We consider a situational awareness as the perception of environmental elements and events, the comprehension of their meaning, and the projection of their status after some variable has changed. Applied to perspective autonomous unmanned aerial vehicles (AUAV) the situational awareness is understood as awareness of terrestrial scene structure and its variability in relation to information about this structure preassigned in the flight task (FT) which preparation is performed on the ground. In this case, ground system of FT preparation and onboard control system could be represented as two components: conventional and intellectual. The intellectual component of a control system is based on: - "intellectual" component of ground system of FT preparation that includes aprioristic knowledge base (informational, situational and algorithmic components); this knowledge base is formed on the base of aprioristic information about a condition of the external environment, a set of the predicted situations and their resolution methods; - "intellectual" component of the onboard control system (technical vision system) providing a posteriori information about a current state of external environment, comparison of a priori and a posteriori information, decision-making methods, formation and implementation of the relevant control. When developing situational and algorithmic components of the aprioristic knowledge base the following logical sequence is used: - FT is represented through a priori formed semantic networks of the typal situations (TS); one of examples of TS is process of targeting the relevant priority object of interest (OI); - each TS is formed of big number of significant events (SE); these SE forms sets (situational vectors Vs), that characterizes the problem sub-situation (PrSS). "Intellectual" component of the onboard control system (technical vision system) performs such tasks as recognition, selection and auto tracking of the priority OI assigned during the preparation of the FT. In a real situation, the priority OI could be unavailable (destroyed, invisible because of weather condition, etc.). In this case, the on-board system decides to retarget to an alternative OI. The example is offered where a method is given for the developing of the situational and algorithmic components of the a priori knowledge base for the case of two OI that are part of a multi-object stationary terrestrial scene.