Frank Kirchner

Adapting measurement noise dynamically can boost legged robot state estimation accuracy by 25%, eliminating the need for cumbersome parameter tuning.

Aleatoric and epistemic uncertainties in gyroscope bias correction reveal critical insights into system reliability under perturbations, with epistemic uncertainty proving more effective at distinguishing operational conditions.

WOLF-VLA enables humanoid robots to understand and execute complex locomotion tasks from natural language, bridging a critical gap in robotic manipulation and locomotion.

Achieve real-time bipedal walking control by cleverly swapping high-fidelity for low-fidelity models in MPC, slashing computation without sacrificing stability.