Wei-Di Chang

McGill University, Intelligent Machines (Sweden)

Papers

Total Citations

H-Index

About

Wei-Di Chang is a robotics researcher whose work sits at the intersection of autonomous underwater vehicles, computer vision, and multi-robot systems. Chang's research focuses on enabling robots to operate intelligently in unstructured, real-world environments — particularly challenging underwater settings where perception and navigation are notoriously difficult. Among Chang's most recognized contributions is a robust multi-robot convoying framework that uses visual tracking-by-detection to allow autonomous agents to follow a leading robot in complex three-dimensional underwater environments. This work, which has garnered 12 citations, addresses a fundamental challenge in coordinated robotics by combining efficient model-based object detection with temporal filtering, removing the need for structured surroundings or external localization systems. Building on this foundation, Chang has also advanced the field of intelligent robotic search, developing a one-shot learning approach that enables field robots to seek out scientifically relevant targets during environmental monitoring missions — moving beyond simple pre-programmed path-tracking toward truly adaptive navigation. This work reflects a broader commitment to making autonomous robots practical tools for real-world scientific data collection. Chang's contributions represent meaningful progress in bridging the gap between laboratory robotics and deployment in demanding natural environments.