Modern rapid developments of robotic and automated systems created novel operating environments for machinery and industrial process operators in which the reduction of the level of control exercised by operators can lead to their losing attention during important machinery or process operation tasks. Loss of attention is currently one of the most important causes of work and traffic related accidents. The problem of detecting the loss of operator's attention attracted substantial attention in recent years. In this work, an EEG and SVM-based Brain-Computer Interface system was developed for determining an operator's attention state. Using a virtual continuous attention vehicle control task, the ability of the system to detect different operator attention states with high degree of reliability was demonstrated.