Show simple item record

dc.contributor.authorBrown, Peter
dc.date.accessioned2017-02-15 17:42:29 (GMT)
dc.date.available2017-02-15 17:42:29 (GMT)
dc.date.issued2017-02-15
dc.date.submitted2017
dc.identifier.urihttp://hdl.handle.net/10012/11315
dc.description.abstractMultibody forward dynamics models of the human body are often used in predictive simulations of human motion. An important component of these models is contact modelling. For example, foot-ground contact plays a crucial role in obtaining accurate results from a walking or running simulation and contact models of joints are necessary to determine accurate joint pressures. Contact models increase multibody system equation complexity (often dramatically) and can introduce nonlinearities and discontinuities into the system equations. This is particularly problematic in predictive simulations, which may determine optimal performance by running a model simulation thousands of times. A desirable contact model should be accurate enough to recreate physiological motion and contact pressures yet still efficient enough to use in an optimisation. A suitable contact model for multibody biomechanics is volumetric contact modelling. Volumetric contact modelling is ideally suited for large, conforming contacts, as is found in biomechanic applications, and has relatively simple, analytical equations (provided the contact surfaces can be approximated as simplified shapes). Another advantage is that volumetric contact can be used to calculate contact pressure, which is difficult to do with simpler point-contact models. In this thesis, volumetric contact was used in two biomechanics models to test its applicability: an anatomical knee model with tibiofemoral contact and a foot-ground contact model. The volumetric knee model was based on another knee model in the literature, with the contact model replaced with volumetric contact. The volumetric model ran faster than real-time and had similar contact forces to the original model. Further improvements are possible by using medical images to determine the contact geometry and including muscles in the model. A friction model is an important part of some biomechanic contact models, particularly the foot-ground contact model. A literature review revealed that many current friction models introduce discontinuities into system equations or are unnecessarily complex. A novel continuous friction model was developed which uses a minimum number of parameters for easy parametrisation. A novel, three-dimensional foot-ground contact model was developed and validated, for future use in a human gait simulation. The foot model used volumetric contact equations for ellipsoidal geometry (which were derived in this thesis, as an improvement on previous sphere-plane contact models). A gait experiment was used to parametrise and validate the model (except for the friction parameters). The model ran over 100 times faster than real-time (in an inverse simulation) and matched experimental normal force and centre of pressure location (with less than 7% root-mean-square error). It was discovered that the designed gait experiment could not be used to determine the friction parameters for the foot-ground model. A possible alternative was suggested, and the validation of the friction portion of the model was left to a future study. In conclusion, volumetric contact can be used to produce a computationally efficient and accurate contact model.en
dc.language.isoenen
dc.publisherUniversity of Waterlooen
dc.subjectmultibody dynamicsen
dc.subjectbiomechanicsen
dc.subjectcontact modellingen
dc.subjectfriction modellingen
dc.titleContact Modelling for Forward Dynamics of Human Motionen
dc.typeMaster Thesisen
dc.pendingfalse
uws-etd.degree.departmentSystems Design Engineeringen
uws-etd.degree.disciplineSystem Design Engineeringen
uws-etd.degree.grantorUniversity of Waterlooen
uws-etd.degreeMaster of Applied Scienceen
uws.contributor.advisorMcPhee, John
uws.contributor.affiliation1Faculty of Engineeringen
uws.published.cityWaterlooen
uws.published.countryCanadaen
uws.published.provinceOntarioen
uws.typeOfResourceTexten
uws.peerReviewStatusUnrevieweden
uws.scholarLevelGraduateen


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record


UWSpace

University of Waterloo Library
200 University Avenue West
Waterloo, Ontario, Canada N2L 3G1
519 888 4883

All items in UWSpace are protected by copyright, with all rights reserved.

DSpace software

Service outages