I am interested in understanding principles of movement generation. Given the continuous stream of movements that biological systems exhibit and their enormous capacity to learn new skills, an account for such versatility and creativity has to assume that movement sequences consist of segments, executed either in sequence or with partial or complete overlap.
My research revolves around the existence and formalization of discrete and rhythmic movement primitives in human arm control. We study this hypothesis using psychophysical experiments and Dynamics Systems modeling
In psychophysical experiments we use Sensuit or Visual Tracking system to record joint angle and endpoint position while performing rhythmic and discrete tasks. One such experiment involves interaction of discrete and rhythmic movements in single joint, dual joints of the same arm and bimanual tasks.
In another set of experiments we investigate online movement correction using target switching paradigm. We have a 3-D visual display using anaglyph glasses.
In Dynamics System modeling, we study Dynamic Motor Primitives (DMP). DMPs are units of action that are formalized as stable nonlinear attractor systems. They are useful framework for biological motor control as they are highly flexible in creating complex rhythmic (limit cycle) and discrete (point attractor) behaviors that can quickly be adapted to the inevitable perturbations of a dynamically changing, stochastic environment.
We use our Dynamic System framework to model and interpret our human behavioral studies.
Motor Control, Robotics, Machine learning
(213) 740 6717