Tzu-Yuan Lin
Tzu-Yuan Lin is a postdoctoral associate in the Biomimetic Robotics Lab at the Massachusetts Institute of Technology, advised by Professor Sangbae Kim. He earned his Ph.D. in Robotics from the University of Michigan under the supervision of Professor Maani Ghaffari, where his research focused on the intersection of machine learning and robot perception. He is best known for developing proprioceptive invariant robot state estimation methods for legged robots, including the DRIFT open-source InEKF library and learning-based contact estimation techniques.
“Lin et al. proposed augmenting the state-of-the-art invariant extended Kalman filter (InEKF) with learned contact detection. However, this approach requires labeled contact data for training and still treats contact as a binary state.”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.