Arena-Rosnav/rosnav-rl

RosNav-RL is a modular DRL framework for ROS 2 with a pluggable architecture, allowing you to switch between RL backends like Stable-Baselines3 and DreamerV3 to accelerate research and deployment.

37
/ 100
Emerging

This project provides a research toolkit for developing and deploying robot navigation agents using reinforcement learning (RL) within the ROS 2 ecosystem. It takes robot sensor data and desired reward functions as input, and outputs trained agents that can issue movement commands (Twist messages) to navigate a robot. Robotics researchers and developers focused on autonomous navigation would find this particularly useful.

Use this if you are a robotics researcher or developer looking to quickly prototype, train, and deploy RL-based navigation policies for your ROS 2-enabled mobile robots, especially when experimenting with different RL algorithms or reward structures.

Not ideal if you are looking for a pre-trained, off-the-shelf navigation solution without needing to engage with reinforcement learning frameworks or customize policies.

robotics autonomous-navigation reinforcement-learning robot-control ROS-2-development
No License No Package No Dependents
Maintenance 10 / 25
Adoption 9 / 25
Maturity 8 / 25
Community 10 / 25

How are scores calculated?

Stars

77

Forks

6

Language

Python

License

Last pushed

Mar 13, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Arena-Rosnav/rosnav-rl"

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