Core ACT3 Reinforcement Learning (RL) Library - Core framework and base implementations of common things such as controllers, glues, observes, sensors, evaluation, and etc
Project description
title: CoRL
1. Core ACT3 Reinforcement Learning Library
This repository and corresponding documentation site are currently under construction. We are still porting items and updating instructions for GitHub.
1.1. Summary
The Core ACT3 Reinforcement Learning library (CoRL) is created and maintained by the Air Force Research Laboratory’s (AFRL) Autonomy Capability Team (ACT3). CoRL is intended to enable scalable deep reinforcement learning (RL) experimentation in a manner extensible to new simulations and new ways for the learning agents to interact with them. The objective is to make RL research easier by removing lock-in to particular simulations.
1.1.1. Benefits
- Makes RL environment development significantly easier
- Provides hyper configurable environments, agents and experiments
- Record observations by adding a few lines of config (instead of creating a new file for each observation)
- Reuse glues/dones/rewards between different tasks if they are general
- Uses an episode parameter provider (EPP) to randomize both domain and curriculum learning
- Has an integration first focus, which means that integrating agents to the real world or different simulators is significantly easier
1.1.2. Related Publications
- CoRL: Environment Creation and Management Focused on System Integration
- Inside the special F-16 the Air Force is using to test out ML
- AFRL, AFTC collaborate on future technology via weeklong autonomy summit
- Demonstrating and testing machine learning applications in aerospace
1.2. Documentation
Documentation for the CoRL repository can be accessed directly as files in this repository, as a public documentation site, or can be built locally as an MkDocs site.
1.2.1. Guides
1.2.2. Documentation Web Site
The full public documentation site is available on GitHub pages.
1.2.3. Local Documentation
A local version of the documentation site can be built using MkDocs.
Build the documentation:
mkdocs build
Follow CLI prompts, as needed, to install all required plugins.
Serve the documentation:
mkdocs serve
1.3. Notices and Warnings
1.3.1. Initial Contributors
Initial contributors include scientists and engineers associated with the Air Force Research Laboratory (AFRL), Autonomy Capability Team 3 (ACT3), and the Aerospace Systems Directorate (RQ).
1.3.2. Citing CoRL
If you use CoRL in your work, please use the following BibTeX to cite the CoRL white paper:
@inproceedings{
title={CoRL: Environment Creation and Management Focused on System Integration},
author={Justin D. Merrick, Benjamin K. Heiner, Cameron Long, Brian Stieber, Steve Fierro, Vardaan Gangal, Madison Blake, Joshua Blackburn},
year={2023},
url={https://arxiv.org/abs/2303.02182}
}
To cite the source code, use the Cite this repository option on GitHub to access the reference.
1.3.3. Distribution Statement
Approved for public release: distribution unlimited.
1.3.3.1. Case Number
Date | Release Number | Description |
---|---|---|
2022-05-20 | AFRL-2022-2455 | Initial release |
2023-03-02 | APRS-RYZ-2023-01-00006 | Second release |
2024-21-03 | AFRL-2024-1562 | Third release |
1.3.3.2. Designation Indicator
- Controlled by: Air Force Research Laboratory (AFRL)
- Controlled by: AFRL Autonomy Capability Team (ACT3)
1.3.3.3. Points of Contact
1.3.3.3.1. Repository Contributors
{{ get_authors() }}
1.3.3.3.2. Documentation Contributors
{{ git_site_authors }}
{{ include_glossary_abbreviations() }}
{{ include_bottom_banner() }}
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file corl-3.16.2.tar.gz
.
File metadata
- Download URL: corl-3.16.2.tar.gz
- Upload date:
- Size: 322.6 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/1.8.2 CPython/3.10.12 Linux/5.15.146.1-microsoft-standard-WSL2
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | e99d21c6292e1b425f40b15ba34b9c8bdff82b9350ffcd27f52ea61b2c7647c7 |
|
MD5 | fedceec979a6eb4627d9b3c6bffb8f77 |
|
BLAKE2b-256 | 87dd83bb53081d0a1fdc58298b212f3f42f56b977a68e4dbf587d626d45ad0e5 |
File details
Details for the file corl-3.16.2-cp310-cp310-manylinux_2_35_x86_64.whl
.
File metadata
- Download URL: corl-3.16.2-cp310-cp310-manylinux_2_35_x86_64.whl
- Upload date:
- Size: 538.3 kB
- Tags: CPython 3.10, manylinux: glibc 2.35+ x86-64
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/1.8.2 CPython/3.10.12 Linux/5.15.146.1-microsoft-standard-WSL2
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 06cb9a555821bd4630be466a3f4a3b992360405441a42b5bc7c45f85e03fd856 |
|
MD5 | 52428b4de6a662154d7ac69adfa2e242 |
|
BLAKE2b-256 | ac4ba667a9136b70a176db969632fa0678a584bf5c349563123777d31c5bd1dc |