Multi-part collaborative voice AI programming expression robot control system and method

Abstract

The invention discloses a multi-part collaborative voice AI programming expression robot control system and method. The system comprises a main control unit, a voice processing module, an AI acceleration unit, a safety storage area, a debugging and program injection interface, an actuator driving circuit and multi-stage safety hardware. According to the system, a structured control intention vector is extracted through voice recognition, and a control code is automatically generated or modified by an AI programming agent in combination with an emotional action mapping library and physical constraints; after digital twinborn simulation verification and automatic debugging optimization, the data are safely burnt to hardware and executed, automatic generation and reliable execution of multi-part collaborative expressions and actions are achieved, programming complexity is remarkably reduced, and interaction naturalness and system safety are improved.

Description

Technical Field

[0001] This invention relates to robot control technology, and more particularly to a multi-part collaborative voice AI programming facial expression robot control system and method. Background Technology

[0002] With the deep integration of artificial intelligence and robotics, biomimetic robots capable of emotional expression have become a cutting-edge research direction in the field of human-computer interaction. However, existing facial expression robot systems still face many challenges in achieving natural, vivid, and multi-part coordinated emotional interaction, such as excessively high programming complexity, long debugging cycles, slow system response, and a lack of intuitive interaction methods, which seriously restrict their popularization and application depth.

[0003] At the level of emotion understanding and decision-making, existing technologies have attempted to establish a correlation between user emotions and robot behavior. Invention patent CN120015062A, a method for expressing the emotional behavior of a humanoid robot, proposes a method based on a random forest model to identify user voice emotions and make decisions about the robot's emotional behavior accordingly. This method effectively improves the emotion recognition rate in complex environments through a separate microphone array, sound source localization, and multi-feature fusion, and constructs a mapping relationship between emotion categories and robot behavior. However, the behavior output described in this method heavily relies on predefined action and voice template libraries. Its decision-making result is actually the selection and invocation of preset templates, rather than the real-time generation of new, refined action control instructions based on emotion intensity parameters. This means that the system lacks true creativity and parameterized behavior generation capabilities, making it difficult to achieve continuous and precise control of action amplitude and speed, such as "increase the smile amplitude by 50%" or "blink slowly." Its expressive richness and adaptive capabilities have inherent limitations.

[0004] At the actuator level, a highly flexible bionic head structure design is crucial for achieving precise facial expressions. The invention patent CN120552120A provides an advanced hardware solution for a multi-expression-driven bionic intelligent robot head. This solution integrates 26 servo motors through a layered, modular mechanical structure and employs a transmission method that coordinates traction lines and a track mechanism to precisely control the movement of multiple parts such as the eyeballs, eyelids, eyebrows, cheekbones, lips, and chin, achieving highly realistic multi-expression driving capabilities. This design effectively solves the problems of stiff facial expressions and insufficient degrees of freedom in traditional robot heads. However, the core contribution of this patent lies in its precise mechanical structure and drive scheme, but it does not address how to intelligently control the coordinated movement of these 26 servo motors to express specific emotions. Its control method still relies on pre-programmed sequences and lacks a higher-level intelligent control system capable of receiving high-level semantic instructions, such as "awkward smile," and automatically converting them into specific motion parameters, trajectories, and timing of the 26 servo motors.

[0005] Neither of these methods addresses the core issue of "how to enable users to dynamically, accurately, automatically, and safely execute control code that drives such complex mechanisms through the most natural voice interaction method." Existing control methods still require professionals to manually write the underlying code, and the debugging process is cumbersome and error-prone, failing to meet the interactive needs of real-time, online, and personalized programming of robot expressions.

[0006] Therefore, there is an urgent need for a full-process control system that integrates advanced intent understanding, AI parametric programming, virtual simulation verification, and automated security deployment to bridge the gap between advanced emotional commands and fine control of multiple underlying actuators, ultimately lowering the programming threshold, improving development efficiency and system reliability, and enabling vivid and natural multi-part collaborative emotional expression. Summary of the Invention

[0007] The first objective of this invention is to provide a multi-part collaborative voice AI programming facial expression robot control system, which aims to solve the problems of existing facial expression robots being complex to program, difficult to debug, slow to respond, and lacking natural interaction methods, thus failing to achieve vivid, safe, and multi-part collaborative emotional expression.

[0008] To address the aforementioned technical issues, a multi-part collaborative voice AI programming facial expression robot control system is provided, comprising a main control unit, a voice processing module, an AI acceleration unit, a secure storage area, actuator drive circuits, and a debugging interface. The main control unit is a multi-core heterogeneous processor, integrating an application processing core for running AI algorithms and the operating system, and a real-time control core for real-time parallel control of multi-channel actuators. The voice processing module integrates hardware audio preprocessing circuitry for hardware-level noise reduction, echo cancellation, and sound source localization of microphone signals, outputting a clean voice signal. The AI ​​acceleration unit accelerates the operation of the voice recognition model, key information extraction model, intent reasoning module, and digital twin. The simulation module performs physical calculations; the secure storage area stores verified system programs, user configuration files, emotion and action mapping libraries, and system recovery images; the actuator drive circuit is a multi-channel drive circuit that communicates with the main control unit via a high real-time bus to drive multiple actuators on the robot; the output of the voice processing module is connected to the input of the AI ​​acceleration unit; the output of the AI ​​acceleration unit is connected to the application processing core of the main control unit; the output of the real-time control core of the main control unit is connected to the input of the actuator drive circuit; the secure storage area is bidirectionally connected to the main control unit; the main control unit is connected to external systems via a debugging and program injection interface.

[0009] In one embodiment, the clean speech signal output by the speech processing module is converted into text by the speech recognition model running by the AI ​​acceleration unit, and the key information extraction model extracts a structured control intent vector containing action type, location, intensity, speed, and frequency from the text; the intent reasoning module running by the AI ​​acceleration unit is used to receive and process the structured control intent vector and output control command parameters for simulation verification by the digital twin simulation module.

[0010] In one embodiment, the emotion-action mapping library stored in the secure storage area is used to map the structured control intent vector to specific actuator parameters; the emotion-action mapping library is constructed through a combination of expert rules and data-driven methods, and includes emotion instructions, facial expression parameters, associated body movements, and intensity codes.

[0011] In one embodiment, the main control unit runs an AI programming agent that employs a retrieval-enhanced generation architecture. The AI ​​programming agent receives the structured control intent vector and the robot's current state information, and automatically generates or modifies executable control code by querying the emotion-action mapping library and the physical constraint library. The AI ​​acceleration unit runs a digital twin simulation module for simulating and verifying the generated control code.

[0012] In one embodiment, the digital twin simulation module is connected to an AI debugging agent. If the simulation reports an error, the AI ​​debugging agent starts a multi-objective optimization algorithm to fine-tune and optimize the parameters or logic of the control code, and starts iterative simulation until the verification is successful.

[0013] In one embodiment, the system employs dual firmware partitions (A / B) and a timer. After verification, the new program is burned to partition B for operation, while partition A backs up the old version of the system program. If an error occurs, the timer triggers the system to automatically roll back to the stable version in partition A.

[0014] In one embodiment, the multi-site actuator includes a face actuator, a neck actuator, a torso actuator, and an arm actuator.

[0015] The second objective of this invention is to provide a multi-part collaborative voice AI programming facial expression robot control method, which aims to solve the problem that existing robot facial expression control methods rely on pre-programming and cannot generate and safely execute multi-part collaborative actions in real time through natural voice commands.

[0016] To address the aforementioned technical problems, a method for controlling a multi-part collaborative voice AI programming facial expression robot is provided, applied to the aforementioned multi-part collaborative voice AI programming facial expression robot control system, comprising the following steps:

[0017] S1. Voice command capture and semantic understanding: The voice signal is preprocessed into a clean voice signal by the voice processing module. The clean voice signal is then processed by the voice recognition model and the key information extraction model to output a structured control intent vector.

[0018] S2. AI-driven parametric programming and code generation: The AI ​​programming agent queries the emotion-action mapping library and physical constraint library based on the structured control intent vector, and automatically generates or modifies executable control code.

[0019] S3. Virtual simulation and automated debugging: The generated control code is simulated and verified by the digital twin simulation module, and the control code is debugged and optimized by the AI ​​debugging agent, iterating the control code until it passes;

[0020] S4. Secure burning and execution: Securely burn the verified control code into the main control unit, and monitor the actuator status in real time during execution to form a closed-loop feedback.

[0021] In one embodiment, the AI-driven parametric programming and code generation further includes the AI ​​programming agent using retrieval-enhanced generation technology to retrieve pre-stored code templates based on intent reasoning results, and modifying and combining them with parameters. The generation process incorporates security rules and evaluates the emotional fidelity of the generated results.

[0022] In one embodiment, the secure programming and execution further includes performing a checksum comparison after programming to ensure integrity, and feeding back the actuator status data collected in real time during execution to the AI ​​programming agent for evaluating the execution effect and optimizing subsequent instructions.

[0023] Implementing the embodiments of the present invention will have the following beneficial effects:

[0024] 1. In the multi-part collaborative voice AI programming facial expression robot control system of this embodiment, the user issues commands through natural speech. The speech processing module performs hardware-level preprocessing on the microphone signal and outputs clean speech to the AI ​​acceleration unit. The AI ​​acceleration unit converts the speech into text through a speech recognition model, and the key information extraction model extracts a structured control intent vector containing action parameters from the text. The intent reasoning module processes the vector and outputs control parameters. The AI ​​programming agent queries the emotion-action mapping library and the physical constraint library based on these parameters and automatically generates or modifies the control code. The digital twin simulation module performs pre-simulation verification of the code, and the AI ​​debugging agent automatically debugs and optimizes it in the simulation environment, forming a closed-loop iteration. After the verified security code is burned through the A / B partitioning mechanism, it is controlled in real time by the main control unit. The core drives multi-channel actuator circuits via a high real-time bus to coordinate the control of facial, neck, torso, and arm actuators to complete emotional expression. Compared with existing technologies, this solution combines multi-core heterogeneous computing, hardware AI acceleration, retrieval-enhanced generation (RAG) technology, digital twin simulation, and multiple security mechanisms to construct a fully automated control chain of "voice → intent understanding → AI programming → simulation verification → automatic debugging → secure burning → execution feedback." This enables the automatic generation and execution of vivid and reliable multi-part coordinated facial expressions based on natural voice interaction. By using an AI agent to understand the subtle differences in user commands and parameterize them to specific actuator actions, the barrier to entry and debugging cycle for robot facial expression programming are significantly reduced, fundamentally improving development efficiency, system reliability, and the naturalness of human-computer interaction.

[0025] 2. The multi-part collaborative voice AI programming facial expression robot control method in this embodiment captures and preprocesses voice commands into clean signals through the voice processing module. After being recognized and extracted by the AI ​​acceleration unit, a structured control intent vector is generated. The AI ​​programming agent automatically generates or modifies the control code based on the vector. The digital twin module performs simulation verification on the code, and the AI ​​debugging agent automatically debugs and optimizes until it passes. Finally, the safety code is burned into the main control unit for execution, forming a closed-loop feedback, realizing fully automatic and intelligent control from natural voice input to safe output of multi-part collaborative actions. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1This is a block diagram of the overall hardware architecture of the multi-part collaborative voice AI programming facial expression robot control system described in Embodiment 1 of the present invention;

[0028] Figure 2 This is a flowchart of the multi-part collaborative voice AI programming facial expression robot control method described in Embodiment 2 of the present invention;

[0029] Figure 3 This is a schematic diagram of the multi-part collaborative control of the multi-part collaborative voice AI programming facial expression robot according to Embodiment 1 of the present invention;

[0030] Figure 4 This is a conceptual diagram of the multi-part collaborative voice AI programming facial expression robot emotion and action mapping library described in Embodiment 1 of the present invention;

[0031] Figure 5 This is a schematic diagram of the AI ​​programming agent workflow of the multi-part collaborative voice AI programming facial expression robot control method described in Embodiment 2 of the present invention;

[0032] Figure 6 This is a schematic diagram of the digital twin simulation and debugging interface of the multi-part collaborative voice AI programming facial expression robot control method described in Embodiment 2 of the present invention;

[0033] Figure 7 This is a flowchart of the secure burning and rollback mechanism of the multi-part collaborative voice AI programming facial expression robot control system described in Embodiment 1 of the present invention. Detailed Implementation

[0034] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0035] It should be noted that when a component is said to be "fixed to" another component, it can be directly attached to the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0037] Example 1

[0038] refer to Figure 1 , Figure 3 , Figure 4 and Figure 7 Embodiment 1 of the present invention provides a multi-part collaborative voice AI programming facial expression robot control system.

[0039] refer to Figure 1 An embodiment of a multi-part collaborative voice AI programming facial expression robot control system includes a main control unit, a voice processing module, an AI acceleration unit, a secure storage area, an actuator drive circuit, and a debugging and program injection interface.

[0040] The main control unit is a multi-core heterogeneous processor, integrating an application processing core (such as the ARM Cortex A series) and a real-time control core (such as the ARM Cortex M series). The application processing core is used to run advanced AI algorithms and operating systems (such as Linux); the real-time control core is used to generate high-precision, multi-channel, real-time parallel control instructions for actuators, ensuring strict synchronization of the movement of each actuator.

[0041] The voice processing module integrates hardware audio preprocessing circuits (such as dedicated DSP chips and matching ADCs) to perform hardware-level noise reduction (NR), echo cancellation (AEC), and sound source localization (DOA) on microphone array signals, outputting a clean voice signal with a high signal-to-noise ratio.

[0042] AI acceleration units (such as integrated NPUs or FPGAs) are used to accelerate the physical computations in speech recognition models, key information extraction models, intent reasoning modules, and digital twin simulation modules.

[0043] The secure storage area (a protected area partitioned using eMMC or SPIFlash) is used to store verified system programs, user configuration files, emotion and action mapping libraries, and system recovery images.

[0044] The actuator drive circuit is a multi-channel, high-precision drive circuit (such as a servo motor drive board), which communicates with the main control unit through a high real-time bus (such as EtherCAT or CAN / FD) to drive the actuators of multiple parts of the robot (including face, neck, torso, and arm actuators).

[0045] The output of the voice processing module is connected to the input of the AI ​​acceleration unit; the output of the AI ​​acceleration unit is connected to the application processing core of the main control unit; the output of the real-time control core of the main control unit is connected to the input of the actuator drive circuit; the secure storage area is bidirectionally connected to the main control unit; the main control unit is connected to external debugging equipment through debugging and program injection interfaces (such as SWD / JTAG).

[0046] Implementing Embodiment 1 of the present invention will have the following beneficial effects: When the multi-part collaborative voice AI programming facial expression robot control system of this solution is working, the user issues commands through natural speech (such as "laugh exaggeratedly and wave"), and the speech processing module performs hardware-level preprocessing on the microphone signal and outputs a clean speech signal to the AI ​​acceleration unit; the AI ​​acceleration unit first converts the clean speech signal into text through its lightweight speech recognition model (such as based on the CNNCTC architecture); then, its key information extraction model (such as based on the lightweight BERT architecture) extracts a structured control intent vector from the text, which contains multi-dimensional parameters such as action type (laugh, wave), part (face, arm), intensity (exaggerated), speed (default), and number of times (1 time); the vector is then passed to the intent reasoning module for processing.

[0047] Please refer to Figure 4 After receiving the vector, the intent reasoning module queries the emotion-action mapping library in the secure storage area. This mapping library is built by combining expert rules with motion capture data learning. It not only maps high-level emotional commands (such as "laugh exaggeratedly") to specific actuator parameters (such as "increase the angle of the corner of the mouth servo to 25°, and contract the orbicularis oculi muscle servo by 40%), but also supports the parsing and mapping of fine parameters such as stroke, rate, execution duration, turning direction, and return flag contained in the structured control intent vector. For example, users can achieve precise and parameterized control of the actuators through specific voice commands such as "raise the left eyelid upward by 5 mm, at a rate of 2 mm per second, hold for 3 seconds and then lower it" or "contract the nose downward by 2 mm for 1 second." "Waving" is mapped to "raise the shoulder servo by 30°, and reciprocate the elbow servo twice."

[0048] Please refer to Figure 5The AI ​​programming agent running on the main control unit (using a retrieval-enhanced generation RAG architecture) is then activated. It receives structured parameters from the intent reasoning module and sensor data from the robot's current state. By querying the emotion-action mapping library and physical constraint library (such as joint angle limits and maximum speed), it retrieves the action primitive code templates for "smiling" and "waving" from the local knowledge base. Based on the intensity parameter of "exaggeration," it modifies the displacement and velocity values ​​in the templates, automatically generating a new, executable multi-part collaborative control code. The multi-part actuators include facial (e.g., eyelids, nose, mouth), neck, torso, and arm actuators. The neck actuator can perform rotation (e.g., turning the head left and right) and pitch (e.g., nodding, raising the head) movements to express rich emotional scenarios. For example, when receiving a negative voice command, it controls the head to make avoidance actions such as lowering the head or turning away. The eye actuator supports independent or collaborative eye movements in eight directions (left, right, up, down, upper left, lower left, upper right, lower right) to simulate natural eye-tracking interaction.

[0049] Please refer to Figure 6 Before being burned into the hardware, the generated code is simulated and verified by a digital twin simulation module (based on a simplified version of an embedded physics engine such as Bullet) run by the AI ​​acceleration unit. The simulation monitors in real time whether collisions occur between virtual actuators, whether the motion trajectory is feasible, and whether any physical constraints are violated. If the simulation reports an error (such as detecting interference between the arm swing trajectory and the head), the connected AI debugging agent will start a multi-objective optimization algorithm (such as Bayesian optimization), automatically fine-tune the code parameters (such as reducing the arm swing amplitude or modifying the torso tilt angle), and start iterative simulation until verification is successful.

[0050] Please refer to Figure 7 The final, verified code is securely burned into a specific memory area (Area B) of the main control unit via the debugging interface. The system employs dual firmware partitioning (A / B) and a hardware watchdog timer. After burning, the checksum (e.g., CRC32) of the hardware-calculated code is compared with the value generated by the simulation end to ensure integrity. The new program runs in Area B, while the old version is backed up in Area A. If the hardware watchdog timer detects an abnormality in the new program's operation (e.g., actuator overcurrent or program deadlock), it immediately triggers a hardware-level rollback to the stable version in Area A, ensuring system security.

[0051] Finally, the real-time control core of the main control unit executes the program, sending precise parallel control commands to the actuator drive circuit via a high real-time bus to drive various actuators (such as those for the face, neck, torso, and arms) Figure 3 As shown, the actuators complete coordinated movements, thereby achieving vivid and natural multi-expression emotions. Simultaneously, real-time status data (position, current, temperature) of the actuators is continuously collected and fed back to the AI ​​programming agent, forming a closed loop for evaluating execution effectiveness and optimizing subsequent instructions.

[0052] Compared to existing technologies, this solution combines multi-core heterogeneous computing, hardware AI acceleration, retrieval-enhanced generation (RAG) technology, digital twin simulation, and multiple security mechanisms to construct a fully automated control chain of "voice → intent understanding → AI programming → simulation verification → automatic debugging → secure burning → execution feedback." This enables the automatic generation and execution of vivid and reliable multi-part collaborative facial expressions based on natural voice interaction. By using an AI agent to understand the subtle differences in user commands and parameterize them to specific actuator actions, it significantly reduces the threshold and debugging cycle for robot facial expression programming, fundamentally improving development efficiency, system reliability, and the naturalness of human-computer interaction.

[0053] It is important to note that the emotion-motion mapping library is not static; it can be updated and optimized through data-driven learning (such as recording motion capture data from real-life performances), thereby making the robot's facial expressions richer and more human-like.

[0054] Furthermore, it is worth noting that the simulation accuracy of the digital twin simulation module can be configured as needed, from rapid coarse simulation to high-precision detailed simulation, in order to achieve a balance between debugging speed and accuracy.

[0055] See Figure 1 In one optional embodiment, the hardware audio preprocessing circuit integrated into the speech processing module includes a dedicated DSP chip (such as the domestically produced CI1122), which has a built-in hardware-accelerated neural network processor for running lightweight speech recognition models offline, further reducing system latency and protecting user privacy.

[0056] Based on actual needs, the NPU (Neural Processing Unit) used in the AI ​​acceleration unit supports INT8 quantization calculation, which can significantly reduce computing power consumption while ensuring recognition accuracy, making it suitable for embedded mobile platforms.

[0057] In another alternative embodiment, the high real-time bus is preferably the EtherCAT bus, which has microsecond-level synchronization accuracy. This ensures that the movements of dozens of actuators distributed in the robot's head, neck, torso, and arms are strictly synchronized, avoiding disjointed movements and stuttering. This is the key to achieving natural coordinated movement.

[0058] Example 2

[0059] The multi-part collaborative voice AI programming facial expression robot control method in this second embodiment differs from the control system in the first embodiment in terms of the protected subject matter, specifically as follows:

[0060] Please refer to Figure 2 and Figure 5A multi-part collaborative voice AI programming facial expression robot control method, as described in another embodiment, is applied to the system described in Embodiment 1 and includes the following steps:

[0061] S1. Voice command capture and semantic understanding: The voice processing module captures the voice signal and performs hardware-level preprocessing to obtain a clean voice signal; the clean voice signal is converted into text by the voice recognition model running by the AI ​​acceleration unit, and then processed by the key information extraction model to output a structured control intent vector;

[0062] S2. AI-driven parametric programming and code generation: The AI ​​programming agent queries the emotion action mapping library and physical constraint library based on the control intention vector, uses retrieval enhancement generation technology to automatically retrieve relevant code templates, and modifies and combines them with parameters such as intensity and speed to generate new executable control code. The generation process embeds safety rules and performs emotion fidelity assessment.

[0063] S3. Virtual Simulation and Automated Debugging: The generated control code is simulated and verified through the digital twin simulation module to monitor collisions, trajectory feasibility, and boundary conditions. If the simulation reports an error, the AI ​​debugging agent will start the optimization algorithm to fine-tune the code parameters or logic and iterate the simulation until the verification is successful.

[0064] S4. Secure Programming and Execution: Securely program the verified control code to the main control unit. After programming, perform a checksum comparison to ensure integrity. The main control unit executes the program, driving multiple actuators to complete coordinated actions. During execution, actuator status data is collected in real time and fed back to the AI ​​programming agent for evaluating execution performance and optimizing subsequent instructions.

[0065] Implementing Embodiment 2 of the present invention will have the following beneficial effects: the voice processing module captures and preprocesses the voice commands into a clean signal, which is then recognized and extracted by the AI ​​acceleration unit to generate a structured control intent vector; the AI ​​programming agent automatically generates or modifies the control code based on the vector; the digital twin module performs simulation verification on the code, and the AI ​​debugging agent automatically debugs and optimizes it until it passes; finally, the security code is burned into the main control unit for execution, forming a closed-loop feedback, realizing fully automatic and intelligent control from natural voice input to safe output of multi-part coordinated actions.

[0066] See Figure 2In one alternative embodiment, in AI-driven parametric programming and code generation, the AI ​​programming agent employs Retrieval Enhanced Generation (RAG) technology. Its local knowledge base stores a large number of pre-validated skill primitive code templates (such as C++ functions and Python script blocks). The agent first retrieves the most relevant templates based on the intent, and then modifies and combines them according to the parameters, which greatly improves the efficiency and reliability of code generation.

[0067] In another embodiment, a multi-part collaborative voice AI programming facial expression robot control system also includes host computer debugging software. This software is connected to the main control unit through a debugging interface and can be used to monitor the system status in real time, visualize the digital twin simulation process, allow manual intervention for debugging, update the emotion and action mapping library, and manage A / B partition firmware, thus facilitating development and maintenance.

Claims

1. A multi-part collaborative voice AI programming facial expression robot control system, characterized in that, include: The main control unit is a multi-core heterogeneous processor that integrates an application processing core for running AI algorithms and operating systems and a real-time control core for real-time parallel control of multi-channel actuators. The voice processing module integrates a hardware audio preprocessing circuit for hardware-level noise reduction, echo cancellation, and sound source localization of microphone signals, outputting a clean voice signal. The AI ​​acceleration unit is used to accelerate the physical computation in the speech recognition model, key information extraction model, intent reasoning module, and digital twin simulation module. The secure storage area is used to store verified system programs, user configuration files, emotion and action mapping libraries, and system recovery images. An actuator drive circuit, which is a multi-channel drive circuit, communicates with the main control unit through a high real-time bus and is used to drive actuators in multiple parts of the robot. The output of the voice processing module is connected to the input of the AI ​​acceleration unit; the output of the AI ​​acceleration unit is connected to the application processing core of the main control unit; the output of the real-time control core of the main control unit is connected to the input of the actuator drive circuit; the secure storage area is bidirectionally connected to the main control unit; the main control unit is connected to the outside through the debugging and program injection interface.

2. The multi-part collaborative voice AI programming facial expression robot control system according to claim 1, characterized in that, The clean speech signal output by the speech processing module is converted into text by the speech recognition model running by the AI ​​acceleration unit, and the key information extraction model extracts a structured control intent vector containing action type, location, intensity, speed, and frequency from the text. The intent reasoning module running by the AI ​​acceleration unit receives and processes the structured control intent vector and outputs control command parameters for simulation verification by the digital twin simulation module.

3. The multi-part collaborative voice AI programming facial expression robot control system according to claim 2, characterized in that, The emotion-action mapping library stored in the secure storage area is used to map the structured control intent vector to specific actuator parameters. The emotion-action mapping library is constructed by combining expert rules and data-driven methods, and includes emotion commands, facial expression parameters, associated body movements, and intensity codes.

4. The multi-part collaborative voice AI programming facial expression robot control system according to claim 3, characterized in that, The main control unit runs an AI programming agent, which adopts a retrieval-enhanced generation architecture. It receives the structured control intent vector and the robot's current state information, and automatically generates or modifies executable control code by querying the emotion-action mapping library and the physical constraint library. The AI ​​acceleration unit runs a digital twin simulation module to simulate and verify the generated control code.

5. The multi-part collaborative voice AI programming facial expression robot control system according to claim 4, characterized in that, The digital twin simulation module is connected to the AI ​​debugging agent. If the simulation fails, the AI ​​debugging agent starts a multi-objective optimization algorithm to fine-tune and optimize the parameters or logic of the control code, and starts iterative simulation until the verification is successful.

6. The multi-part collaborative voice AI programming facial expression robot control system according to claim 1, characterized in that, The multi-part collaborative voice AI programming facial expression robot control system adopts A / B dual firmware partitioning and timers. After the new program is verified, it is burned to partition B for operation, while partition A backs up the old version of the system program. If the operation is abnormal, the system will automatically roll back to the stable version in partition A through the timer.

7. The multi-part collaborative voice AI programming facial expression robot control system according to claim 1, characterized in that, The multi-site actuators include a face actuator, a neck actuator, a torso actuator, and an arm actuator.

8. A method for controlling a multi-part collaborative voice AI programming facial expression robot, applied to the multi-part collaborative voice AI programming facial expression robot control system as described in any one of claims 1-7, characterized in that, Including the following steps: S1. Voice command capture and semantic understanding: The voice signal is preprocessed into a clean voice signal by the voice processing module. The clean voice signal is then processed by the voice recognition model, the feature extraction model and the key information extraction model to output a structured control intent vector. S2. AI-driven parametric programming and code generation: The AI ​​programming agent queries the emotion-action mapping library and physical constraint library based on the structured control intent vector, and automatically generates or modifies executable control code. S3. Virtual simulation and automated debugging: The generated control code is simulated and verified through the digital twin simulation module, and the control code is optimized through AI debugging agent, and the control code is iterated until it passes; S4. Secure burning and execution: Securely burn the verified control code into the main control unit, and monitor the actuator status in real time during execution to form a closed-loop feedback.

9. The multi-part collaborative voice AI programming facial expression robot control method according to claim 8, characterized in that, The AI-driven parametric programming and code generation also includes the AI ​​programming agent using retrieval-enhanced generation technology to retrieve pre-stored code templates based on intent reasoning results, and modify and combine them with parameters. The generation process incorporates security rules and evaluates the emotional fidelity of the generated results.

10. The multi-part collaborative voice AI programming facial expression robot control method according to claim 8, characterized in that, The secure programming and execution also includes comparing the verification code after programming is completed, and feeding back the actuator status data collected in real time during execution to the AI ​​programming agent for evaluating the execution effect and optimizing subsequent instructions.

Images