AUV Project

• Project presentation is ready in docs.
• "AUV Static Environment Simulation" can be run directly in releases.
• To create Unity Environment to train and simulation the Agent, please follow the instructions below.

Prerequests

1. Install Python 3.9.x
2. Install Unity Editor
3. Create a virtual environment

Virtual Environment (Windows)

Run the following commands.

$ python -m venv sample-env
$ .\sample-env\Scripts\activate
$ python -m pip install --upgrade pip
$ pip install -r requirements_py39.txt

Note that on Windows, you may also need Microsoft's Visual C++ Redistributable if you don't have it already. See the PyTorch installation guide for more installation options and versions.

Add Environment Project

Project Import Steps

1. Open Unity Hub.
2. Select Open and choose AUV folder.
3. Ignore warnings and choose Continue
4. Start with Safety Mode
5. Open the Package Manager under Window Menu

6. Click the + button and select Add package from disk.. option
7. Locate to com.unity.ml-agents under root. Select package.json
8. Apply same steps for com.unity.ml-agents-extensions

Setup Scene

1. Double click to SampleScene object under the Scenes folder.