Learning to Compose Hierarchical Object-Centric Controllers for Robotic Manipulation
By Mohit Sharma, Jacky Liang, Jialiang Zhao, Alex LaGrassa, And, Oliver Kroemer. 2020
32 citations recorded by the source corpus.
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.
Subfields: Artificial Intelligence, Control and Systems Engineering
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Silvio Savarese
- Li Fei-Fei
- Danfei Xu
- Ajay Mandlekar
- Chen Wang
- Rui Wang
- Abhinav Kumar Gupta
- Sergey Levine
In Control and Systems Engineering
- Li Fei-Fei
- Silvio Savarese
- Ajay Mandlekar
- Danfei Xu
- Rui Wang
- Chen Wang
- Yuke Zhu
- Roberto Mart'in-Mart'in
Citations in the corpus
- Generalization Through Hand-Eye Coordination: An Action Space for Learning Spatially-Invariant Visuomotor Control (2021, methodology): “However, these methods often rely on knowledge of groundtruth object poses [25, 32] while our method generates suitable action targets directly from raw vision input without explicit supervision.”