Autonomous flight of pocket drones is challenging due to the severelimitations on on-board energy, sensing, and processing power. However, tinydrones have great potential as their small size allows maneuvering throughnarrow spaces while their small weight provides significant safety advantages.This paper presents a computationally efficient algorithm for determiningoptical flow, which can be run on an STM32F4 microprocessor (168 MHz) of a 4gram stereo-camera. The optical flow algorithm is based on edge histograms. Wepropose a matching scheme to determine local optical flow. Moreover, the methodallows for sub-pixel flow determination based on time horizon adaptation. Wedemonstrate velocity measurements in flight and use it within a velocitycontrol-loop on a pocket drone.
展开▼
