UTCS Colloquium/AI: Paul Newman/Oxford University Appearance Based Navigation and the FAB-MAP Algorithm ACES 2.402 Tuesday April 1 2008 2:30 p.m.

Contact Name: 
Jenna Whitney
Date: 
Apr 1, 2008 2:30pm - 3:30pm

There is a signup schedule for this event (UT EID required).

Typ

e of Talk: UTCS Colloquium/AI

Speaker Name/Affiliation: Paul Newma

n/Oxford University

Date/Time: Tuesday April 1 2008 2:30 p.m.

Location: ACES 2.402

Host: Ben Kuipers

Talk Title:

Appearance Based Navigation and the FAB-MAP Algorithm

Talk Abstract:

This talk considers an appearance-based topological approach
to mob

ile robotic navigation and mapping. We shall introduce a
new algorithm

- Fast Appearance Based Mapping (FAB-MAP) -
which is capable of buildin

g large scale (>>km) topological maps
and detecting loop closure with a
cost linear in the size of the map.
Loop closing is the problem of cor

rectly asserting that a robot has
returned to a previously visited area

. It is a particularly hard but
important component of the Simultaneous
Localization and
Mapping (SLAM) problem. Here a mobile robot explores

an
a-priori unknown environment performing on-the-fly mapping
while
the map is used to localize the vehicle. Many SLAM
implementations loo

k to internal map and vehicle estimates to
make decisions about whether
a vehicle is revisiting a previously
mapped area or is exploring a new
region of workspace. We
suggest that one of the reasons loop closing i

s hard in SLAM
is precisely because these internal estimates can despi

te best
efforts be in gross error. FAB-MAP makes no recourse to the metric estimates of the SLAM system it supports and aids—it is

entirely independent. We illustrate the effectiveness of the
algorithm

on several outdoor and indoor data sets producing
both purely topologic

al maps and by integrating topological
constraints within large scale m

etric SLAM maps built with 3D
laser data.

Speaker Bio:
Paul N

ewman obtained an M.Eng. in Engineering Science
from Oxford University

in 1995. He then undertook a Ph.D.
in autonomous navigation at the Aust

ralian Center for Field
Robotics University of Sydney Australia. In 1

999 he returned
to the United Kingdom to work in the commercial sub-sea

navigation industry. In late 2000 he joined the Dept of Ocean
Engi

neering at M.I.T. where as a post-doc and later a research
scientist h

e worked on algorithms and software for robust
autonomous navigation fo

r both land and sub-sea agents. In
early 2003 he returned to Oxford as

a Departmental Lecturer
in Engineering Science before being appointed t

o a University
Lectureship in Information Engineering and becoming a Fe

llow
of New College in 2005. He heads the Oxford Mobile robotics
Re

search group and has research interests in pretty much
anything to do w

ith autonomous navigation but particularly
Simultaneous Localisation an

d Mapping. He is on the editorial
board of the International Journal of
Robotics Research and
The Journal of Field Robotics and a IEEE R.A.S E

uropean
Distinguished Lecturer for 2008 and 2009.