Robot Reinforcement Learning using Crowdsourced Rewards

This work is inspired by the idea that robots should possess the ability to learn new skills with minimum guidance from a dedicated human teacher and they should enhance their abilities by utilizing external resources such as the Internet. In this paper we describe our approach for achieving robot learning by combining basic skill transfer through physical demonstrations, reinforcement learning and web-based crowdsource rewards obtained via remote laboratory. We demonstrate how a robot that is taught very basic motion skills through kinesthetic teaching (i.e. lift, push, etc.), can learn to manipulate several objects through reinforcement learning by taking advantage of the feedback from multiple non-expert crowd workers. Although reinforcement learning is commonly intractable for real robot applications due to high exploration times, we demonstrate that by transferring the training to the crowd, it is possible to achieve robot learning while minimizing the effort of a dedicated teacher. This paper describes the system architecture, as well as the integration of our application to a current remote laboratory framework (Robot Management System). Experimental results with a real robot performing a time consuming learning task show a successful learning performance using crowdsouce feedback.

 

 

Demo Videos

Related Publications

Christian I. Penaloza, Sonia Chernova, Yasushi Mae and Tatsuo Arai: "Robot Reinforcement Learning using Crowdsourced Rewards", IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2013) Cloud Robotics Workshop. Tokyo, Japan. Nov. 3, 2013.

Cloud-based human activity monitoring

This project was submitted by RoboSamurai team to the Cloud Robotics Hackathon 2013. The aim of the competition was to create useful robotic applications that use natural and social human interactions and robot-to-robot collaboration by using cloud-computing and web services. Our projects consited in having one robot that monitors human activity and shares the information to the cloud. In particular, the robot monitored when a person as doing excercise and this activity would be shared to the cloud as a graph. Other robot connected to the cloud, and used the information to encourage the person. Comment from the judges: Team RoboSamurai, truly exploited the full potential of MyRobots.com by a clever usage of the platform. Their application solves a real-life problem, is well implemented, is presented in an engaging ways and, most importantly, it features robot-to-robot collaboration robot-to-human collaboration, and monitoring. By using the specific capabilities of several robots the team is able to track a human working out and encourage him along the way. For this unique application, they win the first prize of 1500$!

 

Demo Videos