Colloquia

Summary

An exoskeleton can have a massive impact on the user’s life by giving them the ability to stand and walk. One of the main limitations of the exoskeleton of the University of Twente is that standing balance, stability, and safety cannot be guaranteed. Therefore, the user is forced to walk with crutches at all times. Research at the University of Twente focuses on model-based control methodologies to self-balance the exoskeleton and user. The inverse kinematics of the exoskeleton and user should be known with a low uncertainty level to build a model-based balance controller. The ultimate aim is to learn or identify an accurate mechanical model of the system that is suitable for computed torque control and state estimation. The exoskeleton can be modeled as an underactuated floating rigid multibody system made up of a floating body and two legs. For the momentum-based balance controller, the location of the Center of Mass (CoM) of the exoskeleton and human must be known, and the location’s first derivative.

The aim is to investigate which inertial parameter sets are needed for describing the location of the CoM and to examine how these parameter sets can be estimated. The work gives a general expression for the location of the CoM of a forked fixed base system with revolute joints and an arbitrary number of bodies. In addition, the Equations of Motion of the bodies are derived in a linear regression matrix formalism by starting with the Newton-Euler Equations for a planar, forked, fixed base system with revolute joints. The matrix formalism exists of a regression matrix post-multiplied with a parameter vector that can be estimated with a Least-square approach where the parameter vector exists of base parameters; base parameters contain dynamic parameters whose values uniquely determine the dynamic model. Furthermore, proof is given of the non-rank deficiency of the regression matrix, and the general shape and sparsity are examined.

This study shows that the location of the CoM can be estimated based on the identifiable parameter sets. Exploring the identifiable parameter sets for a planar, forked, fixed base system with revolute joints and an arbitrary number of bodies provides the first step towards a general matrix formulation for a spatial floating base system with forking and an arbitrary number of bodies, so that it can be used for the exoskeleton and human system.