This project finds its justification in the vast economic and social impact that musculoskeletal disorders have at a global level. Many cases of low back pain, which is the most relevant disorder among them, remain a mystery as far as their pathology is concerned. The assumption underlying this work is that, for those mysterious cases, the problem is located in the sacroiliac joints of the pelvic girdle in form of abnormal biomechanical properties. Unfortunately, at present there is no diagnostic means to assess those properties in an objective manner. The goal of this project is thus to develop a diagnostic tool for an objective non-invasive estimation of the biomechanical properties of the sacroiliac joints in those cases of inexplicable low back pain. The idea behind it is to perform a system identification, with the pelvis being the "system", based on dynamic testing of the pelvic structure in vivo. In order to do this an identification methodology is necessary, based on a dynamic model of the pelvis, as well as an apparatus for the detection of the dynamic response of the pelvis in vivo. In this work this methodology has been investigated, with focus on the model and the identification tools. The development of the apparatus is only briefly described, since it was the subject of a parallel PhD track.