As part of my Bachelor’s Dissertation, I developed a 3D vision only Synthetic Dataset that showcases impossible and possible scenes, which an artificial agent must discriminate using the Violation-of-Expectation paradigm. The work was published in a NeurIPS Workshop. Supervised by Cheston Tan and Marcelo Ang.
Co-authored a Survey paper on Machine Learning Approaches for Modelling Intuitive Physics.
Introduced a novel unsupervised deep learning approach to automated protocol reverse engineering (APRE).
Developed enigma, a software framework API written in Python to simplify the usability and flexibility of testing and conducting APRE analysis using multiple machine learning technique.
Investigated multiple unsupervised machine learning techniques as baselines for comparison against the deep learning approach. This work was publised in IEEE SSCI 2021. Supervised by Bugsy Teo.
Developed an Autonomous Ground Vehicle (AGV) fleet controller using Robotic Operating System (ROS) C++, in the Gazebo simulation environment. Supervised by Dejanira Araiza Illan.
The AGV fleet controller allows the control of single/multiple AGV(s) transport orders, similar to a virtual fleet manager for mobile industrial robots, through an easy-to-use User Interface (UI).
Investigated the Negative Magnus Effect on the flow past a rotating cylinder at different angular velocities using Reynold-averaged Navier Stokes (RANS) models in OpenFOAM.
Studied the effect of varying the mesh motion methodologies, turbulence intensities and transitional RANS models in detecting the Negative Magnus Effect. Solo-presented the work during AIAA Scitech 2020 at Orlando, Florida. Supervised by Harish Gopalan.