Agility Add us as a preferred source on Google

Faces. Address labels. Proprietary IP. A humanoid working in a real facility sees all of it. Our CTO Pras Velagapudi on why protecting that data has to be built into the whole platform — not added on later. New series: Agility Foundations. Watch the full video on YouTube: https://www.youtube.com/watch?v=WpGs7DKe8QY #Agility #MadeForWork #DigitRobot

Introducing Digit 5. #MadeToWork #Agility

Every object a different size, weight, and grip. Rearranging a full room's worth of stuff is the same generalized manipulation Digit uses to take on mixed loads and changing layouts on a real warehouse floor. #agility #DigitRobot

Getting multiple Digits to move in sync is a real test of multi-robot coordination. Each one runs its own balance and whole-body control while staying timed to the others. The same thing that lets a fleet work side by side on a warehouse floor. #DigitRobot #agility

On this holiday, Digit is on grill duty. It turns out precise force control is good for more than payload handling. Happy 4th of July from all of us at Agility. #DigitRobot #humanoidrobotics #4thofjuly

Getting Digit to dance around a mouse is a real test of reactive footstep planning — the same dynamic footwork it uses to navigate clutter on a busy warehouse floor. #Agility #robotics

Turns out a jumping jack class is a solid stress test for whole-body range of motion. Every joint moving in sync, timing, velocity, balance. That's the same control Digit uses on a real warehouse floor.

Digit was wondering if coffee tastes better when you can stick a pinky out, so it swapped out end effectors for just a second. Different tasks require different tools. #robotics #digit #agility #humanoidrobot

#Digit hasn’t had to run during customer deployments yet, but those humanoid marathon videos really got it inspired. The only way to prove if simulated policies actually translate is to just do it. #robotics

Unlike statically stable #robots, a dynamically balanced robot can shift its center of mass to accommodate loads without tipping over. Getting #Digit to stand on one leg pushes the limits of our sim-to-real pipeline training methodologies – even the slightest model mismatches can lead to instability.