Control System design for Drone Autonomy

Currently working as a Research Assistant in the Controls team in the Computational Engineering and Robotics Laboratory (CERLAB) under Prof.Kenji Shimada. 

Current work is focused on augmenting Reinforcement Learning with traditional Control methodologies to get robust stability and performance from Quadrotor Drones and Underwater Vehicles.

Built a wind estimation model within the simulation environment and used this to train multiple RL agents. Experimented with DDPG, Soft Actor Critic and PPO

Also worked on the system integration and testing of the underwater drone: BlueROV and conducted CFD simulation studies to investigate drag and lift coefficients to build a simulation model.