Page 18 - Autonomous Mobile Robots

I

Sensors and Sensor Fusion

Mobile robots participate in meaningful and intelligent interactions with other
entities — inanimate objects, human users, or other robots — through sensing
and perception. Sensing capabilities are tightly linked to the ability to perceive,
without which sensor data will only be a collection of meaningless figures.
Sensors are crucial to the operation of autonomous mobile robots in unknown
and dynamic environments where it is impossible to have complete a priori
information that can be given to the robots before operation.
In biological systems, visual sensing offers a rich source of information to
individuals, which in turn use such information for navigation, deliberation,
and planning. The same may be said of autonomous mobile robotic systems,
where vision has become a standard sensory tool on robots. This is especially
so with the advancement of image processing techniques, which facilitates the
extraction of even more useful information from images captured from mounted
still or moving cameras. The first chapter of this part therefore, focuses on
the use of visual sensors for guidance and navigation of unmanned vehicles.
This chapter starts with an analysis of the various requirements that the use of
unmanned vehicles poses to the visual guidance equipment. This is followed by
an analysis of the characteristics and limitations of visual perception hardware,
providing readers with an understanding of the physical constraints that must be
considered in the design of guidance systems. Various techniques currently in
use for road and vehicle following, and for obstacle detection are then reviewed.
With the wealth of information afforded by various visual sensors, sensor fusion
techniques play an important role in exploiting the available information to

1

© 2006 by Taylor & Francis Group, LLC

FRANKL: “dk6033_c001” — 2006/3/31 — 16:42 — page1—#1