Title | ||
---|---|---|
Learning interaction protocols by mimicking understanding and reproducing human interactive behavior |
Abstract | ||
---|---|---|
Four phases system for learning human interactive behavior is proposed.The system is based on constrained motif discovery for basic action discovery.Controller generation is achieved through a piecewise-linear control generator.Models learned from interacting with multiple people can be combined.A real-world experiment is reported as a proof-of-concept. Understanding human interactive behavior is a key technology required for future robots. To achieve this goal, the robot should be able to recognize key patterns in human-human interactions. Moreover, the robot should be able to generate similar behaviors during its interaction with human partners. In this paper, an unsupervised system is proposed that allows the robot to build a generative model of the interaction protocol using interaction records. A system is evaluated in a guided navigation task and is shown to successfully learn the underlying interaction protocol. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1016/j.patrec.2014.11.010 | Pattern Recognition Letters |
Keywords | Field | DocType |
HRI,Social robotics,Guided navigation,Learning from demonstrations | Robot learning,Computer vision,Social robot,Interaction protocol,Motif (music),Artificial intelligence,Robot,Mathematics,Generative model | Journal |
Volume | Issue | ISSN |
66 | C | 0167-8655 |
Citations | PageRank | References |
4 | 0.41 | 21 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yasser F. O. Mohammad | 1 | 180 | 19.21 |
Toyoaki Nishida | 2 | 1097 | 196.19 |