Hardware plug-in system for providing large model capability for accompanying robot

Abstract

The invention discloses a hardware plug-in system for providing large model capability for an accompanying robot, and the system comprises a plug-in interface module which is connected with a main control system of the accompanying robot; the main processing chip adopts a chip with AI reasoning and training capabilities; the built-in memory is used for storing a pre-trained language-behavior-perception large model; the data acquisition and feedback module is used for interacting with a robot system, receiving sensor data and outputting decision feedback; and the PLC calling interface module is used for carrying out logic signal interaction and model calling with a robot control system. By integrating the large model and the data storage function into the pluggable hardware card, rapid deployment and personalized memory migration of the companion robot large model capability are realized, the computing capability and the response speed are improved, and the emotion continuity and the data privacy of the user are also enhanced.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and robotics technology, and in particular to a hardware plug-in system for providing large model capabilities for a companion robot. Background Art

[0002] Most existing companion robots rely on local or cloud computing to achieve interactive capabilities and intelligent responses. However, cloud computing presents privacy and security issues and network latency issues, while local computing is limited by computing resources and struggles to run large-scale AI models. Furthermore, if a robot is damaged or replaced, the emotional memory between the user and the robot cannot be transferred, resulting in a disruption in the companionship experience and reduced user satisfaction. Summary of the Invention

[0003] In view of the above problems, the present invention is proposed to provide a hardware plug-in system that overcomes the above problems or at least partially solves the above problems and provides large model capabilities for a companion robot.

[0004] According to one aspect of the present invention, a hardware plug-in system for providing large model capabilities for a companion robot is provided, the hardware plug-in system comprising:

[0005] Plug-in interface module, used to connect to the companion robot main control system;

[0006] The main processing chip uses a chip with AI reasoning and training capabilities;

[0007] Built-in memory for storing pre-trained language-behavior-perception models;

[0008] Data acquisition and feedback module, used to interact with the robot system, receive sensor data, and output decision feedback;

[0009] The PLC calling interface module is used for logical signal interaction and model calling with the robot control system.

[0010] Optionally, the chip with AI reasoning and training capabilities specifically includes: ARM+NPU architecture, RISC-VAI chip, or FPGA+ASIC combination.

[0011] Optionally, the companion robot reserves a plug-in interface during design, and realizes automatic recognition and model calling of hardware plug-ins through software drivers within the system.

[0012] Optionally, the automatic identification and model calling of the hardware plug-in through the software driver in the system specifically includes:

[0013] Model loading module, used to identify the model structure in the hardware card and load it into the robot interaction logic;

[0014] Data synchronization module, used to transmit the robot's data information to the hardware card for processing in real time;

[0015] Feedback module, used to generate feedback behavior based on model output;

[0016] Optionally, the data information specifically includes: sensor, voice and image information.

[0017] Optionally, the feedback behavior specifically includes: voice, action and expression.

[0018] Optionally, the hardware plug-in system also includes: when the robot hardware is damaged or upgraded, the user directly inserts the hardware card into another compatible robot, and the system automatically loads the model and user data in the card to maintain a consistent companionship experience.

[0019] The present invention provides a hardware plug-in system that provides large-scale model capabilities for a companion robot. The system includes a plug-in interface module for connecting to the companion robot's main control system; a main processing chip equipped with AI reasoning and training capabilities; built-in memory for storing pre-trained large-scale language-behavior-perception models; a data acquisition and feedback module for interacting with the robot system, receiving sensor data, and outputting decision feedback; and a PLC call interface module for interacting with the robot control system through logical signal exchange and model call. By integrating large-scale model and data storage functions into a pluggable hardware card, the system enables rapid deployment of large-scale model capabilities and personalized memory migration for the companion robot, improving computing power and response speed while also enhancing user emotional continuity and data privacy.

[0020] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 A block diagram of a hardware plug-in system for providing large model capabilities for a companion robot according to an embodiment of the present invention;

[0023] Figure 2A flowchart of system interaction after a hardware card is inserted into the companion robot provided by an embodiment of the present invention;

[0024] Figure 3 A schematic diagram of a user memory migration scenario provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0026] The terms "comprises" and "comprising" and any variations thereof in the description, embodiments, claims and drawings of the present invention are intended to cover non-exclusive inclusions, for example, including a series of steps or units.

[0027] The technical solution of the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0028] The purpose of this invention is to provide a hardware plug-in device that provides a companion robot with built-in large-scale model computing capabilities and data memory migration capabilities. This hardware card can be inserted into the robot through a standardized interface. While enabling model invocation, it continuously records external data during user interaction, enabling the retention and migration of the user's emotional memory.

[0029] The present invention provides a hardware plug-in system for providing large model capabilities for a companion robot, comprising:

[0030] Plug-in interface module, used to connect to the companion robot main control system;

[0031] The main processing chip uses a dedicated chip with AI reasoning and training capabilities (such as ARM+NPU architecture, RISC-VAI chip, or FPGA+ASIC combination);

[0032] Built-in memory for storing pre-trained language-behavior-perception models;

[0033] Data acquisition and feedback module, used to interact with the robot system, receive sensor data, and output decision feedback;

[0034] PLC call interface module, used for logic signal interaction and model call with the robot control system;

[0035] The hardware plug-in system has:

[0036] Hot-swappable capability, allowing it to be safely plugged in and out while the robot is running;

[0037] Data incremental learning capability can fine-tune model parameters based on user usage and retain interactive memory;

[0038] Secure encryption mechanism to prevent user data from being leaked or illegally called.

[0039] The robot is designed with a plug-in interface, and the system's software driver enables automatic recognition and model calling of hardware plug-ins, including:

[0040] Model loading module: identifies the model structure in the hardware card and loads it into the robot interaction logic;

[0041] Data synchronization module: transmits the robot's sensor, voice, image and other information to the hardware card for processing in real time;

[0042] Feedback module: Generates feedback behaviors such as voice, action, and expression based on model output.

[0043] When the robot hardware is damaged or upgraded, the user can directly insert the hardware card into another compatible robot. The system will automatically load the model and user data on the card to maintain a consistent companionship experience.

[0044] like Figures 1 to 3 As shown, the hardware plug-in described in this invention includes a processing module based on an edge AI chip, a built-in large language model, and a multimodal data analysis module. Upon detecting the insertion of the hardware card, the robot's main control system loads the plug-in's model services via a PLC interface and transmits sensory data, such as voice, image, and tactile, to the plug-in for processing. The plug-in generates feedback signals, which the robot then responds with, such as voice responses and facial expressions.

[0045] At the same time, the user's interaction data and behavior patterns are continuously recorded in the plug-in storage area, forming a personalized behavior model and emotional profile. After replacing the robot body, simply insert the plug-in into the new robot to load the previous interaction model, realizing the "memory" transfer.

[0046] Beneficial effects: The present invention integrates large models and data storage functions into a pluggable hardware card, thereby realizing the rapid deployment of large model capabilities and personalized memory migration of companion robots, which not only improves computing power and response speed, but also enhances the user's emotional continuity and data privacy.

[0047] The above specific implementation methods further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A hardware plug-in system that provides large model capabilities for a companion robot, characterized in that: The hardware plug-in system includes: Plug-in interface module, used to connect to the companion robot main control system; The main processing chip uses a chip with AI reasoning and training capabilities; Built-in memory for storing pre-trained language-behavior-perception models; Data acquisition and feedback module, used to interact with the robot system, receive sensor data, and output decision feedback; The PLC calling interface module is used for logical signal interaction and model calling with the robot control system.

2. A hardware plug-in system for providing large model capabilities for a companion robot according to claim 1, characterized in that: The chips with AI reasoning and training capabilities specifically include: ARM+NPU architecture, RISC-VAI chip, or FPGA+ASIC combination.

3. A hardware plug-in system for providing large model capabilities for a companion robot according to claim 1, characterized in that: The companion robot reserves a plug-in interface during design, and realizes automatic recognition and model calling of hardware plug-ins through software driver within the system.

4. A hardware plug-in system for providing large model capabilities for a companion robot according to claim 3, characterized in that: The automatic identification and model calling of the hardware plug-in by the software driver in the system specifically includes: Model loading module, used to identify the model structure in the hardware card and load it into the robot interaction logic; Data synchronization module, used to transmit the robot's data information to the hardware card for processing in real time; The feedback module is used to generate feedback behavior based on the model output.

5. A hardware plug-in system for providing large model capabilities for a companion robot according to claim 4, characterized in that: The data information specifically includes: sensor, voice and image information.

6. The hardware plug-in system for providing large model capabilities for a companion robot according to claim 4, characterized in that: The feedback behaviors specifically include: voice, action and expression.

7. The hardware plug-in system for providing large model capabilities for a companion robot according to claim 1, characterized in that: The hardware plug-in system also includes: When the robot hardware is damaged or upgraded, the user simply inserts the hardware card into another compatible robot, and the system automatically loads the model and user data on the card to maintain a consistent companionship experience.

Images