Millimeter wave MIMO communication signals can provide accurate user localization in addition to high data rates. Previous work derives methods for position and orientation estimation assuming, however, unrealistic mmWave MIMO architectures or channel models. In this paper, we consider a band-limited frequency selective channel model and a hybrid MIMO architecture at the base station (BS) and the user equipment (UE). We develop a time domain algorithm that provides an estimation of the channel and the user position and orientation. The proposed approach is based on a compressive channel estimation stage that operates in parallel in the angular and delay domains. Experimental results show that the proposed approach outperforms previous work when operating with hybrid architectures and realistic channel realizations.