With only a few seconds of data, the resulting geometry and dynamics predictions are more accurate that vision-only methods, including deep shape reconstruction. Videos on the website, and details in the paper.
arxiv.org/abs/2504.18719
Michael Posa
@michaelposa.bsky.social
1.5K followers 48 following 4 posts
Asst. Prof. at the University of Pennsylvania, part of the GRASP Lab. PI of the Dynamic Autonomy and Intelligent Robotics Lab. dair.seas.upenn.edu