TY - JOUR
T1 - Robot learning in homes
T2 - 32nd Conference on Neural Information Processing Systems, NeurIPS 2018
AU - Gupta, Abhinav
AU - Murali, Adithyavairavan
AU - Gandhi, Dhiraj
AU - Pinto, Lerrel
N1 - Funding Information:
ACKNOWLEDGEMENTS This work was supported by ONR MURI N000141612007. Abhinav was supported in part by Sloan Research Fellowship and Adithya was partly supported by a Uber Fellowship.
Funding Information:
This work was supported by ONR MURI N000141612007. Abhinav was supported in part by Sloan Research Fellowship and Adithya was partly supported by a Uber Fellowship.
Publisher Copyright:
© 2018 Curran Associates Inc.All rights reserved.
PY - 2018
Y1 - 2018
N2 - Data-driven approaches to solving robotic tasks have gained a lot of traction in recent years. However, most existing policies are trained on large-scale datasets collected in curated lab settings. If we aim to deploy these models in unstructured visual environments like people's homes, they will be unable to cope with the mismatch in data distribution. In such light, we present the first systematic effort in collecting a large dataset for robotic grasping in homes. First, to scale and parallelize data collection, we built a low cost mobile manipulator assembled for under 3K USD. Second, data collected using low cost robots suffer from noisy labels due to imperfect execution and calibration errors. To handle this, we develop a framework which factors out the noise as a latent variable. Our model is trained on 28K grasps collected in several houses under an array of different environmental conditions. We evaluate our models by physically executing grasps on a collection of novel objects in multiple unseen homes. The models trained with our home dataset showed a marked improvement of 43.7% over a baseline model trained with data collected in lab. Our architecture which explicitly models the latent noise in the dataset also performed 10% better than one that did not factor out the noise. We hope this effort inspires the robotics community to look outside the lab and embrace learning based approaches to handle inaccurate cheap robots.
AB - Data-driven approaches to solving robotic tasks have gained a lot of traction in recent years. However, most existing policies are trained on large-scale datasets collected in curated lab settings. If we aim to deploy these models in unstructured visual environments like people's homes, they will be unable to cope with the mismatch in data distribution. In such light, we present the first systematic effort in collecting a large dataset for robotic grasping in homes. First, to scale and parallelize data collection, we built a low cost mobile manipulator assembled for under 3K USD. Second, data collected using low cost robots suffer from noisy labels due to imperfect execution and calibration errors. To handle this, we develop a framework which factors out the noise as a latent variable. Our model is trained on 28K grasps collected in several houses under an array of different environmental conditions. We evaluate our models by physically executing grasps on a collection of novel objects in multiple unseen homes. The models trained with our home dataset showed a marked improvement of 43.7% over a baseline model trained with data collected in lab. Our architecture which explicitly models the latent noise in the dataset also performed 10% better than one that did not factor out the noise. We hope this effort inspires the robotics community to look outside the lab and embrace learning based approaches to handle inaccurate cheap robots.
UR - http://www.scopus.com/inward/record.url?scp=85063310771&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85063310771&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:85063310771
SN - 1049-5258
VL - 2018-December
SP - 9094
EP - 9104
JO - Advances in Neural Information Processing Systems
JF - Advances in Neural Information Processing Systems
Y2 - 2 December 2018 through 8 December 2018
ER -