Human-Robot Collaboration & Human-Object Interaction

HRC:

This project explores the generation of robot-appropriate behaviors when it works with a human in collaborative tasks. The robot is to understand human motion, model it, and understand how to best work with the human to achieve collaborative goals in smoother, easier ways for the human user.

HRC Work:

• Collaborative Human-Robot Motion Generation using LSTM-RNN
• Dynamic Robot Motion Prediction Updates in Physical Human-Robot Interactive Tasks
• Learning Human-Robot Collaboration Insights through the Integration of Muscle Activity in Interaction Motion Models.

HOI:

Another angle of this work is to help robots understand the type of interactions that humans have with objects. Such understanding is critical for robots to better serve humans.

• Visual-Semantic-Pose Graph Mixture Networks for Human-Object Interaction Detection (VSP-GMN)

Human Pose Estimation:

As well as understanding how humans move over time. This can help robots to learn to imitate humans better, estimate their real-time positions, and study their movements in relation to objects and robots.

• Graph Attention Spatio-Temporal Convolutional Nets (GAST-Nets)