Strategic navigation for robotic wheelchair based on geometrical information of augmented reality markers

This paper proposes a mobile robot navigation method based on augmented reality technique using not only AR markers’ patterns but their configurations. Mobile requires self-location to navigate toward destination. The authors propose system for wheelchairs that uses position and orientation of markers as well information connoted in the marker patterns. technology is applied estimate relative location robot. consists three elements: markers, web camera, motor-driven wheelchair, WHILL Model CR. When camera wheelchair recognizes an marker, it creates action command according both ID recognized pattern arrangement marker. Identified posture are used calculate complement rotation angle adjust planned route. have designed techniques plan path motorized nonholonomic enter narrow corridor where robots cannot easily drive by turning or circular routing. radius arc trajectory information. Experiments conducted guide with width 0.55[m] 0.7[m]. results confirm effectiveness our proposal because error final arrival point 5.7[mm] from ideal position.