Simultaneous localisation and mapping (SLAM) robotics expert
I work as a Principal Computer Vision Engineer within the Applied Technologies and Data Vision team. My journey in the realm of computer vision began in 2013 when I completed my PhD in Robotics vision and machine learning at the Australian Center for Field Robotics, The University of Sydney.
One of my notable accomplishments includes developing a collaborative object-level SLAM (Simultaneous Localisation and Mapping) system with shape prior. This work involved integrating multiple disciplines such as SLAM, pose-graphs, optimisation, neural networks, and multi-agent systems. The system effectively fused information from cameras, inertial sensors, and lidar. Currently, I am actively involved in expanding this system to include new modalities, such as language and chat models.
SLAM is the Simultaneous Localisation and Mapping (SLAM) problem that asks if it is possible for a mobile robot to be placed at an unknown location in an unknown environment and for the robot to incrementally build a consistent map of this environment while simultaneously determining its location within this map. This is a fundamental modelling problem that forms the basis for many robotics challenges.