Increasing user acceptance by augmented robot intelligence: The lesson we got from the semantics of human communication

2^nd International Conference on Robotics and Artificial Intelligence

May 23-24, 2019 | Vienna, Austria

Eleni Efthimiou

Institute for Language and Speech Processing, Greece

Keynote: JET

Abstract

Research on assistive robots has received special focus within the domain of robotics and is continuously gaining ground, also boosted by demographic data and related AAL supportive policies worldwide. Having in mind devices which need to address real user needs and be capable of interacting with users in some sort of “human” like manner, it has become mandatory to find robust ways for augmenting robot intelligence in order to enable devices overcome basic interaction shortages which are easily spotted during validation by end user populations.

One predominant parameter for user acceptance is proven to be satisfaction of the human need for communication with an “intelligent” companion or assistant, if a device has to gain user trust and be systematically used within a specific mid- to long-term time frame. In this context, we exploit the paradigm of exposure of assistive devices in real use conditions, to discuss the degree of user acceptance and the need to augment robot intelligence in the context of multimodal HRI. Focus is placed on those NLP tools and resources which may increase the span of human-robot communication by engaging standard NLP approaches in combination with signals of human embodied expression which can lead to enhanced performance of robotic devices when they interact with humans.

Biography

Eleni Efthimiou is Research Director at the Institute for Language and Speech Processing /ATHENA RC, where she heads the Embodied Interaction and Robotics Group, focusing on multimodal human communication, assistive interfaces and multimodal Human-Robot Interaction. In 1986, she received her Ph.D. degree in Generative Linguistics from the University of Salzburg. Her main research interests focus on natural language processing, sign language (SL) technologies and optimization of HCI/HRI. From 1995 to date, she has designed and developed various interaction environments with emphasis on SL based interfaces, while she has also developed methodologies for human multimodal data collection. In 2005, she founded the Sign Language Technologies Team at ILSP, a group of excellence activating in wide scale SL resources and technologies with emphasis on synthetic signing, machine translation and information retrieval from SL video. She is an editorial board member of Universal Access in the Information Society (UAIS) Journal.

E-mail: eleni_e@ilsp.gr

• PDF
• HTML