Edmond Tong

I am currently pursuing a PhD in Robotics at Queensland University of Technology (QUT) under the supervision of Prof. Niko Sünderhauf and Prof. Jonathan Roberts. I completed an MS in Robotics from the University of Michigan (2024), and I hold a BS in Mechanical Engineering from Brigham Young University (2018).

My research interests lie at the intersection of computer vision, robotics, and machine learning. I am interested in advancing robotic capabilities by exploring novel visual and semantic representations derived from human video demonstrations. Specifically, my goal is to enable robots to learn complex, long-horizon manipulation policies by efficiently extracting generalized skills, object affordances, and sequential task logic from unstructured human-centric data, thereby allowing for robust policy generalization across diverse, real-world environments.

Email

Research

Utilizing large language models to associate affordances with specific parts of objects, and ground them using a VLM to generate task oriented grasp.

Utilizing large language models to associate affordances with specific parts of objects, thereby localizing them to facilitate manipulation.

Teaching