Open Source Reinforcement Learning Algorithms

AirSim is an open-source, cross platform simulator for drones, cars and more vehicles, built on Unreal Engine with an experimental Unity release in the works. It supports software-in-the-loop simulation with popular flight controllers such as PX4 & ArduPilot and hardware-in-loop with PX4 for physically and visually realistic simulations. It is developed as an Unreal plugin that can simply be dropped into any Unreal environment. AirSim's development is oriented towards the goal of creating a platform for AI research to experiment with deep learning, computer vision and reinforcement learning algorithms for autonomous vehicles. For this purpose, AirSim also exposes APIs to retrieve data and control vehicles in a platform independent way. AirSim is fully enabled for multiple vehicles. This capability allows you to create multiple vehicles easily and use APIs to control them.

Brax is a fast and fully differentiable physics engine for large-scale rigid body simulations, built on JAX. It is designed for research in reinforcement learning and robotics, enabling efficient simulations and gradient-based optimization.

Highly modularized Reinforcement Learning library for real/simulation robots to learn behaviors. Our ultimate goal is to develop an artificial intelligence (AI) program with which the robots can learn to behave as their users wish.

The Teachingbox uses advanced machine learning techniques to relieve developers from the programming of hand-crafted sophisticated behaviors of autonomous agents (such as robots, game players etc...) In the current status we have implemented a well founded reinforcement learning core in Java with many popular usecases, environments, policies and learners. Obtaining the teachingbox: FOR USERS: If you want to download the latest releases, please visit: http://search.maven.org/#search|ga|1|teachingbox FOR DEVELOPERS: 1) If you use Apache Maven, just add the following dependency to your pom.xml: <dependency> <groupId>org.sf.teachingbox</groupId> <artifactId>teachingbox-core</artifactId> <version>1.2.3</version> </dependency> 2) If you want to check out the most recent source-code: git clone https://git.code.sf.net/p/teachingbox/core teachingbox-core Documentation: https://sourceforge.net/p/teachingbox/documentation/HEAD/tree/trunk/manual/

Closed Loop Simulation System (CLSquare) is an integrated architecture to train, test and compare reinforcement learning controllers on different plants. CLSquare provides simulated plants as well as interfaces to real plants.

The Free Connectionist Q-learning Java Framework is an library for developing learning systems. Keywords: qlearning, artificial intelligence, alife, neural nets, neural networks, machine learning, reinforcement learning unsupervised learning agents lejos

Parallel Reinforcement Evolutionary Artificial Neural Networks (PREANN) is a framework of flexible multi-layer ANN's with reinforcement learning based on genetic algorithms and a parallel implementation (using XMM registers and NVIDIA's CUDA).

Physical Symbolic Optimization (Φ-SO) - A symbolic optimization package built for physics. Symbolic regression module uses deep reinforcement learning to infer analytical physical laws that fit data points, searching in the space of functional forms.

This is a third year computer science project. A software system for simulating and animating Reinforcement Learning (RL) algorithms mainly for modular robots.

General purpose agents using reinforcement learning. Combines radial basis functions, temporal difference learning, planning, uncertainty estimations, and curiosity. Intended to be an out-of-the-box solution for roboticists and game developers.