Anthony Tzes

Agile Mode can dynamically adjust obstacle avoidance strategies, achieving impressive real-time performance in cluttered environments despite inherent limitations.

A dual stress signal can detect nearly five times more collision risks than traditional geometric hazard monitors in autonomous navigation.

Robots can now navigate cluttered spaces more effectively by learning adaptive safety margins that optimize both safety and efficiency in real-time.

Novice users can achieve impressive success rates in robot manipulation and social tasks with minimal training, thanks to an immersive, language-assisted teleoperation system.

Humanoid robots can now safely mimic your moves in real-time, thanks to a vision-based safety net that prevents them from bumping into you (or themselves).

Forget painstakingly tuning MPC controllers by hand: this method learns optimal humanoid locomotion policies by aligning MPC cost functions with high-fidelity RL data.