Socially Capable Humanoid Robots - a New Control Architecture for Virtual School Teachers

SMU scientist developed a novel unified control architecture (TDC) for socially capable humanoid robots thus introducing V-robots as classroom teachers using Augmented Reality.

Digital Elite Inc. in cooperation with EASEL today announced it developed a new unified robotic control architecture, called cascaded Temporal Disc Controllers (TDC), for socially capable humanoid robots designed to overcome multiple challenges of classic robot behavior control.

EASEL Scientists at the Semmelweis Medical University have developed virtual replicas of advanced socially capable robots (FACE and Zeno - made by Texas-based Hanson Robotics) as educational companions to children. The research was carried out as part of a research project in an effort to create expressive agents for symbiotic education and learning.

Augmented Reality Robots are readily acceptable to our children as educators.

Barnabas Takacs, SMU EASEL

The unique robotic control architecture, called Cascaded Temporal Disc Controllers (CTDC) the SMU team has developed, addresses four critical aspects of difficult robot control problems, namely: 1) to avoid repetitive behaviors, 2) allowing high level controls (such as “wave”, “read”, “be happy”, etc.) to be sent to the robots independent of low-level platform specific implementations, 3) intuitive user interfaces to control emotional facial displays, and 4) built in safety zones defined in TDC parameter space to avoid self collisions.

To evaluate the robots' true social capabilities virtual models were treated and assessed as if they were real human faces and subjected to psychological validation and testing using automatic facial analysis tools based on Paul Eckman's Facial Action Coding System (FACS). Public validation included a FORBES Flow event, where a large number of people interacted with one of the virtual facial models to express their mood while none of them could detect that it was not a real human face. Key findings are being presented at international robotic science conference in Asia this month.