Framework
Framework of cognition layer. What are the tasks of cognition layer? How decision making works?
Conceptual Framework
Gathering information in an uncontrolled field with autonomy requires a powerful and robust decision making architecture. Autonomous vehicle must have metrics for awareness. It should be able to track data of surroundings, position, target, events of environment and self condition with its awareness metrics.
Cognition Layer
Cognition layer contains tasks which are responsible from decision making and taking real time actions. To be able to fulfill cognition layer tasks, management of drone's control unit, chain of actions depending to output of vision layer and tracking events in environment, e.g., collision or mechanical issues, are needed. This study focuses to cognitive decision process, in other words cognition layer of autonomous environmental monitoring system.
Monitoring tasks can be divided to levels in regards of tracking. In first level of monitoring, definition of environment and recognition of targets should be done. Recognition is subjective in a manner of quality of inspection, e.g., image observation may be effected from angle of camera or daylight. In next level of monitoring, observer should be certain to not overlook any part or target of field. In respect to levels of monitoring a cognition layer for autonomous monitoring system is designed to act in three decision states; path tracking, target acquisition and emergency. To handle each decision state cognition layer is separated to three modules; coverage module, investigation module and safety module. While each module has separated action and decision logic, each module is in effect of other modules for their actions-decisions. Design of modules gives a flexibility of usage for different monitoring applications.
Last updated