voxelgym2D 0.2

Gym environment for 2D grid path planning

voxelgym2D

A gym environment for voxel/grid based reinforcement learning for path planning.

Results with SB3 (v1.6.2) : PPO :smile:

Here are the results of training a PPO agent on the onestep-v0 using the example here. Below you will find the episode reward and episode length over steps during training. As the agent learns, the episode reward increases and the episode length reduces are the agent learns to identify the goal and reach it in the shortest possible path.

Installation

git clone https://github.com/harisankar95/voxelgym2D.git ~/path/to/repo
cd ~/path/to/repo
pip install .

# to aditionally install stable_baselines 3 and pytorch (optional)
pip install .[sb3]

or directly from github :smile:

pip install git+https://github.com/harisankar95/voxelgym2D.git

Development

To install the package in development mode, run the following command in the root directory of the repository:

pip install -e .[dev]

# to aditionally install stable_baselines 3 and pytorch (optional)
pip install -e .[dev,sb3]

Usage

import voxelgym2D
import gym

env = gym.make("voxelgym2D:onestep-v0")
env.reset()
env.render()

Examples

The examples can be found here.

License

This project is licensed under the terms of the MIT license.

Documentation

The documentation can be found here.

Changelog

0.1.0

• Initial release of voxelgym 2D environments tested with stable_baselines 3 (v1.6.2) and python 3.8

TODO

• Add 2D environments
• Test with gym 0.26.2
• Add documentation

Known issues

• Currently only supports gym==0.21.0 :neutral_face:,hence setuptools==65.5.0 is required to install gym.

Contributing

Contributions are welcome! Please open an issue or a pull request.

References

• OpenAI Gym
• Stable Baselines 3