Real Time Visual Servoing Using Raspberry Pi

Bhumika Jain, Vipra Bohara

Video processing is becoming widely used technique in robotics & automation systems to mimic human like sensing capabilities. Video processing of powerful platform has demonstrated widespread possibilities such as face detection around ball trajectory mapping games especially cricket robotics racer industrial vision-based inspection system. Video processing of powerful platform such as Intel & AMD processor has been widely researched & applied in main stream technique however there is very limited research on video processing on single board computers due to them constrains of processing power & memory. This work is aimed at development of a real time processing system based on raspberry pi single board computer which is a low-cost credit card size single board computer. The proposed work aims to demonstrated real time hardware control along with real time video processing to enfaces on the embedded nature of the single board computers & their application proximity to hardware control realm. Thus, a real time visual system is developed using raspberry pi SBC a web camera & opto isolated DC motor driver circuit. Vision processing is carried out by employing open CV (Image and Video Processing) in python.