Task-Directed Exploration in Continuous POMDPs for Robotic Manipulation of Articulated Objects.
Aidan CurtisLeslie Pack KaelblingSiddarth JainPublished in: CoRR (2022)
Keyphrases
- articulated objects
- degrees of freedom
- manipulation tasks
- human body
- motion analysis
- articulated motion
- reinforcement learning
- shape model
- single view
- visual hull
- human motion
- partially observable markov decision processes
- rigid objects
- image sequences
- mobile robot
- object recognition
- multi view
- computer vision
- rigid parts