This is a repository for a mobile robot control using Unity Machine Learning Agent (ML-Agents) and Robot Operating System (ROS).
The Reinforcement Learning packages are obtained from ML-Agents toolkit, which is an open-source project that enables games and simulations to serve as an environment for training agents.
Some of the software and hardware requirements (Preferred) are listed below:
- Windows10 (64bit), Ubuntu18.04
- Python 3.6.13
- Anaconda 4.5.4
- Tensorflow 2.3.4
- mlagents 0.20.0
- Unity version 2020.3.26f1
- ML-Agents 1.4.0-Preview
- ROS TCP Connector 0.7.0-Preview
- URDF Importer
- CPU: Intel Xeon(R) W-1250 CPU @ 3.30GHZ * 12
- GPU : GeForce RTX 3080
- Driver : 470.161.03
- Memory : 10GB