My research spans robotics, haptics, multi-modal perception, and artificial intelligence, at the intersection of computer science and engineering. There are two highly related themes in my robotics research: one is the focus on “contact sport”, i.e., the contact and interaction between a robot or a part/tool it holds and the environment, and the other is real-time adaptiveness of robots to uncertainty and uncertain changes in an environment based on perception.
Under the first theme, my work includes compliant motion planning involving complex contact scenarios for high-precision robotic assembly and manipulation in the presence of uncertainty, haptic simulation and rendering of fine manipulation involving both rigid and deformable objects, and continuum manipulation of objects using a highly flexible continuum/soft robot, such as an elephant trunk robot.
Under the second theme, my work on real-time adaptive motion planning (RAMP) addresses the problem of how to enable a high-‐degree of freedom (DOF) robot, such as a mobile manipulator, to operate in dynamic environments full of obstacles of unknown motions.
In bridging the two themes, the recent work of my group is focused on interleaving perception and manipulation to manipulate objects in an unknown environment or to model and recognize unknown objects, based on visual, depth, and/or tactile sensing, using conventional or continuum robot manipulators.
Both themes still require a lot of research, but advances are crucial to making robots useful in a wide range of applications and in unstructured, human-centered environments.