Embodiment & Hardware
I develop robotic hands and tactile hardware as part of a holistic approach to robot design, where morphology, sensing, and learning are tightly coupled rather than treated as separate layers. In line with embodied and developmental robotics, this pillar focuses on how hand structure, tactile sensors, and physical embodiment shape what robots can perceive and how they can act in complex real-world environments. My goal is to enable dexterous, robust, and adaptive manipulation through hardware that is not only functional, but also informative, scalable, and suitable for deployment in industrial and assistive settings.
A central challenge is designing tactile hardware that remains robust, compact, and informative across different tasks, contact conditions, and sensing modalities. Current directions include biomimetic and vision-based tactile sensors, sensor-agnostic representations, and hand morphologies that support both perception and manipulation without over-specializing to a single benchmark. I am especially interested in how hardware design can better support transferable learning, safer physical interaction, and knowledge-intensive R&D collaborations that bridge lab prototypes with practical robotic systems.
