CareAvatars

We use SMPL-X models to represent a human avatar’s shape and expression, and integrate it with an actuatable musculoskeletal backbone to model kinematics and dynamics, and a XPDB based soft-tissue system to simulate deformation.

Soft-tissue Modeling

We use extended particle based dynamics (XPBD) to simulate the soft-tissues covering human body. When physical contact occurs between a robot arm and a human body, the human body will deform.

Muscle Driven Actuating System

We present a muscle driven actuating system for SMPL-X. The muscle is modeled with Hill-type Model. In these demos, each red segment represents a muscle segment.

Adaptive

Adaptive

The musculoskeletal system is adaptive to different body shapes.

Face

Our careavatars have high-fidelity faces with expressions and movable eyes. Also, we integrate a high-fidelity mouth with teeth and actuable tongues.

Expression

Expression

Our careavatars can have different facial expressions.

Mouth

Mouth

We integrate our avatars with a high-fidelity mouth which can open and close, and has tongue and teeth inside. It is essential for assistive feeding.

Tongue

Tongue

There is a movable tongue in the mouth, which is also actuable.

Eyes

Eyes

The avatars can move their eyeballs, making it possible to simulate eyegaze.