Teaching

Worcester Polytechnic Institute

RBE 595: Optimal Control and Reinforcement Learning (Instructor, 2026). A unified treatment of optimal control and reinforcement learning as two perspectives on the same problem, sequential decision-making. The course moves from convex optimization, dynamic programming, LQR, the Hamilton-Jacobi-Bellman equation, and model predictive control to policy gradients, PPO, SAC, model-based and offline reinforcement learning, and sim-to-real transfer on physical robots.

RBE 595: Optimal Control (Instructor, 2025). Foundations of optimal control for robotics: convex optimization and quadratic programming, the Linear Quadratic Regulator and Model Predictive Control, nonlinear MPC for trajectory planning and control, and learning-based approaches that combine data-driven models with optimal control. Assignments use real or simulated robotic platforms.

RBE 502: Robot Control (Instructor, 2024 to 2026). Linearization and state-space modeling, feedback control, Lyapunov stability, nonlinear control, trajectory and motion planning, compliance and force control, impedance, adaptive, and robust control, with hands-on projects on robotic manipulators and mobile robots.

Earlier

  • Guest Lecturer, ROB 6003 Foundations of Robotics, New York University (2019 to 2022). Lectures on robot-arm dynamics with the Lagrange and Newton-Euler approaches.
  • Teaching Assistant, MEAM 510 Design of Mechatronic Systems, University of Pennsylvania (2017).
  • Graduate Teaching Assistant, edX Robotics: Dynamics and Control, University of Pennsylvania (2017).