In this paper we give an overview of our work on a self-pace asynchronous BCI that responds every 0.5 seconds. A statistical Gaussian classifier tries to recognize three different mental tasks; it may also respond unknown for uncertain samples as the classifier has incorporated statistical rejection criteria. We report our experience with different subjects. We also describe three brain-actuated applications we have developed: a virtual keyboard, a brain game, and a mobile robot (emulating a motorized wheelchair). Finally, we discuss current research directions we are pursuing in order to improve the performance and robustness of our BCI system, especially for real-time control of brain-actuated robots.