Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation

TL;DR

This paper introduces Mobile ALOHA, a low-cost teleoperation system enabling imitation learning for complex bimanual mobile manipulation tasks, demonstrating significant success rate improvements through data augmentation.

Contribution

The work presents a novel mobile whole-body teleoperation system and shows how co-training with static datasets enhances mobile manipulation performance.

Findings

Co-training increases success rates by up to 90%.

Mobile ALOHA enables autonomous complex tasks like cooking and door opening.

System effectively combines mobile base and whole-body control for manipulation.

Abstract

Imitation learning from human demonstrations has shown impressive performance in robotics. However, most results focus on table-top manipulation, lacking the mobility and dexterity necessary for generally useful tasks. In this work, we develop a system for imitating mobile manipulation tasks that are bimanual and require whole-body control. We first present Mobile ALOHA, a low-cost and whole-body teleoperation system for data collection. It augments the ALOHA system with a mobile base, and a whole-body teleoperation interface. Using data collected with Mobile ALOHA, we then perform supervised behavior cloning and find that co-training with existing static ALOHA datasets boosts performance on mobile manipulation tasks. With 50 demonstrations for each task, co-training can increase success rates by up to 90%, allowing Mobile ALOHA to autonomously complete complex mobile manipulation tasks such as sauteing and serving a piece of shrimp, opening a two-door wall cabinet to store heavy cooking pots, calling and entering an elevator, and lightly rinsing a used pan using a kitchen faucet. Project website: https://mobile-aloha.github.io