EricChen0104/ppo-icm-maze-exploration

A curiosity-driven PPO + ICM reinforcement learning agent for autonomous maze exploration and victim rescue — built to evolve into a full SLAM-based search and rescue system.

21
/ 100
Experimental

This project helps roboticists and AI researchers develop autonomous agents capable of navigating unknown mazes to find and rescue hidden "victims." It takes maze layouts and an agent's visual observations as input, producing trained navigation strategies that prioritize exploration and victim discovery. This is useful for researchers building search-and-rescue robots or exploring advanced AI navigation.

No commits in the last 6 months.

Use this if you are developing or researching AI agents for autonomous exploration and victim location in complex, unknown environments.

Not ideal if you need a production-ready search and rescue system, as this is an early-stage research prototype.

robotics autonomous-navigation search-and-rescue AI-research environmental-exploration
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 4 / 25
Maturity 15 / 25
Community 0 / 25

How are scores calculated?

Stars

8

Forks

Language

Python

License

MIT

Last pushed

Jul 23, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/EricChen0104/ppo-icm-maze-exploration"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.