Interactive robot task training through dialog and demonstration
Proceedings of the ACM/IEEE international conference on Human-robot interaction
Hi-index | 0.00 |
Human-robot interaction is a growing research domain;there are many approaches to robot design, depending onthe particular aspects of interaction being focused on. Inthis paper we present an action-based framework that providesa natural means for robots to interact with humansand to learn from them. Perception and action are the essentialmeans for a robot's interaction with the environment;for successful robot performance it is thus importantto exploit this relation between a robot and its environment.Our approach links perception and actions in a unique architecturefor representing a robot's skills (behaviors). Weuse this architecture to endow the robots with the abilityto convey their intentions by acting upon their environmentand also to learn to perform complex tasks from observingand experiencing a demonstration by a human teacher.We demonstrate these concepts with a Pioneer 2DX mobilerobot, learning various tasks from a human and, whenneeded, interacting with a human to get help by conveyingits intentions through actions.