Michael Kuniavsky
Papers
2
Total Citations
6
H-Index
1
About
Michael Kuniavsky is a leading researcher in Human-Robot Interaction (HRI), specializing in how robots can perceive and learn from subtle, non-verbal human social cues to improve their performance and safety. His work focuses on enabling robots to detect their own errors and anticipate negative outcomes by reading the implicit reactions of people nearby—such as confusion, smirks, or giggles. In his highly cited 2023 paper, "The Bystander Affect Detection (BAD) Dataset for Failure Detection in HRI," Kuniavsky introduced a novel dataset and framework that allows robots to identify their mistakes by observing bystanders' spontaneous reactions, a key step toward more autonomous and socially aware machines. His 2024 follow-up, "“Bad Idea, Right?” Exploring Anticipatory Human Reactions for Outcome Prediction in HRI," extends this concept by showing how robots can predict impending failures from anticipatory human behaviors before an error fully occurs. Though early in his career, Kuniavsky’s work has already garnered attention for its creative approach to leveraging natural human responses, offering a practical path for robots to repair their own errors and operate more intuitively alongside people.
Research Focus
Key Achievements
Top Papers
- 1
- 2