Dimensionality reduction and motion coordination in learning trajectories with dynamic movement primitives

Abstract

Dynamic Movement Primitives (DMP) are nowadays widely used as movement parametrization for learning trajectories, because of their linearity in the parameters, rescaling robustness and continuity. However, when learning a movement with a robot using DMP, many parameters may need to be tuned, requiring a prohibitive number of experiments/simulations to converge to a solution with a locally or globally optimal reward. We propose here strategies to palliate this dimensionality problem: the first is to explore only along the most significant directions in the parameter space, and the second is to add a reduced second set of Gaussians that would optimize the trajectory after fixing the Gaussians approximating the demonstrated movement. Both strategies result in less Gaussian computations and better performance on learning algorithms.

To further speed up the learning and allow for a better biased exploration, we also propose to coordinate the motion of different joints, by computing a coordination matrix initialized with the demonstrated movement and then automatically updating it by eliminating the degrees of freedom least affecting task performance.

Our three proposals have been experimentally tested and the obtained results show that similar (or even better) performance can be obtained at a significantly lower computational cost by reducing the dimensionality of the exploration space.