Robot Learning for Navigation, Interaction, and Complex Tasks using Large Language and Foundation Models

 

How can robots learn everyday tasks such as tidying up an apartment or assisting humans in their everyday domestic chores? 

In this PhD project, you can investigate different strategies for robots to learn. You will become an expert in Robot Learning and will implement learning algorithms that can learn from natural language task descriptions or human demonstrations. You will achieve this by using Large Language Models, Vision-Language Models and other large-scale foundation models. In the process, you will become fluent in PyTorch, one of the most versatile and widely-used deep learning frameworks in industry. You get the chance to work with a mobile robot that is equipped with a versatile arm, and set up impressive demos and experiments.

This project is supported by funding from the Australian Research Council and is done in collaboration between QUT and the University of Adelaide. You will be embedded in a team that spans both universities and will have the opportunity to travel and be co-supervised by experts from both unis.


Chief Investigators

Partners

Other Partners

This project is supported by funding from the Australian Research Council and is done in collaboration between QUT and the University of Adelaide. You will be embedded in a team that spans both universities and will have the opportunity to travel and be co-supervised by experts from both unis.