Setting up a Reinforcement Learning Task with a Real-World Robot

03/19/2018
by   A. Rupam Mahmood, et al.
0

Reinforcement learning is a promising approach to developing hard-to-engineer adaptive solutions for complex and diverse robotic tasks. However, learning with real-world robots is often unreliable and difficult, which resulted in their low adoption in reinforcement learning research. This difficulty is worsened by the lack of guidelines for setting up learning tasks with robots. In this work, we develop a learning task with a UR5 robotic arm to bring to light some key elements of a task setup and study their contributions to the challenges with robots. We find that learning performance can be highly sensitive to the setup, and thus oversights and omissions in setup details can make effective learning, reproducibility, and fair comparison hard. Our study suggests some mitigating steps to help future experimenters avoid difficulties and pitfalls. We show that highly reliable and repeatable experiments can be performed in our setup, indicating the possibility of reinforcement learning research extensively based on real-world robots.

READ FULL TEXT
research
09/20/2018

Benchmarking Reinforcement Learning Algorithms on Real-World Robots

Through many recent successes in simulation, model-free reinforcement le...
research
07/06/2020

robo-gym – An Open Source Toolkit for Distributed Deep Reinforcement Learning on Real and Simulated Robots

Applying Deep Reinforcement Learning (DRL) to complex tasks in the field...
research
03/13/2018

Flexible Robotic Grasping with Sim-to-Real Transfer based Reinforcement Learning

Robotic manipulation requires a highly flexible and compliant system. Ta...
research
03/15/2022

Bi-Manual Manipulation and Attachment via Sim-to-Real Reinforcement Learning

Most successes in robotic manipulation have been restricted to single-ar...
research
09/09/2019

A Survey on Reproducibility by Evaluating Deep Reinforcement Learning Algorithms on Real-World Robots

As reinforcement learning (RL) achieves more success in solving complex ...
research
04/16/2020

MobiAxis: An Embodied Learning Task for Teaching Multiplication with a Social Robot

The use of robots in educational settings is growing increasingly popula...
research
03/23/2022

Asynchronous Reinforcement Learning for Real-Time Control of Physical Robots

An oft-ignored challenge of real-world reinforcement learning is that th...

Please sign up or login with your details

Forgot password? Click here to reset