Search papers, labs, and topics across Lattice.
9
2
6
4
Transforming kinematically feasible motion references into dynamically accurate trajectories could revolutionize how robots learn complex contact-rich behaviors.
A humanoid robot can dodge 95% of thrown balls using only depth data from a head-mounted camera, showcasing the potential of perception-aware safety in dynamic environments.
A humanoid robot can learn to hit a backhand in under an hour using just a single demonstration, revolutionizing the efficiency of robotic skill acquisition.
MPC-RL achieves superior humanoid locomotion and manipulation performance by integrating efficient MPC guidance, challenging the traditional RL training paradigms.
Domain randomization doesn't just make your robot policies more robust; it fundamentally warps the optimization landscape, potentially guiding your search towards better contact-rich behaviors.
Guaranteeing stability for complex robot locomotion just got easier: HALO learns low-dimensional models that accurately predict stability regions in the full state space.
Real-world flight tests show control barrier functions can effectively constrain a human pilot's inputs on an F-16, enforcing safety limits without overly restricting maneuverability.
Legged robots can now navigate complex environments with context-aware safety margins, thanks to a new framework that understands object semantics and adjusts safety protocols accordingly.
Humanoid robots can now walk more robustly on uneven terrain thanks to a hierarchical MPC approach that cleverly incorporates arm and torso dynamics for improved stability.
