Search papers, labs, and topics across Lattice.
Affiliation:
5
0
4
SEED-UMI is presented, a framework in which both the human and the robot wear the same exoskeleton: joint encoders become a physically shared measurement, and wrist cameras mounted to the exoskeleton observe the same outer mechanism during both human data collection and robot policy rollouts.
Efficiently decoupling grasp synthesis from task understanding allows AdaRoboVLG to generalize across robotic hands while maintaining high performance in dynamic environments.
PRISM reveals that high-precision industrial manipulation requires a dataset that captures the complexities of contact-rich tasks, challenging the limitations of existing datasets focused on simpler operations.
Reusing cached plans can cut computational costs by over 80% without sacrificing performance, but only if you adaptively manage prediction mismatches.
Humanoid robots can now groove in sync for a full four minutes on stage, thanks to a model-based control framework that anticipates disturbances and proactively adjusts foot placement.
