Predictive control for human-robot collaboration for object handling

In a human-robot collaboration scenario, human motion behavior modeling and analysis may be a new human-centered approach for designing a task, e.g., jointly handling an object. In the HiSMoT project, we are investigating human-robot motion models and control. In this task, we are looking for someone who develops a predictive controller for a human motion model in order to allow humans and robots to handle an object in collaboration.

This master thesis aims to extend the current latent space model for pick and place operation which has been developed by the team, and apply a predictive controller for controlling. The result will be compared to previous works and state-of-the-art. For further details, please contact us at the following address.