Warning:

This page is provided for historical and archival purposes only. While
the seminar dates are correct, we offer no guarantee of informational
accuracy or link validity. Contact information for the speakers, hosts
and seminar committee are certainly out of date.

----------------------------------------------------------------------

Title: Mobile Robots and Sensors for Aircraft Inspection

In April 1988, Aloha Airlines surprised Boeing and the rest of the world by
demonstrating that the 737 could (1) without warning lose its fuselage skin
circumferencially from floorline to floorline and longitudinally over a
distance of about eight seat spacings, and (2) be flown and landed in this
condition.  The Aloha "incident" resulted in the birth of research programs
in aging aircraft inspection and maintenance.  These have been directed
primarily at fracture mechanics, nondestructive testing instruments and
methods, and human factors studies.  At CMU we undertook two relatively
modest projects that focus on using mobile robots for nondestructive testing
instrument deployment, both teleoperated and automated.

The first project, jointly conducted by CMRI and my lab in the Robotics
Institute, resulted in the ANDI (Automated NonDestructive Inspector) robot.
This project emphasized demonstration of mobility capabilities; ANDI is
equipped with suction cups that make it mobile over most of the aircraft
skin surface.  For inspection it has an eddy current sensor, and for this
sensor's alignment and the robot's guidance it has four cameras, but
optimizing and demonstrating the sensors' data delivery capabilities has
been a secondary priority relative to actuation, control, and tether
management issues.

The second project, entirely in my lab (with a small company we spun off),
resulted in the CIMP (Crown Inspection Mobile Platform) robot.  CIMP has
certain mobility advantages over ANDI, i.e., it is wireless vs tethered by
an umbilicus, but unlike ANDI it can operate over only a fraction of the
aircraft skin surface.  CIMP was developed to demonstrate that a wireless
mobile robot could deliver **useful** inspection signals to an inspector
and/or flaw detection algorithms, i.e., signals that human inspectors would
accept as adequate for them to "sign off" on the integrity of the inspected
airplane.  Because 90% of aircraft inspection is visual, the primary sensor
modality carried by CIMP is state-of-the-art high resolution 3D-stereoscopic
video.

My talk will review the problem and how it is being approached by the aging
aircraft research community, describe the design and performance features of
ANDI and CIMP, and present data, including videotape of the two robots in
action and imagery from CIMP's inspection cameras taken recently during a
demonstration on a Boeing 747 in Northwest Airlines maintenance and
inspection hangar in Minneapolis.