[Apologies for multiple postings]

**************************************************
*
*
*    GRAND Seminar
*
*    http://cs.gmu.edu/~robotics/Main/GrandSeminar
*
*
**************************************************


*Title*
Mobile Robot Perception for Long-term Autonomy


*Time/Venue*
CS conference room, ENGR 4201
*2PM*, October 05, Wed.

*Speaker*

Gabe Sibley
http://faculty.cs.gwu.edu/~gsibley
Assistant Professor
George Washington University


*Abstract*

If mobile robots are to become ubiquitous, we must first
solve fundamental problems in perception. Before a mobile
robot system can act intelligently, it must be given -- or
acquire -- a representation of the environment that is
useful for planning and control. Perception comes before
action, and the perception problem is one of the most
difficult we face.

An important goal in mobile robotics is the development of
perception algorithms that allow for persistent, long-term
autonomous operation in unknown situations (over weeks or
more). In our effort to achieve long-term autonomy, we have
had to solve problems of both metric and semantic
estimation. In this talk I will describe two recent and
interrelated advances in robot perception aimed at enabling
long-term autonomy.

The first is relative bundle adjustment (RBA). By using a
purely relative formulation, RBA addresses the issue of
scalability in estimating consistent world maps from vision
sensors. In stark contrast to traditional SLAM, I will show
that estimation in the relative framework is constant-time,
and crucially, remains so even during loop-closure events.
This is important because temporal and spatial scalability
are obvious prerequisites for long-term autonomy.

Building on RBA, I will then describe co-visibility based
place recognition (CoVis). CoVis is a topo-metric
representation of the world based on the RBA landmark
co-visibility graph. I will show how this representation
simplifies data association and improves the performance of
appearance based place recognition. I will introduce the
"dynamic bag-of-words" model, which is a novel form of query
expansion based on finding cliques in the co-visibility
graph. The proposed approach avoids the -- often arbitrary
-- discretization of space from the robot's trajectory that
is common to most image-based loop-closure algorithms.
Instead, I will show that reasoning on sets of co-visible
landmarks leads to a simple model that out-performs
pose-based or view-based approaches, in terms of precision
and recall.

In summary, RBA and CoVis are effective representations and
associated algorithms for metric and semantic perception,
designed to meet the scalability requirements of long-term
autonomous navigation.


*Short Bio*

Gabe Sibley is a robotics scientist and assistant professor
in Computer Science at George Washington University. He was
formerly in the University of Oxford in the Mobile Robotics
Group. He did his PhD at the University of Southern
California and at NASA-JPL, where he worked on long-range
data-fusion algorithms for planetary landing vehicles,
unmanned sea vehicles and unmanned ground vehicles. His core
interest is in probabilistic perception algorithms and
estimation theory that enable long-term autonomous operation
of mobile robotic systems, particularly in unknown
environments. He has extensive experience with vision based,
real-time localization and mapping systems, and is
interested in fundamental understanding of sufficient
statistics that can be used to represent the state of the
world. His research uses real-time, embodied robot systems
equipped with a variety of sensors -- including lasers,
cameras, inertial sensors, etc. -- to advance and validate
algorithms and knowledge representations that are useful for
enabling long-term autonomous operation.


-- 
*Jyh-Ming Lien*
Assistant Professor, George Mason University
+1-703-993-9546
http://cs.gmu.edu/~jmlien