Current Research

My research interests lie at the intersection of nonlinear control and optimization theory, with a particular focus on developing novel control solutions for complex robotic systems. The central objective of my research is to develop computationally tractable optimization-based control methodologies that unify formal control theory with advanced numerical optimization techniques and, ultimately, to realize versatile and dynamic maneuvers experimentally on a variety of robotic platforms including home assistant robots and robotic exoskeletons. 

Real Time Robotic Motion Planning

  • Hardware accelerated real-time robot motion planning on a configurable parallel computing chip

  • Community-based open source development platforms for extremely fast robot optimization and simulation

Intelligent and Adaptive Feedback Control Theory for Robotics

  • Multivariate adaptive feedback control policy design using deep reinforcement learning

  • Adaptive optimal control learning by computationally efficient trajectory optimization

  • Stochastic verification of the robustness and stability of learning-based control policies

Assistive Robotic Technology for Medical Assistive and Rehabilitation Devices

  • Enhanced mobility for the elderly and people with paraplegia using a powered lower-limb exoskeleton

  • Clinical study of a powered lower-limb exoskeleton for people with paraplegia: energy efficiency, user comfort, augmented rehabilitation, and safety

  • Decentralized feedback control design for interactive human-exoskeleton systems

Advanced Legged Locomotion

  • Networked control design for coordinated locomotion and manipulation of home assistant humanoids

  • Autonomous visual navigation of legged robots in the complex real-world environment