6-robot

6-robot / jie_3d_nav

Public

一套基于 ROS 2 Humble 的 3D 导航系统，通过 Web 界面交互。本系统已在智元科技 D1 机器狗以及留形科技 Odin 1 空间定位模组上测试通过。

93
24
100% credibility
Found May 11, 2026 at 93 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

An open-source 3D navigation system for ROS2 robots that imports maps from scans or simulations, lets you view and edit them via web or GUI, plans paths, and supports real robot testing.

How It Works

1
📱 Watch the demo video

You find this 3D robot navigation tool through a fun video showing a dog-like robot exploring rooms and fields smoothly.

2
💻 Run the easy setup script

On your Ubuntu computer, you copy the project folder and run a simple script that installs everything needed.

3
Pick your map type
☁️
Point cloud or 2D scan

Load a colorful 3D scan from your robot's sensors.

🏗️
Simulation world

Use a ready-made scene like a building or field.

4
👁️ Open the 3D viewer

A web page or window pops up showing your map as colorful blocks you can spin around.

5
🎯 Click start and goal spots

Tap green walkable areas to place start and goal markers, drag to set direction, and watch a smooth path appear instantly!

6
💾 Save your custom map

Tweak blocks if needed, then save the map package for reuse anytime.

🚀 Robot navigates perfectly

Your robot follows the path flawlessly, exploring safely thanks to your prepared 3D map!

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 93 to 93 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is jie_3d_nav?

jie_3d_nav is a GitHub ROS 2 Humble project delivering a full 3D navigation stack with Web and GUI interfaces for map handling and path planning. Users import PCD point clouds, ROS 2D occupancy grids, or Gazebo worlds into OctoMap format, edit voxels via Qt tools, save/load map packages, and test planning over Web at http://robot-ip:8080. Built in Python with C++ nodes, it launches via ros2 launch commands like web_octomap.launch.py for browser-based start/goal selection and nav.launch.py for hardware integration.

Why is it gaining traction?

It stands out in GitHub ROS planning and perception repos by bundling map import, editing, and 3D planning into simple launches, skipping manual OctoMap tweaks common in ROS Humble tutorials. The Web viewer for real-time path viz and point picking hooks devs needing quick iteration without RViz setup, plus hardware-tested configs for ROS humble docker images streamline deployment on Ubuntu 22.04.

Who should use this?

ROS developers prototyping 3D nav on legged robots like AgiBot D1 or modules like Odin 1, especially those extending GitHub ROS controls or industrial stacks. Ideal for teams handling uneven terrain who want GUI/Web tools over raw octomap_msgs, or evaluators of ROS-GZ sim-to-real pipelines.

Verdict

Grab it for ROS Humble desktop experiments—solid docs, install script, and Bilibili/YouTube demos make setup fast despite 93 stars and 1.0% credibility score signaling early maturity. Pair with ros-humble-ros1_bridge for hybrid setups, but test thoroughly before production.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.