Implementation and performance evaluation of optical flow navigation system under specific conditions for a flying robot

摘要

Honeybees land on a flower [1], [2], estimate traveled distance [3], and avoid obstacles [4] merely based on optical flow (OF) reckonings. Like OF reckoning, OF navigation used in flying robots uses biological and computer vision activities to estimate distance and avoid obstacles. Currently, most flying robots use Global Navigation Satellite System (GNSS)/inertial navigation system (INS) integration for navigation/positioning purposes. Navigation is defined as estimation/determination of the position, velocity, and orientation of a vehicle. If the position alone is considered, it is called positioning/ localization data and is one input for guidance and control systems in a flight computer [5]. Accuracy and safety during an autonomous flight rely heavily on the positioning data. However, under indoor and in some outdoor environments similar to dense forests and caves, GNSS signals are not accessible or dependable [6]. In environments like this, without any supplemental sensing devices, flying robots are routinely incapable of sensing and collision avoidance. For this reason, the use of such systems is not appropriate for indoor and outdoor spaces where GNSS signals are not reliable/available.