T. Padır

This work introduces Universal Navigation Interface (UNI), a robot-free data collection paradigm that uses a four-wheeled rollator walker and smartphone to collect physically constrained human demonstrations and demonstrates closed-loop transfer to a powered wheelchair in curb, staircase, and curb-cut scenarios.

A robot can autonomously navigate emergency evacuations, achieving a 92.4% success rate in facilitating door traversal and equipment delivery.

Finger-grazing and palm-rolling aren't just fancy robot moves—they slash the number of tactile interactions needed for accurate 3D shape reconstruction by over a third, while boosting accuracy by more than half.

Achieve faster and more reliable robot motion planning by reactively switching between simplified models within a Nonlinear Model Predictive Control framework, guided by deep reinforcement learning.