ponta17

https://github.com/Yutarop/waypoints_editor

When robot navigation scales from hundreds of meters to several kilometers, managing waypoints as raw numeric values quickly becomes painful and error-prone.This provides a GUI tool that lets you visually edit waypoints on a map, automatically generate them from start and goal positions, and batch-edit multiple waypoints at once, making medium- to large-scale navigation much easier to manage.

https://github.com/Yutarop/comic-generator

Last week, Google released Nano Banana Pro, and I was amazed by its drawing capabilities. So, I decided to experiment and built a tool that can generate multi-page manga.Even though there are still some typos in the speech bubbles, the level of detail in the artwork is incredible. Here’s a demo of some manga generated with it:https://github.com/Yutarop/comic-generator/blob/main/demo/le...For...

https://github.com/Yutarop/turtlebot3_agent

I developed an AI agent that controls a TurtleBot3 in the real world. The repository includes a demo video showing the TurtleBot3 following a prompt to move in a square shape, with each side 0.5 meters long. If you're interested, please have a look.

https://github.com/Yutarop/sign_detection

I've been working on a ROS 2 package that detects route blockage signboards used in the Tsukuba Challenge. The approach relies entirely on point cloud data and intensity information from a 3D LiDAR sensor without camera input.

https://github.com/Yutarop/ga-pixel-art

This project is a simple but fun demonstration of a genetic algorithm applied to image generation. It starts with a population of random images and evolves them over generations to resemble a target image. The output is an animated GIF that shows the entire evolution process.

https://github.com/Yutarop/ros-mcp

I've been experimenting with ROS (Robot Operating System) and the Model Context Protocol (MCP), and noticed a gap: while there were some MCP servers for ROS, most were limited. They either only supported topic communication or had hard-coded topic names, which severely limited their reusability.To address this, I built a general-purpose MCP server for ROS that allows interaction with ROS...

https://github.com/Yutarop/turtlebot3_agent

I've developed an AI agent for the TurtleBot3 mobile robot using ROS2. This agent can interpret commands such as “move to (X, Y)” or “check surroundings,” and dynamically selects from a toolkit of actions, including movement, sensor data retrieval (from odometry, LiDAR, or camera), and basic navigation decisions. So far, I've tested the agent in the Gazebo simulator. If you're interested,...

https://github.com/Yutarop/turtlebot3_agent

A week ago, I shared a project called Turtlesim Agent — an AI agent that lets users control the classic turtlesim turtle using natural language commands like “draw a triangle” or “make a red star.”Now, I’ve taken the same language-to-robot-behavior idea and applied it to a real robot: the TurtleBot3.TurtleBot3 Agent interprets natural language instructions and selects from a set of tools to...

https://github.com/Yutarop/two_claps_open

Recently, I watched Iron Man and was inspired by the scene where Tony Stark claps twice to activate his computer. (https://www.youtube.com/watch?v=OT2b5KzMoC0&t=101s) I thought, “I want to try that!".So I wrote a simple script that opens Google Chrome — and optionally a local file — when I clap twice.Technically, I recorded sample clap sounds and analyzed their frequency content in MATLAB...

https://github.com/Yutarop/turtlesim_agent

I'm a grad student studying robotics, with a particular interest in the intersection of LLMs and mobile robots. Recently, I discovered how easily LangChain enables the creation of AI agents, and I wanted to explore how such agents could interact with simulated environments.So, I built TurtleSim Agent, an AI agent that turns the classic ROS 2 turtlesim turtle into a creative artist.With this...