Robotics Developer | RoboMaster Engineer | ROS2 / Embedded Control / Robot Learning

I am a robotics developer focused on robot control, embedded systems, simulation, and learning-based robotics.

I have worked on RoboMaster robot control systems, embedded communication, chassis and gimbal control, and robot software development. I am also exploring ROS2, MuJoCo, Pinocchio, reinforcement learning, humanoid robots, robotic arms, and embodied AI systems.

• 🔧 Interested in robot control, embedded development, and real-time systems
• 🤖 Focused on RoboMaster, mobile robots, humanoid robots, and robotic arms
• 🚀 Working with C/C++, Python, ROS2, MuJoCo, Pinocchio, RBDL, FreeRTOS, and Linux
• 🧠 Learning reinforcement learning, motion planning, robot dynamics, and sim-to-real
• 🎯 Goal: building robots that can perceive, move, decide, and interact with the real world

• Robot Motion Control
• Embedded Robot Software
• ROS2 Robot System Development
• Robot Dynamics and Inverse Dynamics
• Motion Planning and Trajectory Control
• Reinforcement Learning for Robotics
• Sim-to-Real Transfer
• Humanoid and Legged Robot Control
• Robotic Arm Planning and Manipulation
• Vision-Language-Action Models for Robotics
• AI Applications for Engineering Productivity

A RoboMaster four-swerve-wheel infantry robot control project, including chassis control and gimbal control.

Highlights

• Four-swerve-wheel chassis motion control
• Gimbal yaw / pitch control
• Shooting mechanism control
• CAN bus motor communication
• IMU data processing and attitude compensation
• Vision communication interface
• Robot power control and device status checking
• FreeRTOS-based multi-task scheduling
• STM32 embedded development with C

Keywords: RoboMaster, STM32, FreeRTOS, Embedded Control, Chassis Control, Gimbal Control, CAN, IMU

An AI-powered campus recruitment job matching agent for resume analysis, job recommendation, and career planning.

Offer Catcher helps students upload their resume, extract structured profile information, match suitable campus recruitment positions, and generate multi-dimensional analysis with resume optimization suggestions.

Highlights

• Resume upload and parsing for PDF / DOCX files
• AI-powered structured resume analysis
• Intelligent job recommendation based on candidate profiles
• Multi-dimensional matching score dashboard
• Skill gap diagnosis and improvement suggestions
• Resume optimization with STAR-style rewriting examples
• Career roadmap generation for target positions
• Full-stack implementation with React, TypeScript, FastAPI, and Docker
• DeepSeek / Claude-compatible AI API integration

Keywords: AI Job Matching, Campus Recruitment, Resume Analysis, FastAPI, React, TypeScript, Docker, Career Planning

A customized ESP32-based AI voice interaction project for vehicle-mounted robot development.

This project explores how embedded devices can serve as an AI interaction入口, connecting large language models such as Qwen / DeepSeek with hardware-side control through voice interaction and MCP-style communication.

Highlights

• ESP32-S3 embedded development
• AI voice interaction入口
• Vehicle-mounted robot scenario
• MCP-based interaction concept
• Qwen / DeepSeek model integration exploration
• Customized development based on an open-source ESP32 AI project

Keywords: ESP32, Embedded AI, Voice Interaction, MCP, Vehicle Robot, C++, LLM

• Building ROS2-based robot systems
• Improving embedded robot control skills
• Learning MuJoCo simulation and reinforcement learning
• Migrating robot dynamics code from RBDL to Pinocchio
• Studying humanoid robot control and locomotion
• Exploring robotic arms, imitation learning, and VLA models
• Building practical AI agents for real engineering and job-search scenarios

RoboMaster · STM32 · FreeRTOS · ROS2 · Embedded Control · Robot Dynamics · MuJoCo · Pinocchio · Reinforcement Learning · ESP32 · AI Job Matching · Full-Stack AI Application · Embodied AI

• GitHub: @REALGSY
• Name: KIYOKU / REALGSY
• Focus: Robotics, Control, Simulation, Embedded Systems, Reinforcement Learning, AI Applications

Building robots that move, think, and act in the real world.