A Human Robot Interaction Toolkit with Heterogeneous Multilevel Multimodal Mixing

Vijver, B. van de (2016) A Human Robot Interaction Toolkit with Heterogeneous Multilevel Multimodal Mixing.

[img]
Preview
PDF
3MB
Abstract:In the modern world robots are being used more and more for everyday tasks, but mostly they are being used for industrial purposes. A relatively new application field is autonomous conversation, which requires robots to have social interaction with humans. For social interaction it is important to have fluent and lifelike robot behavior, without the need of explicit constant control. This requires us to no longer see the robot as a puppet, that always needs its puppeteer, but to see it as an actor, which has a director that asks for certain behavior. During the presentation Heterogeneous Multilevel Multimodal Mixing (HMMM) will be introduced, which mixes (external) behavior requests (for example from an external director and the internal autonomous behavior) into the viable behavior. The resulting behavior needs to be fluent and lifelike in order to create good social interaction, but it also needs to be without dead time and it needs to react on the environment. The result is a re-usable human robot interaction toolkit which has the capability to accept behavior requests from other software, resulting in fluent and lifelike behavior. It does not need constant control and it can autonomously break any requested behavior when necessary. The resulting robot seems to have and positive effect on others and it can be used for social interaction with humans.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:50 technical science in general, 54 computer science
Programme:Embedded Systems MSc (60331)
Link to this item:http://purl.utwente.nl/essays/71171
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page