High sensitivity GPS receivers have extended the use of GNSS navigation to environments which were previously deemed unsuitable for satellite signal reception. Under adverse conditions the signals become attenuated and reflected. High sensitivity receivers achieve signal reception by using a large number of correlators and an extended integration time. Processing the observation data in dynamic and rapidly changing conditions requires a careful and consistent treatment. Code-based autonomous solutions can cause major errors in the estimated position, due primarily to multipath effects. A custom procedure of autonomous GPS positioning has been developed, boosting the positioning performance through appropriate processing of code and Doppler observations. Besides the common positioning procedures, robust estimation methods have been used to minimise the effects of gross observation errors. In normal conditions, differential GNSS yields good results, however, under adverse conditions, it fails to improve significantly the receiver's position. Therefore, a so-called conditional DGPS has been developed which determines the position differentially by using data from the strong signals only. These custom-developed procedures have been tested in different conditions in static and kinematic cases and the results have been compared to those processed by the receiver.