This paper presents an algorithm for a mobile robot to actively explore the environment in order to achieve global localization. During the localization process interesting regions for future exploration are selected based on the detected features and on the hypotheses generated by the localization algorithm. The localization process is improved by presenting unique features to the robot's sensors. The proposed algorithm provides highly robust global localization in real world environments with very low computational effort spent in finding exploration goal points. Experimental results are given demonstrating the effectiveness of the algorithm in a number of different situations.