PhD Profile: Reshma Kunjumon

What led you to this topic?

I once completed an internship that included work on obstacle tracking with Camera and LiDAR data using Kalman Filter. It sparked an interest in autonomous vehicle navigation and its potential applications in unstructured environments.

Why is your research important and how will it make a difference?

My research focuses on helping an autonomous vehicle understand its environment better, thereby helping it to make better decisions when navigating autonomously. I make use of multiple sensors like Camera and LiDAR and combine their data intelligently to get more accurate information from them. This is important when navigating offroad terrains like forests or grasslands because unlike structured roads and highways, unstructured environments possess more challenging scenarios like lack of lane markings, unclear object boundaries, irregular layouts, environmental constraints such as shadows and illumination changes. Intelligently designed vehicles or robots that are capable of navigating off road terrains autonomously can be very helpful for applications such as firefighting or search and rescue in disaster-struck areas, where human access is limited or dangerous.

What inspired you to start a career in research?

I had done a small project on obstacle tracking in autonomous vehicles using sensor fusion as part of an internship. It led me to follow the works of people like Andrej Karpathy and Andreas Geiger who work in this area and really got me interested in the perception side of autonomous vehicles. Andrej Karpathy is my inspiration in this field.

Why did you choose to complete a PhD at IISRI?

When searching for research facilities in Australia working on autonomous vehicles, IISRI was one of the very few facilities with ongoing cutting-edge research and industry projects in this area. I had been following IISRI’s research updates for a while. Their work on the autonomous truck convoy seemed very fascinating at the time. When I had the opportunity to visit IISRI, I was given a tour around the facility and that was a game changer. The best part about IISRI is that the research here is not limited to papers. The resources and opportunities provided by them ensure that your research is realized on an application level.

What have been your highlights so far?

Two years into my PhD, I have realised how important it is to talk about my research and spread the word. I have gained skills in how to convey my research to a large, generalised audience without technical jargon. Deakin also provides a lot of platforms for this purpose. I was able to be a finalist at the 2023 Deakin VYT competition, and that felt like a great achievement.

What would you like to do once your PhD is completed?

I would like to start a career in Perception Engineering, hopefully in autonomous vision. I am currently looking for internship opportunities in this area.

This article was originally published in the 2024 IISRI Annual Report.