Research projects A Multi-Fidelity Control Variate Approach for Policy Gradient Estimation RL, policy gradients, multi-fidelity control variates, variance reduction, sim-to-real transfer Sparse Policies in Noncooperative Games Information sparsity, regularized dynamic programming, feedback Nash equilibrium Second-Order Algorithms for Zero-Sum Nash Games Second-order algorithms that provably only converge to strict Nash equilibria Auto-Encoding Bayesian Inverse Games Multi-modal distribution inference, uncertainty, generative modelling, differentiable programming Differentiable Adaptive Game-Theoretic Planner Adaptive game-theoretic planning enabled by a differentiable game solver Other projects MPC Designing and Analysis a model predictive control approach design and analysis Numerical Optimization Algorithms Implementation of nonlinear programming algorithms Multi-Sensor Perception of Intelligent Vehicles pedestrian and obstacle detection using multi-sensor information Rapidly Learn Manipulation Tasks by Demonstration Hackathon challenge by Franka Emika at ERF2022 Learn Driving Policy from Pixels Imitation learning driving policies from image observations ScrubBot: Autonomous Field Cleaning a robotic solution for cleaning a field where obstacles present Reinforcement Learning via Corrective Feedback interactive imitation learning research Autonomous Delivery motion planning and control for an autonomous quadrotor drone Event2Video: Reproduction Project a reproduction project that maps event data to high dynamic range videos using deep neural network Quadrotor Tracking Control quadrotor controllers implementation for tracking tasks
