The effectiveness of perception algorithms, and thus the ability of Robots to act autonomously in the real-world, relies heavily on the way we encode information from the 3D environment. Until today, the scientific community has not yet concluded to a single representation of the 3D domain, as is for example the case for images, where they are for the most part represented as 2D matrices of pixel values. Consequently, the encoding of geometric information remains an active field of research. Recently, Neural Fields have shown promising results as a representation of 3D space for perception tasks in the field of Robotics. The goal of this project will be to utilize the recent advancements in NFs for the purpose of improving spatial perception of mobile Robots, as this area is relatively unexplored.
I am a PhD candidate, at the School of Informatics - University of Edinburgh, sponsored by the Edinburgh Centre for Robotics, under the supervision of Prof. Chris Xiaoxuan Lu. My research objective is to improve the real-time spatial perception in robotics, utilizing the recent advancements in Neural Fields.
In November 2021, I completed with distinction the Artificial Intelligence master's programme at the University of Edinburgh. For my master's dissertation project I worked on a novel inference and learning algorithm for generative capsule models, under the supervision of Prof. Chris Williams.
In 2019, I was awarded the Diploma of Electrical and Computer Engineering, from the University of Patras, in Greece (graduated 4th in my class out of 202 students – GPA: 8.11 out of 10). I completed my ECE Diploma thesis under the supervision of Prof. Athanassios Skodras, researching the use of CNNs for real-time hand-gesture recognition via sEMG signals. Our work was presented at the IISA'19 international conference.
Moreover, working under the supervision of Prof. Konstantinos Moustakas at the VVR group, I researched the use of Haptics in self-driving vehicles, by taking part in the 2nd Student Challenge in Automotive Haptics, at the WHC'19 in Tokyo, Japan, where our team received the "Best Student Innovation Challenge Award".