Robotics Engineering Undergraduate | ROS1 / ROS2 | Computer Vision | Embodied Intelligence
机器人工程本科生｜ROS1 / ROS2｜计算机视觉｜具身智能
I am an undergraduate student in Robotics Engineering at Changchun University of Technology (expected graduation: 2026).
My interests focus on robot perception, embodied interaction, autonomous navigation, and embedded control systems.
I am still learning and deeply value rigorous experimentation, reproducibility, and teamwork in academic competitions and engineering practice.
我是长春工业大学机器人工程专业本科生（预计 2026 年毕业）。
研究与实践兴趣聚焦于机器人感知、具身交互、自主导航与嵌入式控制系统。
目前仍在持续学习，重视在学科竞赛与工程实践中的严谨实验、可复现性与团队协作。
- 🔭 Currently working on ROS1 / ROS2, Nav2, SLAM, and multimodal robot system integration.
🔭 当前主要方向：ROS1 / ROS2、Nav2、SLAM 与多模态机器人系统集成 - 🌱 Continuously improving in Python, OpenCV, YOLO, PaddleOCR, and vision-language interaction pipelines.
🌱 持续提升：Python、OpenCV、YOLO、PaddleOCR 与视觉-语言交互链路 - 🛠 Practical experience on STM32, Arduino, NVIDIA Jetson, and Raspberry Pi for rapid prototyping and deployment.
🛠 具备 STM32、Arduino、NVIDIA Jetson、Raspberry Pi 平台原型开发与部署经验 - 🏆 Since 2023, serving as the student lead of a university engineering training laboratory.
🏆 自 2023 年起担任校内工程训练实验室学生负责人 - 📫 Contact: 2303886347@qq.com | Douyin: xuandj
📫 联系方式：2303886347@qq.com｜抖音：xuandj - Hiwonder社区博客主页 | The homepage of Hiwonder's community blog： https://forum.hiwonder.com.cn/u/mr.dj/summary
- Embodied intelligence and human-robot interaction in open environments
开放环境下的具身智能与人机交互 - Vision-based perception, OCR understanding, and robust real-time pipelines
基于视觉的感知、OCR 理解与实时鲁棒处理链路 - SLAM, localization, and navigation for complex mobile platforms
面向复杂移动平台的 SLAM、定位与导航 - Integrated control systems combining embedded controllers and AI modules
融合嵌入式控制器与 AI 模块的一体化控制系统
2025.09 — 2025.12 | ROS1 · YOLO · PaddleOCR · Qwen3 · TTS · Docker
- Designed a live-comment acquisition pipeline based on
mss screenshot + win32gui window locking + YOLO region detection + PaddleOCR + TCP, improving robustness in dynamic streaming interfaces.
设计并实现mss 截图 + win32gui 窗口锁定 + YOLO 区域检测 + PaddleOCR + TCP弹幕采集链路，提升动态直播界面下的鲁棒性。 - Trained a two-stage YOLO detector (comment area / single comment), and improved OCR stability through NumPy-based adaptive cropping.
训练两级 YOLO 检测器（评论区 / 单条评论），并通过 NumPy 自适应裁剪提升 OCR 稳定性。 - Integrated OCR text with Qwen3 JSON outputs, TTS synthesis, and Unitree SDK motion APIs to realize closed-loop voice-motion interaction.
将 OCR 文本与 Qwen3 JSON 输出、TTS 合成和 Unitree SDK 动作接口打通，实现语音-动作闭环交互。
2024.09 — Present | ROS1 · ROS2 · Nav2 · SLAM Toolbox · AMCL · OpenCV
- Completed communication architecture and sensor integration on Jetson, including LiDAR, depth camera, IMU, odometry, TF, and point-cloud modules.
在 Jetson 平台完成通信架构与传感器集成，包括 LiDAR、深度相机、IMU、里程计、TF 与点云模块。 - Implemented mapping, localization, and single-/multi-goal navigation with
slam_toolbox,AMCL, and Nav2 stacks.
基于slam_toolbox、AMCL与 Nav2 实现建图、定位及单点/多点导航能力。 - Combined global/local planners, depth perception, OpenCV pipelines, and inverse kinematics for autonomous navigation and manipulation tasks.
结合全局/局部规划、深度感知、OpenCV 处理链路与逆运动学，支持自主导航与操作任务。
2024.02 — 2025.09 | STM32F103 · Raspberry Pi · OpenCV · RS232 / TTL
- Developed multi-sensor fusion logic (photoelectric, infrared, grayscale) for pose estimation and control strategy execution.
构建光电、红外、灰度等多传感融合逻辑，用于姿态估计与控制策略执行。 - Achieved stable ~30 FPS Tag-code recognition on Raspberry Pi using OpenCV, and connected to STM32 via RS232-to-TTL communication.
在 Raspberry Pi 上基于 OpenCV 实现约 30 FPS 的 Tag 码稳定识别，并通过 RS232 转 TTL 接入 STM32 控制器。 - Iteratively validated control logic through two years of competitions involving 16 university teams.
在连续两年、16 所高校参赛队伍的对抗环境中迭代验证控制逻辑。
- National First Prize, National College Students Information Security and Countermeasures Competition (Humanoid Agent Track), 2025
国家一等奖，全国大学生信息安全与对抗技术竞赛（人形智能体赛道），2025 - National Special Prize, China College Students Engineering Practice and Innovation Ability Competition (Solar Vehicle Track), 2025
国家特等奖，中国大学生工程实践与创新能力大赛（太阳能汽车赛道），2025 - National First Prize & National Champion, China College Students Intelligent Robot Creative Competition (Wheeled Robot Combat), 2025
国家一等奖、全国冠军，中国大学生智能机器人创意大赛（轮式格斗），2025 - National Second Prize, China Robot and Artificial Intelligence Competition (Composite Agent Track), 2025
国家二等奖，中国机器人及人工智能大赛（复合智能体赛道），2025 - National First Prize & National Champion, China Intelligent Robot Combat and Competition (Wheeled Robot Combat), 2024
国家一等奖、全国冠军，中国智能机器人格斗及竞技大赛（轮式格斗），2024 - National Silver Award, China College Students Engineering Practice and Innovation Ability Competition, 2023
国家银奖，中国大学生工程实践与创新能力大赛，2023
Changchun University of Technology · School of Electrical and Electronic Engineering
B.Eng. in Robotics Engineering · 2022.09 — Present · CET-4
长春工业大学 · 电气与电子工程学院
机器人工程本科 · 2022.09 — 至今 · CET-4
Core courses / 核心课程：
Robotics, Robot Operating System, Image Processing and Machine Vision, Sensors and Detection Technology, Arduino Technology and Applications.
机器人学、机器人操作系统、图像处理与机器视觉、传感器与检测技术、Arduino 技术与应用。
Since 2023, I have served as the student lead in the university Engineering Training Center laboratory.
I assist with cross-team coordination among mechanical, electrical, and automation groups, and support project execution, technical sharing, and newcomer training.
自 2023 年起，我担任校工程训练中心实验室学生负责人。
主要协助机械、电气、自动化团队开展协同工作，并参与项目推进、技术交流与新成员培养。