected robot also contains the steganographic watermark. By detecting whether the data returned by the robot contains the steganographic watermark, the robot data can be traced and the integrity verified.

[0109] In this embodiment, the task scheduling platform is used to open the network capability to the first ROS message, and cooperate with the terminal unit to perform the robot task according to the first ROS message. Specifically, the task scheduling platform is provided with a digital twin interface, and the digital twin interface can use the 3GPP open network capability (Network Exposure Function, NEF) to provide terminal units with business services such as positioning accuracy and bandwidth prediction. For example, the terminal module can package the request for services such as positioning accuracy and bandwidth prediction into a first ROS message, and send it to the access and transport layer through direct communication between devices or directly send it to the access and transport layer. After the first ROS message is converted by the protocol of the core network layer, it is finally sent to the task scheduling platform. The task scheduling platform can run services such as positioning accuracy and bandwidth prediction through the digital twin interface, and package the data generated by the service into a second ROS message. After the second ROS message is converted by the protocol of the core network layer, it is sent back to the terminal module requesting the service through the access and transport layer, thereby realizing the operation of the service.

[0110] Specifically, the first ROS message can be the original data such as point cloud collected by the terminal unit. After processing such as protocol conversion at the core network layer, the task scheduling platform runs the rendering program to render the point cloud data in the first ROS message, thereby obtaining media data with a certain bit rate and standardized format.

[0111] In this embodiment, the task scheduling platform also uses an AI task allocation engine to combine network status data provided by the core network layer to optimize multi-robot task decomposition. Specifically, in the case of multiple terminal units, the core network layer can obtain the network status data of each terminal unit, detect terminal units with poor network status (for example, in high-latency areas) and terminal units with good network status, and prioritize the first ROS messages sent by terminal units with good network status, thereby dynamically avoiding high-latency areas.

[0112] By setting up a task scheduling platform and security and privacy modules, the application and service layer can provide media application services to terminal units.

[0113] In this embodiment, by Figure 1 and Figure 2 The 3GPP-based robot dynamic networking system shown can implement Figure 4 The dynamic networking mechanism shown.

[0114] Reference Figure 4, the terminal unit can initiate a service request to the access and transport layer, and the access and transport layer requests the core network layer to allocate network slices; the access and transport layer monitors the link quality between the terminal unit, and in the case of a link quality degradation alarm (for example, RTT is greater than 50ms, packet loss rate is greater than 8%), the access and transport layer calculates the optimal path through the AI ​​routing optimizer as direct communication between devices. After the core network layer updates the path to direct communication between devices, it triggers the terminal unit to switch the path, so that the terminal unit establishes direct communication between devices with other terminal units, rather than directly establishing a connection with the access and transport layer, so that the terminal unit and the access and transport layer always maintain communication.

[0115] pass Figure 4 The dynamic networking mechanism shown can achieve the following technical effects:

[0116] 1. Significant improvement in technical performance

[0117] Scenario effect: In industrial inspection scenarios (dynamic occlusion and complex environments), the average latency of robot collaboration tasks is reduced from 120ms to 65ms (a 45% decrease), and reliability (SLA compliance rate) is improved to 99.99% (in line with URLLC slicing requirements).

[0118] Technical support:

[0119] The AI ​​route optimizer's multi-path prediction mechanism dynamically avoids base station overload areas (for example, triggering D2D handover 500ms in advance when congestion is predicted).

[0120] D2D multi-hop relay reduces latency jitter from ±30ms to ±8ms through node collaboration within 3 hops (based on TS23.303 ProSe protocol).

[0121] (2) Enhanced robustness

[0122] Anti-interference capability: In 5G NR-U and Wi-Fi 7 dual-connection mode (DC primary and backup links), the packet loss rate is reduced from 15% to 3% in a strong electromagnetic interference environment (based on real-time channel switching detection of the QoE module).

[0123] Self-healing capability: When the D2D network is broken, AI-driven topology management can rebuild the local Mesh within 200ms (patented technology dynamic redundant path planning).

[0124] 2. Resource efficiency optimization

[0125] (1) Dynamic reuse of network resources

[0126] Slice sharing gain: The network slice orchestrator uses a "time-division multiplexing" mechanism to automatically allocate bandwidth to drone video slices during idle periods of the AGV control slice, increasing resource utilization from 60% to 92%.

[0127] Spectral efficiency: In NR-U and Wi-Fi 7 dynamic spectrum sharing (DSS) mode, the average spectral efficiency reaches 8.7bps / Hz (a 35% increase compared to traditional fixed allocation).

[0128] (2) Reduction of terminal energy consumption

[0129] Adaptive sleep mechanism: The low-power mode of the D2D link (RRC_Inactive state extension) reduces the robot's communication energy consumption by 45% (battery life is extended to 4.3 hours in a 1-hour mission cycle).

[0130] Green routing selection: The reinforcement learning algorithm prioritizes low-energy consumption paths (e.g., robots with battery power <30% prefer to use Wi-Fi7 backhaul instead of D2D relay).

[0131] In this embodiment, by Figure 1 and Figure 2 The 3GPP-based robot dynamic networking system shown can implement Figure 5 The protocol / media interoperability mechanism shown.

[0132] Reference Figure 5 After receiving the first ROS message from the terminal unit, the access and transport layer calls the protocol / media interoperability gateway in the core network layer to perform protocol conversion. Specifically, the protocol / media interoperability gateway, as a protocol conversion engine, parses the first ROS message according to the predefined state machine model:

[0133] ROS Topic → Protobuf intermediate format (field mapping)

[0134] Protobuf → HTTP / 3 over QUIC (header encapsulation is 3GPP standard fields)

[0135] This obtains 3GPP protocol data, which the edge UPF verifies and forwards to the application and service layer, ensuring URLLC latency requirements (<10ms). The application and service layer then performs rendering and other processing. The application and service layer can return a JSON response to the terminal unit through the access and transport layer. The protocol conversion engine processes it into a second ROS message adapted for the terminal unit. The specific content of the second ROS message can be a media format video, etc., for the terminal unit to play and other processing.

[0136] In this embodiment, the terminal unit can also upload LiDAR point cloud data as the first ROS message. The core network layer encodes the LiDAR point cloud data in the first ROS message into a standardized MPEG-21 DID structure, and the edge media encapsulation engine converts the media format according to the DID description:

[0137] Point Cloud → 3D mesh simplification (based on edge GPU computing) → H.265 frame sequence (TS26.265). Finally, the digital twin platform applied in the service layer receives and renders the H.265 frame sequence, and requests core network bandwidth prediction through the NEF interface to adjust the bit rate.

[0138] pass Figure 5 The protocol / media interoperability mechanism shown can achieve the following technical effects:

[0139] 1. Technical performance breakthrough: low-latency and highly reliable communication

[0140] (1) Improved protocol conversion efficiency

[0141] End-to-end latency: The ROS→HTTP / 3 conversion latency is reduced from 15ms in the traditional solution to 3ms (an 80% reduction), meeting the URLLC slicing latency requirements (<10ms).

[0142] Technical support:

[0143] Lightweight state machine model: Based on XML predefined mapping rules (such as directly binding ROS Topic to HTTP / 3 Request-URI), it avoids the complexity of traditional syntax parsing (such as regular expression matching).

[0144] Edge UPF offload verification: Complete protocol header encapsulation at the edge node (QUIC connection ID reuse reduces handshake overhead) and reduce core network processing delay.

[0145] (2) Media processing fidelity and real-time performance

[0146] Point cloud → H.265 conversion quality: Under GPU acceleration, the accuracy loss of 3D mesh simplification is ≤5% (compared to the average of 12%), and the video stream PSNR value reaches 42dB (UHD standard).

[0147] Indicator optimization:

[0148] Real-time frame rate: 30FPS point cloud stream conversion latency is stable at 25ms (edge ​​GPU utilization rate is 65%, traditional centralized processing requires 100ms).

[0149] Adaptive bitrate: Dynamically adjusts the H.265 compression rate (QP value) based on core network bandwidth prediction, saving 30% bandwidth (e.g., 20Mbps→14Mbps).

[0150] 2. Resource efficiency optimization: dual gains in computing and transmission

[0151] (1) Optimization of computing resource utilization

[0152] Protocol conversion load reduction: Protobuf intermediate format reduces serialization / deserialization CPU overhead (traditional JSON parsing takes 15% CPU → Protobuf takes 5%).

[0153] Edge hardware acceleration: Dedicated media processing ASICs (such as Intel QuickAssist) in edge UPF carry out H.265 encoding, freeing up the GPU for robot SLAM algorithms.

[0154] (2) Dynamic Adaptation of Transmission Resources

[0155] Metadata compression: MPEG-21 DID identification compresses point cloud description information from 1.2KB to 256B (a 79% reduction), reducing signaling overhead.

[0156] Bandwidth efficiency: The DID standardized format enables mixed streaming across media types (e.g., point cloud + video sharing a single path), with a multiplexing gain of 40%.

[0157] In this embodiment, Figure 3 Each terminal unit in the robot shown can choose to work in the first networking mode, the second networking mode or the third networking mode.

[0158] In this embodiment, the first networking mode is as follows Figure 6 As shown. Figure 6All terminal units belonging to the same robot are grouped into the same terminal unit group, while terminal units belonging to different robots are grouped into different terminal unit groups. For example, terminal units 1, 2, and 3 belonging to robot 1 belong to the same terminal unit group, while terminal unit 4 belonging to robot 2 belongs to a separate terminal unit group, while terminal units 2 and 4 belong to different terminal unit groups. Each terminal unit in the same terminal unit group, such as terminal unit 1, terminal unit 2 and terminal unit 3, conducts direct device-to-device communication, and one of the terminal units (specifically, a terminal unit can be randomly selected, or the terminal unit with the best communication quality with the access and transport layer, such as terminal unit 1) communicates with the outside of the terminal unit group, including communication with the access and transport layer, and direct device-to-device communication with terminal units belonging to other terminal unit groups, such as terminal unit 4, terminal unit 5, and terminal unit 6, while other terminal units (including terminal unit 2 and terminal unit 3) in the same terminal unit group as terminal unit 1 maintain communication concealment with the outside of the terminal unit group, for example, by hiding the device IP address, domain name and other network internal topology information through the Security and Border Proxy (SEPP) in 3GPP, so that the external network cannot directly discover terminal unit 2 and terminal unit 3, or by hiding the User Identity Hiding (SUPI) in 3GPP. Concealment) ensures that only the authorized device that holds the private key in the home network (i.e., terminal unit 1 in the same terminal unit group) can decrypt and identify the user identities of terminal unit 2 and terminal unit 3. Other devices (such as the access and transport layer or terminal unit 4) cannot obtain the real user identification of terminal unit 2 and terminal unit 3, thereby achieving communication concealment between terminal unit 2 and terminal unit 3.

[0159] In this embodiment, by having one terminal unit in the same terminal unit group be responsible for communicating with the outside world, and the other terminal units be responsible for keeping communications with the outside world hidden, direct device-to-device communication is carried out between the terminal units in the terminal unit group, so that communication can be carried out in units of terminal unit groups. Each terminal unit can send and receive data, and the number of communicating units is reduced, which is conducive to reducing the load on equipment such as the access and transmission layers, thereby improving communication efficiency.

[0160] In this embodiment, the first networking mode divides the terminal unit groups into robot-based units, i.e., each robot corresponds to a terminal unit group, making it easier to control the entire robot. The first networking mode can be set to normal mode or default mode. For example, if no abnormality occurs, the first networking mode is used by default.

[0161] In this embodiment, the second networking mode is as follows Figure 7 As shown. Figure 7Although terminal units 2, 4, and 5 belong to different robots, they are combined into the same terminal unit group (the first terminal unit group) because they are within the same combination range. Terminal units that are not within the combination range are combined into a corresponding second terminal unit group in other ways. For example, terminal units 1 and 3 are not within the combination range, but they both belong to robot 1. Therefore, terminal units 1 and 3 are combined into the same second terminal unit group; terminal unit 6 is not within the combination range, but other terminal units (terminal unit 5) that belong to robot 3 have been combined into the first terminal unit group. Therefore, terminal unit 6 belongs to a second terminal unit group alone.

[0162] In this embodiment, the second networking mode can realize the division of terminal unit groups across robots, that is, multiple terminal units belonging to different robots can also be combined into the same terminal unit group (first terminal unit group). Since direct inter-device communication is carried out between the terminal units in the same terminal unit group, one of the terminal units is selected to be responsible for external communication, and the other terminal units keep the communication hidden from the outside. Therefore, from the perspective of external control, the terminal units in the first terminal unit group are equivalent to a robot as a whole, but in fact, the terminal units in the first terminal unit group can belong to different robots. For example, each terminal unit is located in different spatial positions and realizes the functions of different robots. Therefore, the first terminal unit group composed of the second networking mode can realize the functional combination of different robots and cross-space combination, thereby breaking through the limitations of the robot's physical form and spatial movement.

[0163] For example, if robot 1 is a drone, robot 2 is a robotic arm, and robot 3 is an AGV, a combination range can be set based on factors such as functional requirements. Terminal Unit 2 on robot 1 (specifically, a camera module mounted on the drone), Terminal Unit 4 on robot 2 (specifically, the robotic arm itself), and Terminal Unit 6 on robot 3 (specifically, a laser detection module on the AGV) are all set within the combination range. This combination results in a first terminal unit group consisting of Terminal Unit 2, Terminal Unit 4, and Terminal Unit 6. This first terminal unit group is equivalent to a robot consisting of a camera module with flight capabilities, a robotic arm with gripping and other functions, and a laser detection module with ground driving capabilities. Its spatial distribution and mobility are stronger than those of a single physical robot, enabling a wider range of perception and movement. The second networking mode can be set to enhanced mode, or it can be used when some terminal units on certain robots fail while others are functioning normally. In this case, the functioning terminal units on the robots can be combined with the terminal units on the other robots to form the first terminal unit group, thereby improving the efficiency of terminal unit utilization.

[0164] In this embodiment, the third networking mode is as follows Figure 8 As shown. Figure 8 , each terminal unit is identified as a corresponding terminal unit group, that is, the terminal units do not communicate directly with each other, but directly connect to the access and transport layers respectively. The third networking mode can be used when communication conditions are good.

[0165] In this embodiment, by running the first networking mode, the second networking mode and the third networking mode, the advantages of the 3GPP communication protocol can be utilized to achieve flexible switching of the robot's functional forms and flexible control of the robot.

[0166] The robot dynamic networking system based on 3GPP in this embodiment, the intelligent robot dynamic networking mechanism provided by the 3GPP system can dynamically establish communication groups according to task requirements and spatial distribution, ensure the collaborative work between robots, support wide-area communication and cross-regional collaboration, and flexibly adjust the communication range and structure according to changes in tasks to ensure that task execution is not restricted by communication; in order to achieve intercommunication between different platforms and devices, the protocol and media type interoperability mechanism provided by the 3GPP system can effectively solve the interoperability problem between robots. Through protocol conversion, media type conversion and data format standardization, the 3GPP system can ensure that robots from different manufacturers can achieve seamless communication and collaboratively complete tasks. In addition, the 3GPP system can also support the transmission of multimodal information and the collaborative work of AI models, further improving the efficiency of information exchange between robots and the intelligent level of task execution.

[0167] In this embodiment, the robot can automatically adjust its communication strategy based on its own state and surrounding environment. Specifically, the robot can perform the following four stages:

[0168] 1. Multimodal Environment Perception Stage

[0169] Autonomous condition monitoring (robot self-examination)

[0170] Battery level detection: Like a mobile phone battery level reminder, it will automatically switch to power saving mode (such as reducing video transmission) when the battery level is lower than 30%

[0171] Computational load prediction: Similar to computer CPU usage monitoring, it can predict whether there will be a lag in the next 5 seconds and decide whether to ask other robots to help.

[0172] Motion state recognition: Use the built-in gyroscope to determine whether you are stationary, moving at a constant speed, or making a sharp turn, and adjust the signal transmission strength

[0173] Environmental perception (robots perceive the surrounding network environment)

[0174] Signal strength measurement: Similar to the mobile phone signal bar display, it scans the signal quality of surrounding 5G base stations and WiFi hotspots every second

[0175] Interference Mapping: Generates a heat map of signal interference in three-dimensional space (e.g., metal equipment in a factory will appear as a red interference area)

[0176] Nearby device detection: Automatically discover other robots that can be directly connected within 200 meters (similar to mobile phone Bluetooth search)

[0177] 2. Dynamic Resource Modeling (Building a Communication Resource Map)

[0178] Network status analysis

[0179] Service quality score sheet: Use six indicators (delay, jitter, etc.) to score each communication line, similar to the road condition rating of navigation software

[0180] Dynamic topology modeling: Using AI to draw a real-time network structure diagram (e.g., robot movement causing signal weakening on certain paths)

[0181] Resource Forecast

[0182] Channel quality prediction: Using historical data to predict whether a channel will be congested in the future (similar to weather forecasting)

[0183] Spectrum efficiency evaluation: Analyze the actual transmission efficiency of different frequency bands (e.g., determine whether 5G high or low frequency bands are more appropriate)

[0184] 3. Intelligent Decision Engine (the robot’s communication brain)

[0185] Protocol Adaptation

[0186] Communication mode switching: switch to "Express Line" (URLLC mode) in emergencies, and use "Normal Broadband" (eMBB mode) in normal times

[0187] Business priority management: assign dedicated channels to critical instructions (such as emergency stop commands), similar to ambulance priority

[0188] Multi-network integration: connect to 5G and 4G channels simultaneously, like dual-SIM dual-standby on a mobile phone to improve stability

[0189] Transmission optimization

[0190] Video compression strategy: Automatically reduce image quality when network quality is poor (e.g., from 4K to 720P)

[0191] Multi-path transmission: 5G and WiFi are used to transmit data at the same time, just like using two pipes to deliver water at the same time.

[0192] Intelligent error correction: When the bit error rate is high, a special coding method is used to enhance anti-interference ability

[0193] 4. Collaborative execution layer (multi-robot collaboration)

[0194] Resource Scheduling

[0195] Channel allocation: Like a traffic controller, it dynamically allocates road resources and assigns appropriate communication frequency bands to each robot.

[0196] Edge computing collaboration: complex computing tasks are distributed to nearby computing nodes for processing (similar to asking neighbors to help pick up packages)

[0197] Knowledge sharing: robots regularly synchronize learning outcomes (e.g., sharing newly discovered signal blind spots)

[0198] Emergency Response

[0199] Quick connection recovery: When encountering obstacles, the backup antenna direction is switched within 2 milliseconds

[0200] Dual-link keep-alive: The primary and backup communication lines are on standby in real time to ensure 99.999% uninterrupted connection

[0201] Emergency support mode: Prioritizes the transmission of core commands in extreme situations (such as evacuation commands during fires)

[0202] For example, in the factory AGV cart scenario: when multiple logistics robots pass by metal shelves (signal interference area), they automatically switch to D2D direct connection to form a fleet and maintain communication through the relay function of the lead vehicle; in the urban security patrol scenario: when the patrol robot detects that the battery is low, it actively transfers the video monitoring task to a nearby robot and returns to recharge itself; in the disaster rescue scene scenario: in the base station damaged area scenario, the robots automatically form a Mesh network and transmit the on-site video back to the command center through multi-hop relay.

[0203] The 3GPP-based robot dynamic networking system in this embodiment has the following technical effects:

[0204] 1. Intelligent management is implemented. Dynamic resource perception and D2D communication modules enable the robot to automatically adjust its communication strategy based on its own status and surrounding environment. An AI-driven route optimizer uses a reinforcement learning model to predict link quality and make optimal path decisions in real time. A collaborative task scheduling platform utilizes network status data to optimize task allocation and improve overall work efficiency.

[0205] 2. This architecture solves the problems of protocol heterogeneity and media diversity in dynamic environments through the deep integration of 3GPP and the robotics field, and can serve key scenarios such as Industry 4.0 and smart cities.

[0206] 3. Expanding New Interfaces: The dynamic networking and protocol and media type interoperability mechanisms of intelligent robots based on 3GPP systems offer significant advantages in addressing issues such as WLAN susceptibility to interference and data leakage, and the short transmission distance and limited bandwidth of Bluetooth technology. These advantages help improve the communication stability, security, and data transmission efficiency of intelligent robot systems, providing strong support for the widespread application of intelligent robots.

[0207] 4. For dynamic networking, the invention uses the SMF to issue new rules to the UPF via the N4 interface, redirecting traffic to the D2D channel. It also uses the 3GPP interface protocol to implement D2D communication and maintain synchronization with the core network's cycle status.

[0208] 5. 3GPP systems (such as 5G NR-U) offer wider coverage than WiFi and Bluetooth. This means robots can maintain stable connections in large facilities or outdoor environments without frequent network switching. And by supporting seamless handover across base stations, robots can automatically and smoothly switch from one base station to another as they move within the network coverage area without interrupting service.

[0209] 6. Higher data rates and support for more users. 3GPP standards, especially 5G NR, offer extremely high data rates and ultra-low latency, making them ideal for applications requiring fast responses, such as real-time video streaming and remote control. Furthermore, compared to WiFi and Bluetooth, 3GPP systems can support more concurrent connections, which is particularly important in environments where large numbers of intelligent robots are deployed.

[0210] 7. Dynamic resource allocation and optimization allows the creation of multiple virtual networks on the same physical infrastructure, each optimized for specific types of traffic. This allows for flexible resource allocation based on the needs of different applications, improving overall efficiency. Furthermore, a reinforcement learning model is used to adjust routing paths based on real-time network status to adapt to changing network conditions, a capability not available in WiFi and Bluetooth.

[0211] 8. Enhanced interoperability and standardization: 3GPP-defined standards are global, ensuring interoperability between devices produced by different manufacturers. Detailed Quality of Service (QoS) management is provided, allowing priorities to be set based on different types of data flows (voice, video, data), ensuring the smooth execution of critical tasks.

[0212] 9. Cross-layer collaboration and management are more convenient. ProSe (Proximity Services) technology supports direct communication between devices, enhancing communication capabilities and reliability within local areas, and is particularly suitable for dense deployment scenarios.

[0213] It should be noted that, unless otherwise specified, when a feature is referred to as being "fixed" or "connected" to another feature, it may be directly fixed or connected to the other feature, or it may be indirectly fixed or connected to the other feature. In addition, the descriptions of up, down, left, right, etc. used in this disclosure are only relative to the relative positional relationships of the components of the present disclosure in the accompanying drawings. The singular forms of "a", "" and "the" used in this disclosure are also intended to include the plural forms, unless the context clearly indicates otherwise. In addition, unless otherwise defined, all technical and scientific terms used in this embodiment have the same meaning as those generally understood by those skilled in the art. The terms used in the specification of this embodiment are only for describing specific embodiments and are not intended to limit the invention. The term "and / or" used in this embodiment includes any combination of one or more related listed items.

[0214] It should be understood that, although the present disclosure may adopt the term first, second, third etc. to describe various elements, these elements should not be limited to these terms.These terms are only used to distinguish the elements of the same type from each other.For example, without departing from the scope of the present disclosure, the first element may also be referred to as the second element, and similarly, the second element may also be referred to as the first element.The use of any and all examples or exemplary language ("for example", "such as" etc.) provided by the present embodiment is only intended to better illustrate embodiments of the present invention, and unless otherwise required, the scope of the present invention will not be limited.

[0215] It should be appreciated that embodiments of the present invention can be implemented or practiced by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable memory. The methods can be implemented in a computer program using standard programming techniques - including a non-transitory computer-readable storage medium configured with a computer program, wherein the storage medium so configured causes the computer to operate in a specific and predefined manner - according to the methods and figures described in the specific embodiments. Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with the computer system. However, if desired, the program can be implemented in assembly or machine language. In any case, the language can be a compiled or interpreted language. In addition, the program can be run on a programmed application-specific integrated circuit for this purpose.

[0216] In addition, the operations of the processes described in this embodiment may be performed in any suitable order, unless otherwise indicated in this embodiment or otherwise clearly contradicted by the context. The processes described in this embodiment (or variations and / or combinations thereof) may be performed under the control of one or more computer systems configured with executable instructions, and may be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) that is executed collectively on one or more processors, by hardware, or a combination thereof. A computer program includes multiple instructions that can be executed by one or more processors.

[0217] Furthermore, the method can be implemented in any type of computing platform that is operably connected to a suitable computer, including but not limited to a personal computer, a minicomputer, a mainframe, a workstation, a network or distributed computing environment, a separate or integrated computer platform, or in communication with a charged particle tool or other imaging device, etc. Various aspects of the present invention can be implemented as machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, an optical read and / or write storage medium, RAM, ROM, etc., so that it can be read by a programmable computer, and when the storage medium or device is read by the computer, it can be used to configure and operate the computer to perform the process described herein. In addition, the machine-readable code, or portions thereof, can be transmitted over a wired or wireless network. When such media includes instructions or programs that implement the above steps in conjunction with a microprocessor or other data processor, the invention of this embodiment includes these and other different types of non-transitory computer-readable storage media. When programmed according to the methods and techniques of the present invention, the present invention also includes the computer itself.

[0218] The computer program can be applied to input data to perform the functions of the present embodiment, thereby converting the input data to generate output data that is stored in a non-volatile memory. The output information can also be applied to one or more output devices such as a display. In a preferred embodiment of the present invention, the converted data represents a physical and tangible object, including a specific visual depiction of the physical and tangible object produced on the display.

[0219] The above are merely preferred embodiments of the present invention. The present invention is not limited to the aforementioned embodiments. As long as the technical effects of the present invention are achieved by the same means, any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention. Within the scope of protection of the present invention, various modifications and variations of the technical solutions and / or implementation methods may be made.

Claims

1. A 3GPP-based robot dynamic networking system, characterized in that: The 3GPP-based robot dynamic networking system includes: Terminal layer; the terminal layer includes a plurality of terminal units corresponding to robots; Access and transport layer; the access and transport layer is used to control the dynamic connection topology between each terminal unit and the core network layer; Core network layer; the core network layer is used to provide 3GPP protocol access and dynamic resource allocation to each terminal unit.

2. The 3GPP-based robot dynamic networking system according to claim 1, characterized in that: The terminal units include: Multimodal collaboration module; the multimodal collaboration module includes heterogeneous sensors, dynamic resource sensing units, inter-device direct communication units and action execution units; the heterogeneous sensors are used to collect sensor data, the dynamic resource sensing unit is used to sense its own communication resource data, and the inter-device direct communication unit is used to perform inter-device direct communication. Protocol and media adapter; the protocol and media adapter is used to encapsulate the sensor data and the communication resource data into a first ROS message and send it to the outside, and receive a second ROS message from the outside to parse and obtain an action instruction; The action execution unit is used to execute the action instruction.

3. The 3GPP-based robot dynamic networking system according to claim 2, characterized in that: The access and transport layer includes: Heterogeneous wireless access module; the heterogeneous wireless access module is used to establish a heterogeneous wireless connection with the terminal unit and to send and receive the first ROS message and the second ROS message; Dynamic topology management module; the dynamic topology management module is used to perform QoE monitoring according to the first ROS message, perform reinforcement learning to predict link quality according to the QoE monitoring result, and switch the connection path of the terminal unit according to the link quality prediction result.

4. The 3GPP-based robot dynamic networking system according to claim 3, characterized in that: The switching of the connection path of the terminal unit according to the link quality prediction result includes: Determining a next hop of a first terminal unit according to a link quality prediction result; the first terminal unit is any terminal unit; When the next hop of the first terminal unit is the heterogeneous wireless access module, triggering the first terminal unit to connect to the heterogeneous wireless access module; When the next hop of the first terminal unit is a second terminal unit, a direct device connection between the first terminal unit and the second terminal unit is triggered; the second terminal unit is a terminal unit other than the first terminal unit.

5. The 3GPP-based robot dynamic networking system according to claim 2, characterized in that: The core network layer includes: A network function virtualization module; the network function virtualization module is used to run a network slice orchestrator to dynamically allocate slice resources to each terminal unit; Protocol / media interoperability gateway; the protocol / media interoperability gateway is used to run the protocol state machine model and perform communication protocol conversion corresponding to the first ROS message and / or the second ROS message.

6. The 3GPP-based robot dynamic networking system according to claim 1, characterized in that: The 3GPP-based robot dynamic networking system also includes: Application and service layer: The application and service layer is used to provide media application services to the terminal unit.

7. The 3GPP-based robot dynamic networking system according to claim 6, characterized in that: The application and service layer includes: A task scheduling platform; the task scheduling platform is used to open network capabilities to the first ROS message and cooperate with the terminal unit to perform the robot task according to the first ROS message; Security and privacy module; the security and privacy module is used to run the 3GPP enhanced security mechanism to perform data traceability and integrity verification on the first ROS message.

8. The 3GPP-based robot dynamic networking system according to claim 6, characterized in that: The application and service layer is connected to the core network layer via a 3GPP API, and the core network layer is connected to the access and transport layer via an NGAP / Xn / N2 interface.

9. The 3GPP-based robot dynamic networking system according to any one of claims 1 to 8, characterized in that: The terminal layer includes a plurality of robots, and any of the robots includes one or more terminal units; Each of the terminal units is combined into one or more terminal unit groups, and any of the terminal unit groups includes one or more of the terminal units; For any of the terminal unit groups, one of the terminal units in the terminal unit group communicates with the outside world of the terminal unit group, while the other terminal units in the terminal unit group keep their communications with the outside world hidden, and direct device-to-device communication is carried out between the terminal units in the terminal unit group.

10. The 3GPP-based robot dynamic networking system according to claim 9, characterized in that: Each of the terminal units is combined into one or more terminal unit groups, including: When working in the first networking mode, all the terminal units belonging to the same robot are combined into the same terminal unit group, and the terminal units belonging to different robots are combined into different terminal unit groups; When operating in the second networking mode, determining a combination range, combining all the terminal units belonging to the combination range into a first terminal unit group, and for any terminal unit not belonging to the combination range, combining the terminal unit alone or with terminal units belonging to the same robot into a corresponding second terminal unit group; When working in the third networking mode, each of the terminal units is determined to be a corresponding terminal unit group.

Images