Adaptive Mobile Manipulation for Articulated Objects In the Open World

TL;DR

This paper presents an adaptive mobile manipulation system enabling robots to operate articulated objects like doors and cabinets in unstructured environments, using online learning and a low-cost hardware platform, significantly improving success rates.

Contribution

The work introduces a full-stack adaptive manipulation system with online learning capabilities and a low-cost hardware platform for real-world articulated object manipulation.

Findings

Success rate increased from 50% to 95% after online adaptation.

System operates effectively with less than an hour of online learning per object.

Validated on 20 objects across 4 buildings in real-world settings.

Abstract

Deploying robots in open-ended unstructured environments such as homes has been a long-standing research problem. However, robots are often studied only in closed-off lab settings, and prior mobile manipulation work is restricted to pick-move-place, which is arguably just the tip of the iceberg in this area. In this paper, we introduce Open-World Mobile Manipulation System, a full-stack approach to tackle realistic articulated object operation, e.g. real-world doors, cabinets, drawers, and refrigerators in open-ended unstructured environments. The robot utilizes an adaptive learning framework to initially learns from a small set of data through behavior cloning, followed by learning from online practice on novel objects that fall outside the training distribution. We also develop a low-cost mobile manipulation hardware platform capable of safe and autonomous online adaptation in unstructured environments with a cost of around 20,000 USD. In our experiments we utilize 20 articulate objects across 4 buildings in the CMU campus. With less than an hour of online learning for each object, the system is able to increase success rate from 50% of BC pre-training to 95% using online adaptation. Video results at https://open-world-mobilemanip.github.io/