Learning Modulation for DS

This lecture covers the concept of learning and adaptive control for robots, focusing on the modulation of dynamical systems (DS). Topics include adapting controllers, theoretical approaches to modulation, local rotations, limit cycles, diverging systems, and the parameterization of modulations. The instructor explains how modulations can generate new properties for DS, improve stability, and reshape dynamics using trajectory data. The lecture also delves into Gaussian Process Regression (GPR) for learning modulations, external modulations, and applications like tactile exploration and obstacle avoidance. By the end, students will understand the benefits of modulating DS for reactive motion and nonlinear transformations.