Max Robotics — Global Service Robot Directory + US Certification

About

Mina Marmpena is a researcher specializing in social robotics, human-robot interaction, and affective computing, with a particular focus on enabling humanoid robots to communicate emotions through body language. Her work sits at the intersection of deep learning and robotics, addressing one of the most compelling challenges in the field: making robots feel genuinely expressive and socially engaging. Marmpena's most influential contribution, "How does the robot feel? Perception of valence and arousal in emotional body language" (2018, 30 citations), established a foundational framework for understanding how humans perceive emotional signals in robotic movement, using the Pepper robot as a testbed. Building on this, she pioneered the use of generative deep learning models — specifically Variational Autoencoders (VAEs) and Conditional VAEs — to automatically generate diverse, nuanced emotional body language, moving the field beyond rigid, hand-coded animations toward data-driven approaches capable of producing natural variation over sustained interactions. Her 2022 paper further demonstrated that data-driven methods can make robots appear more trustworthy and socially present. With over 60 cumulative citations, Marmpena's research has meaningfully advanced the scientific understanding of robotic affective expression, offering practical tools for deploying emotionally intelligent robots in real-world social environments.