ROS2Learn: a reinforcement learning framework for ROS 2

2019/03/14 by Yue Leire Erro Nuin, Nestor Gonzalez Lopez, Nuin, Yue Leire Erro +11
Computer Science · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Robotics (cs.RO) #cs.AI #cs.LG #cs.RO

paper · pdf · doi:10.48550/arxiv.1903.06282

arxiv created 2019/03/18 · arxiv updated 2019/03/19

Abstract

We propose a novel framework for Deep Reinforcement Learning (DRL) in modular robotics to train a robot directly from joint states, using traditional robotic tools. We use an state-of-the-art implementation of the Proximal Policy Optimization, Trust Region Policy Optimization and Actor-Critic Kronecker-Factored Trust Region algorithms to learn policies in four different Modular Articulated Robotic Arm (MARA) environments. We support this process using a framework that communicates with typical tools used in robotics, such as Gazebo and Robot Operating System 2 (ROS 2). We evaluate several algorithms in modular robots with an empirical study in simulation.

Related

• gym-gazebo2, a toolkit for reinforcement learning using ROS 2 and Gazebo
• Evaluation of Deep Reinforcement Learning Methods for Modular Robots
• Deep Reinforcement Learning for Robotics: A Survey of Real-World Successes
• PIC4rl-gym: a ROS2 modular framework for Robots Autonomous Navigation with Deep Reinforcement Learning
• Taxonomy and Trends in Reinforcement Learning for Robotics and Control Systems: A Structured Review