Description

This thesis presents an attention-based deep learning model for human motion prediction in collaborative robot operations, focused on handover tasks. Using handover phases & the robot's End Effector position, it predicts future human poses, enhancing adaptability & safety in dynamic human-robot interactions. Trained & evaluated on data of human volunteers & an anthropomorphic robot, results demonstrate attention mechanisms & phase-awareness improve motion prediction in collaborative systems.

The work is under the scope of the following projects:

• SOCIAL PIA: Cooperative Social PIA model for Cybernetics Avatars (Moonshot Research and Development Program) (web)