Magneto-Inertial Data Sensory Fusion Based on Jacobian Weighted-Left-Pseudoinverse

Abstract

This paper presents a method for sensory fusion of magneto-inertial data for estimation of joint angles of serial kinematic chain with rotational degrees-of-freedom. Method is named Magneto-Inertial tracking based on JAcobian PseudoInverse (MIJAPI). Method incorporates a kinematic model of the mechanism and the estimation relies on the inverse kinematics solution based on the Jacobian inverse utilizing the Moore-Penrose weighted left pseudoinverse of the mechanism. Jacobian matrix is used to solve an overdetermined system in a least squares approach due to available redundant measurements resulting from constraints related to attachments of magneto-inertial sensors and kinematic model.