Research Interests
Learning-Based Control, Multi-Agent Systems, Deep Reinforcement Learning
As a research assistant at NIRO Lab, Musfiq’s primary objective is to develop a robust control algorithm using deep reinforcement learning techniques, optimal control, and control theory of continuous time dynamical systems to maximize safety in autonomous decision making.
He is also working on a project on multi-robot decision making and consensus where he intends to design a robot swarm following the concepts of algebraic graph theory, consensus in multi-agent systems and Boid’s flocking algorithm.