Soft Bar DeformationDescription
In this project, the overall system is a simulation of a soft bar glued to handles. This includes finite elements (triangles) to model the bar itself and additional spring terms to model the interface between the bar and the handles. The displacement of the mesh elements is computed by minimizing the total energy of the static system. In addition, the mesh is controlled with two handles that define planar translations and orientations of the glued points. In this setting, the optimal control signal that brings the deformed position as close as possible to the target position is obtained using the inverse kinematics method. The energy density of a neo-Hookean triangle element is used in computing the simulation.
Chenhao Li
Reinforcement Learning for Robotics
My research interests focus on the general field of robot learning, including reinforcement learning, developmental robotics and legged intelligence.