Scholarlink

This paper designs and implements Robot Dog, a multifunctional desktop pet robot with low cost and extensibility. It runs on Arduino Nano, ESP32-S3-DevKitC-1, and Arduino Nano RP2040 Connect which support natural language interaction with large language model (LLM) and automatic speech recognition with ASR module. Furthermore, Bluetooth is used for manual control and infrared sensor is used to detect cliff and avoid falling. A novel tri-controller architecture is proposed such that the RP2040 Connect is used to implement IoT, i.e., to monitor the states of Robot Dog and send fall alerts to Arduino Cloud through IoT controller. The ESP32-S3 calls Tongyi Qwen LLM API to translate user dialogue represented by natural language to walking or greeting. The Robot Dog system with sensor fusion, hardware integration and layered interaction model: high-level goal driven by LLM and low-level servo control command at 50 Hz. Experiments confirm the system’s effectiveness for AI-powered interaction on embedded platforms, with applications in education and entertainment.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Design and Implementation of a Multifunctional Desktop Pet Robot Dog Based on Arduino Nano and ESP32-S3

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Di Li,
  • Junkai Lin,
  • Siqi Hou,
  • Yanyan Ji

摘要

This paper designs and implements Robot Dog, a multifunctional desktop pet robot with low cost and extensibility. It runs on Arduino Nano, ESP32-S3-DevKitC-1, and Arduino Nano RP2040 Connect which support natural language interaction with large language model (LLM) and automatic speech recognition with ASR module. Furthermore, Bluetooth is used for manual control and infrared sensor is used to detect cliff and avoid falling. A novel tri-controller architecture is proposed such that the RP2040 Connect is used to implement IoT, i.e., to monitor the states of Robot Dog and send fall alerts to Arduino Cloud through IoT controller. The ESP32-S3 calls Tongyi Qwen LLM API to translate user dialogue represented by natural language to walking or greeting. The Robot Dog system with sensor fusion, hardware integration and layered interaction model: high-level goal driven by LLM and low-level servo control command at 50 Hz. Experiments confirm the system’s effectiveness for AI-powered interaction on embedded platforms, with applications in education and entertainment.