Jhonatan Kobylarz
Papers
1
Total Citations
41
H-Index
1
About
Jhonatan Kobylarz is a researcher at the forefront of non-verbal human-robot interaction, specializing in electromyography (EMG)-based gesture recognition and transfer learning. His most cited work, "Thumbs up, thumbs down: non-verbal human-robot interaction through real-time EMG classification via inductive and supervised transductive transfer learning" (2020, 41 citations), introduces a novel method for classifying ternary gestures—specifically "thumbs up" and "thumbs down"—using the Myo armband. By employing inductive and supervised transductive transfer learning, Kobylarz addresses the challenge of adapting EMG models across different users and sessions, significantly improving real-time classification accuracy without extensive retraining. This contribution is pivotal for enabling intuitive, non-verbal communication between humans and robots, reducing the need for voice commands or physical contact. His work has implications for assistive technologies, prosthetics, and collaborative robotics, where seamless interaction is critical. Kobylarz’s research demonstrates a commitment to bridging machine learning and human-robot interfaces, offering scalable solutions that enhance adaptability and user experience. With a focus on practical, real-world applications, his studies continue to influence the development of more responsive and autonomous robotic systems.
Research Focus
Key Achievements
Top Papers
- 1