Science and Engineering Faculty
School of Electrical Engineering and Computer Science
Arren is currently a Postdoctorate Research Fellow at the CyPhyLab, Queensland University of Technology. He is working on visual place recognition and visual odometry under extreme conditional variance, e.g. during bright sunny days and stormy nights. The research will have outcomes in the development of persistent navigation for robots over the space of days, weeks and months.
Arren completed his PhD Dissertation at the CyPhyLab, Queensland University of Technology, and it is currently 'under review'.
Arren began his PhD in 2009 in the field of Intelligent and Cognitive Robotics in a project entitled "Talking with Robots: Evolving Grounded Language for Embodied Agents". Arren's PhD was titled 'Developing Grounded Representations for Robots through the Principles of Sensorimotor Coordination'. The research looked at unsupervised object discovery and affordance learning by correlating change in sensory input with motor commands.(Partially Observable) Markov Decision Processes are developed from robot experience and are used to represent the coordination between sensory and motor data. The framework is designed to incorporate processes of segmentation, recognition, attention, affordance learning and planning. Contributing to the project as a whole - the representations can be used with language development techniques among populations of robots to generate words that label and describe objects.
He graduated with a Bachelor of Engineering (Mechatronics) with first class honours from the University of Queensland in 2008. During his honours year he worked with the Commonwealth Scientific and Industrial Research Organisation (CSIRO) as a summer and honours scholar.more...