rl-texplore-ros-pkg: Reinforcement learning framework, agents, and environments with ROS interface. This project is a framework for running reinforcement learning experiments through ROS. Agents and Environments communicate actions, states, and rewards through a set of ROS messages. The code includes numerous environments (gridworlds, mountain car, cart pole, etc) as well as agents. It also includes a framework for model based agents where various model learning and exploration modules can be inserted along with a variety of available planners (value iteration, policy iteration, prioritized sweeping, uct, parallel uct). It also includes the TEXPLORE algorithm, which uses random forest models, along with an architecture to run model-based RL algorithms in real-time. This repository has been developed by Todd Hester at the University of Texas at Austin.
References in zbMATH (referenced in 1 article )
Showing result 1 of 1.
- Geramifard, Alborz; Dann, Christoph; Klein, Robert H.; Dabney, William; How, Jonathan P.: RLPy: a value-function-based reinforcement learning framework for education and research (2015)