Incremental Semiparametric Inverse Dynamics Learning

TL;DR

This paper introduces an incremental semiparametric method combining parametric rigid body models and nonparametric kernel techniques to learn robot arm dynamics without prior mechanical knowledge.

Contribution

It proposes a novel incremental semiparametric approach that integrates parametric and nonparametric models for inverse dynamics learning.

Findings

Successfully learned the dynamics of the iCub robot arm

Demonstrated advantages of combined parametric and nonparametric methods

Validated the approach through experimental results

Abstract

This paper presents a novel approach for incremental semiparametric inverse dynamics learning. In particular, we consider the mixture of two approaches: Parametric modeling based on rigid body dynamics equations and nonparametric modeling based on incremental kernel methods, with no prior information on the mechanical properties of the system. This yields to an incremental semiparametric approach, leveraging the advantages of both the parametric and nonparametric models. We validate the proposed technique learning the dynamics of one arm of the iCub humanoid robot.