Jan Ondras

University of Southern California, University of Cambridge, Cornell University

Papers

Total Citations

H-Index

About

Jan Ondras is a researcher at the intersection of human-robot interaction, social signal processing, and robotic manipulation. His work focuses on enabling robots to understand and replicate human social behaviors, from intimate self-disclosure to nonverbal communication. In his highly cited 2019 paper (29 citations), Ondras demonstrated that computers can elicit and analyze intimate self-disclosure through multimodal verbal and nonverbal cues, with significant implications for mental health support technologies. His 2020 work (27 citations) pioneered audio-driven synthesis of robot upper-body motion, allowing robots to naturally mirror human body language during conversation. Ondras has also advanced robotic telepresence by developing systems that automatically replicate operator head movements and facial expressions on humanoid platforms. More recently, he has tackled challenging problems in deformable object manipulation for robotic dough shaping and social dining scenarios, where his bite timing prediction models enable robots to feed individuals during group meals. His research uniquely bridges social cognition and physical interaction, creating robots that can both understand human emotional states and move with human-like expressiveness.