Learning to Compose Hierarchical Object-Centric Controllers for Robotic Manipulation

Autor:	Sharma, Mohit, Liang, Jacky, Zhao, Jialiang, LaGrassa, Alex, Kroemer, Oliver
Rok vydání:	2020
Předmět:	Computer Science - Robotics Computer Science - Artificial Intelligence Computer Science - Machine Learning
Druh dokumentu:	Working Paper
Popis:	Manipulation tasks can often be decomposed into multiple subtasks performed in parallel, e.g., sliding an object to a goal pose while maintaining contact with a table. Individual subtasks can be achieved by task-axis controllers defined relative to the objects being manipulated, and a set of object-centric controllers can be combined in an hierarchy. In prior works, such combinations are defined manually or learned from demonstrations. By contrast, we propose using reinforcement learning to dynamically compose hierarchical object-centric controllers for manipulation tasks. Experiments in both simulation and real world show how the proposed approach leads to improved sample efficiency, zero-shot generalization to novel test environments, and simulation-to-reality transfer without fine-tuning. Comment: Accepted as Plenary Talk at CoRL'20. First two authors contributed equally. For results see https://sites.google.com/view/compositional-object-control/
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2011.04627 Zobrazit plný text záznamu View this record from Arxiv