My name is Rishi Ramakrishnan and I am a Postdoctoral Research Associate at the Australian Centre for Field Robotics at the University of Sydney. 

I graduated from the University of Sydney in 2011 with a Bachelor of Engineering (Mechatronic) Honours 1 and a Bachelor of Science, with a double major in Computer Science and Applied Mathematics. I completed my PhD in April 2016 at the Australian Centre for Field Robotics in the Rio Tinto Centre for Mine Automation under the supervision of Dr Juan Nieto and Dr Steve Scheding.

My PhD was focused on achieving illumination invariance in visual data in remote sensing and robotics applications. The aim was to improve the performance of classification, localisation and clustering algorithms by reducing the detrimental impact of occlusions and geometry in outdoor environments. I utilised a multi-modal sensor system to provide automatic field based calibration of visual data, which has the advantage over current systems in that it does not require external hardware or highly parameterised atmospheric modellers.

Currently, I work on classification of objects in urban environments using Convolutional Neural Network approaches. I am investigating how to incorporate 3D detection characteristics into the framework and how we can train the network to reduce catastrophic classification failures (eg. classifying people as cars).