Robot Learning in Homes: Improving Generalization and Reducing Dataset Bias

by   Abhinav Gupta, et al.

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 architecture which explicitly models the latent noise in the dataset also performed 10 effort inspires the robotics community to look outside the lab and embrace learning based approaches to handle inaccurate cheap robots.


page 2

page 4

page 5

page 7


MuSHR: A Low-Cost, Open-Source Robotic Racecar for Education and Research

We present MuSHR, the Multi-agent System for non-Holonomic Racing. MuSHR...

An Open Tele-Impedance Framework to Generate Large Datasets for Contact-Rich Tasks in Robotic Manipulation

Using large datasets in machine learning has led to outstanding results,...

The Lab vs The Crowd: An Investigation into Data Quality for Neural Dialogue Models

Challenges around collecting and processing quality data have hampered p...

3D Annotation Of Arbitrary Objects In The Wild

Recent years have produced a variety of learning based methods in the co...

Data Generation for Learning to Grasp in a Bin-picking Scenario

The rise of deep learning has greatly transformed the pipeline of roboti...

All You Need is LUV: Unsupervised Collection of Labeled Images using Invisible UV Fluorescent Indicators

Large-scale semantic image annotation is a significant challenge for lea...

A First Look at Dataset Bias in License Plate Recognition

Public datasets have played a key role in advancing the state of the art...

Please sign up or login with your details

Forgot password? Click here to reset