New Robot Navigation using vanishing points

A clever sort of orientation-system for indoor-robots has been developed by Pratap Tokekar. It uses vanishing points of side-walls, ceiling and the floor, to determinate an orientation point for navigation. The robot uses a standard laptop-webcam and software, to analyse the vanishing point if the room, calculated by lines of walls, floor and ceilings. The robot also uses visual clues, like orange traffic cones, to recognize specific locations. This technology applies to typical indoor usage like in offices or restaurants.

More on this in the blog of Pratap Tokekar. There is a detailed project report available as PDF. From the report:

The first step in estimating the vanishing point is the extraction of the lines parallel in the world, within the image. Canny edge detector was used to first extract the edges in the image. Change in the illumination conditions is one of the biggest problems associated with vision-based techniques. To reduce the eff ect of this problem, histogram equalisation was performed on the grayscale image. This ensured suffcient contrast within the image, to extract the required edges using the Canny edge detector.

Well, what would happen, if you send this robot to a post-modern, deconstructivated building with ranked wall? Happy malfunction than, little robot.


Tags: indoor, research, homebrew, camera
Jan_22:2009 .020200 Comments(0)

 Related Entries


Add your Flavour to the Article