Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

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

About

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

Zipeng Fu, Tony Z. Zhao, Chelsea Finn• 2024

Related benchmarks

TaskDatasetResultRank
Navigation and PlacementTask 2 Environment Center Configuration
Navigation Success Rate100
6
Navigation and PlacementTask 2 Environment Left Configuration
Navigation Success Rate60
6
Navigation and PlacementTask 2 Environment Right Configuration
Navigation Success Rate60
6
Precise Drawer ManipulationTask 1 Environment Left Configuration
Navigation Success Rate20
6
Precise Drawer ManipulationTask 1 Environment Center Configuration
Navigation Success Rate40
6
Precise Drawer ManipulationTask 1 Environment Right Configuration
Navigation Success Rate20
6
Showing 6 of 6 rows

Other info

Follow for update