ishiguro symbiotic human-robot interaction project
#284: ERICA: A Robot Made to Look Human, with Dylan Glas
Glas discusses how ERICA was designed, the uncanny valley, the software architecture of ERICA, and some of the research studies that ERICA has been involved in. Dylan Glas is a Senior Robotics Software Architect at Futurewei Technologies, a research division of Huawei in Silicon Valley. He was previously a senior researcher in social robotics at Hiroshi Ishiguro Laboratories at ATR and a Guest Associate Professor at the Intelligent Robotics Laboratory at Osaka University. He was the chief architect for the ERICA android in the ERATO Ishiguro Symbiotic Human-Robot Interaction Project. His research interests include social human-machine interaction, ubiquitous sensing, network robot systems, teleoperation for social robots, and machine learning.
New advance with conversational robots
The new research stems from a symbiotic human-robot interaction project from the Japan Science and Technology Agency. This is based around the development of a multimodal conversation control system together with a multi-robot conversation control system. These have been designed to create a robot that possess a much higher degree of human-like presence than any comparable robot today. A secondary aim is to create a robot with a'sense of conversing'. This outcome of the project is to design a new generation of conversational robots, starting with a child-like android dubbed'ibuki'.
Erica, the humanoid robot, is chatty but still has a lot to learn
Jefferson Graham visits the Kyoto, Japan Hiroshi Ishiguro Laboratories to meet Erica, a lifelike humanoid. KYOTO, Japan --Erica greets me, asks my name and how I'm enjoying my trip to Japan. She nods her head as I speak and her eyes wander as if she's going to work the room, but she seems only to be listening to me. This eerie humanoid just happens to be a "conversational" robot companion. And the conversation only goes so far: She's reciting from a digital file that has been placed inside her as I read into a microphone from a script of questions researchers have given me to prompt her.