Haptic Predictive Control for Robot-Assisted Dressing

Robots are going to do many thing for us in the future, including helping people dress up. Zackory Erickson, Henry M. Clever, and their colleagues have come up with a deep haptic model predictive control for robot-assisted dressing. The predictions are made using haptic and kinematic observations from the robot’s end effector.

The robot predicts the forces a garment will apply to a person’s body. As the researchers explain:

We evaluated our approach with a PR2 robot that attempted to pull a hospital gown onto the arms of 10 human participants. With a 0.2s prediction horizon, our controller succeeded at high rates and lowered applied force while navigating the garment around a persons fist and elbow without getting caught.

[Source]