The world of sports is being transformed by the rise of automated systems which enable the monitoring of athletes. Such systems provide objective assessments about the performance of athletes, the loads they are exposed to or their tactical behavior. Presently, data collection, preparation and analysis are still often performed in a manual or semi-automatic manner. This process can be time-consuming and subjective, impeding accurate long-term monitoring of athletes. Streamlining this process may facilitate the data acquisition considerably and enable a transition of the subsequent analysis from subjective impressions to objective and accurate evaluations. This thesis addresses several challenges in turning the above-mentioned desiderata into practical solutions. It provides innovative approaches for improving and extending the acquisition, filtering and analysis of positional and inertial data. The contributions of this thesis have the potential to considerably facilitate activity monitoring in sports and thus provide a basis for future studies in sports science.