A vision-based navigation system in structural road environment

Abstract

Combining the knowledge of structural road and edge detecting method, a machine vision algorithm was proposed to understand the surrounding environment. With the behaviorbased path planning and intelligent preview controlling, a basic vision based navigation system was created and implemented on our autonomous robot test bed ATRV2. Experimental results showed that the proposed vision based system can recognize drivable area in realtime and drive the ATRV2 along a reasonable path.