NavStik Hackerspace | Visual Odometry using Optic Flow
NavStik Hackerspace
navstik hackerspace, UAV, hackerspace
portfolio_page-template-default,single,single-portfolio_page,postid-164,ajax_fade,page_not_loaded,,qode-title-hidden,side_area_uncovered_from_content,qode-theme-ver-10.0,wpb-js-composer js-comp-ver-4.12,vc_responsive

Visual Odometry using Optic Flow

Artificial Intelligence, Computer Vision, Sensor Fusion, UAV
About This Project

Project discussed here is aimed at developing and integrating the optic flow algorithm for visual odometry and demonstrate autonomous flight in a GPS-denied/indoor environment.  Image processing and correlation is used for estimating motion of camera which is fixed with UAV. Using this estimated motion, the path followed by MAV can be traced or it could follow a pre-defined trajectory. The setup  consists of a downward facing camera and a single board computer along with NavStik. Processing is done in real-time with acceptable range of update rate. No specialized sensors were used. All processing is done on-board hence a high bandwidth communication link is not a necessity.