Max Robotics — Global Service Robot Directory + US Certification

About

Fei Dong’s research lies at the intersection of human-robot interaction, affective computing, and multimodal emotion recognition. His work focuses on enabling robots to perceive and respond to human emotional states through the fusion of speech, gesture, and motion data. Dong’s most cited paper (31 citations) introduces a bi-modal emotion recognition system that integrates gesture and speech information using RT middleware, achieving robust classification of four emotions through decision-level fusion with weighted cues. He further advanced the field by combining acceleration sensors with camera images for real-time gesture recognition, employing fuzzy logic to facilitate casual communication between humans and robots. Dong also pioneered the concept of “Fuzzy Atmosfield” to model communication atmosphere based on the emotional states of both humans and robots, demonstrating how weighted fusion and fuzzy logic can estimate human emotion from bimodal cues while generating robot emotions through emotional contagion. His notable achievements include applying emotion recognition to violin music for mascot robot systems, showcasing the versatility of his methods. With over 60 total citations, Dong’s contributions have laid foundational work for emotionally intelligent robotic systems that can interpret and react to human affective signals in natural, intuitive ways.