visits since 08 February, 2005
Welcome to my web site. I am a research fellow at the Queensland Brain Institute and the School of Information Technology and Electrical Engineering at the University of Queensland in Australia. My research interests are robotics, neuroscience, and animal navigation. On this page you'll find links to my research work as well as to my other activities.
Our paper "Spatial Cognition for Robots: Robot Navigation from Biological Inspiration" has just been published in IEEE Robotics and Automation Magazine (2008 Impact Factor 3.0). The abstract and front page image is provided below:
If you see a rat scuttling through your backyard, you might want to stop and consider the superiority of the rat at creating and exploiting spatial representations compared with the most advanced robot. Chances are that the rodent you see has a nest that is many hundreds of meters, possibly kilometers, from your backyard, and yet the rodent can unerringly return to its home. If your yard has some ripe seeds or fruits (and no predatory pets), the rat may return some days later, further demonstrating its ability to store and recall the spatial layout of its range. The rat runs under leaves and through drains, with few clear landmarks in a world that is under constant perceptual change in terms of appearance, texture, and odor. Clearly, the rat can build amap over large ranges in a real-world environment under constant change and use and maintain that map more than its 2-3 year lifetime. As yet, a robot cannot.
18 inch T-800 Cyberdyne Model 101 Terminator:
My paper titled "Persistent Navigation and Mapping using a Biologically Inspired SLAM System" has just been published in The International Journal of Robotics Research (2008 Impact Factor 2.8).
I've replicated the abstract below:
The challenge of persistent navigation and mapping is to develop an autonomous robot system that can simultaneously localize, map and navigate over the lifetime of the robot with little or no human intervention. Most solutions to the simultaneous localization and mapping (SLAM) problem aim to produce highly accurate maps of areas that are assumed to be static. In contrast, solutions for persistent navigation and mapping must produce reliable goal-directed navigation outcomes in an environment that is assumed to be in constant flux. We investigate the persistent navigation and mapping problem in the context of an autonomous robot that performs mock deliveries in a working office environment over a two-week period. The solution was based on the biologically inspired visual SLAM system, RatSLAM. RatSLAM performed SLAM continuously while interacting with global and local navigation systems, and a task selection module that selected between exploration, delivery, and recharging modes. The robot performed 1,143 delivery tasks to 11 different locations with only one delivery failure (from which it recovered), traveled a total distance of more than 40 km over 37 hours of active operation, and recharged autonomously a total of 23 times.
This week we had our second Thinking Systems retreat at O'Reilly's Rainforest Retreat. Was a great few days in mother nature with some spectacular views. I used the Autostitch software to create a nice huge panorama - click on the image to open up a full size one, and then make sure to hover your mouse over the new image until you get the option to click on it again to enlarge it.
IEEE GOLD Magazine has just published an article I wrote titled "The Real Robot Revolution". In it I talk about what I think society should be focusing on over the next couple of decades with respect to robots - not the extreme scenarios as presented in the Terminator and Matrix movies, but the gradual but steady entry of 'mundane' robotics into everyday society. You can view my article by clicking on the picture below, or you can view the entire magazine article here (my article is on page 12 of the March 2009 GOLDRush Issue).
My work as part of the Thinking Systems project is going to be on ABC's Catalyst television program. Session times are:
A link to a the video and transcript is available here.
Here is a video on Youtube of our suburb mapping stuff, with some introduction and brief explanation of how it all works. Click the play button to start the video.
If you have come here from the New Scientist website, I suggest you have a look at our publications.
The New Scientist article can be found here.
Paper accepted to the journal IEEE Transactions on Robotics for publication in October 2008. The paper will form part of a special issue on Visual SLAM and presents work showing that it is possible to map an entire city suburb using just a low quality web camera mounted on the back of a car. This work follows on from preliminary work in this conference paper.
You can see a photo of the experimental setup just below :)
London, Norway, and Portugal Trip
Finally got around to putting up some photos from my overseas trip, just click on the piccie below to open up the gallery.
Brisbane City Hall Hollywood New Years Ball
Went to the ball and decided to actually make an effort with a costume...which of course had to be The Terminator. Here's a couple of pics:
That's a red LED I soldered some wires to which led down into my pocket where I had two AA batteries in series. Wasn't exactly the most comfortable setup having this hot, spiky thing right next to my eyeball but it was well worth the sacrifice, as it got me a best costume award on the night :) The grenades were foam bells which I cut the bottom of the bell off, and painted gold, with a black cardboard cylinder as the body of the grenade. The ammo belt was made from felt. Leather jacket was from St Vinnies.
Australasian Conference on Robotics and Automation
Presented a paper at the conference on outdoor mapping and navigation. Nice seafood dinner at a local restaurant.
Paper accepted to the International Conference on Robotics and Automation
Found out today that my paper was accepted for ICRA 2008, held in California in 2008. 641 papers accepted out of 1476 submitted.
Trip to the Conference on Spatial Information Theory 2007
Just spent a lovely week at a conference just outside of Melbourne at Mt Eliza. You can access the picture gallery by clicking on the photo album icon below. There's also a nice panoramic photo of the bay below it.
Click on the album icon above to go the picture album.
Once you click on the picture and it takes you to the next page, hold the mouse cursor over the image and wait for the expand picture icon to appear, before clicking to see the image at full zoom
Virtual Neurogenesis in Dentate Gyrus
DO YOU WANT TO DO A PHD OR HONOURS PROJECT IN ROBOTICS?
If you are a highly motivated individual with a good track record and are interested in doing a PhD or honours project in robotics or a related topic, contact me now.
My Thesis Accepted for Publication as a Book
Just got the good news that my thesis has been accepted for publication as a book by Springer Verlag, as part of the Springer Tracts in Advanced Robotics (STAR).
A video showing how a robot uses the RatSLAM system and experience mapping algorithm to explore an large indoor environment, and then navigate to goals.
Some videos of our docking system in action:
Have moved into my own shared office in preparation for my new position under the Queensland Brain Institute and the School of Information Technology and Electrical Engineering.
Had my graduation ceremony today, click on the photo to see a few happy snaps.
Some of the final permanent binding copies of my thesis, others are already enroute to the central University Library and my department. Ran into a few other very relieved students who were handing in their final temporary or permanent binding copies. Only graduation left now.
Article in the Courier Mail about what life is like as a robotics researcher. I must say one thing: the robots rarely catch fire.
Journal paper accepted to the Robotics and Autonomous Systems journal.
Am OFFICIALLY a doctor now, yay.
Got around to putting up my photos from the 2006 International Conference on Robotics and Automation which was held in Beijing, China.
2006 Young Achiever Awards
Read the UQ News story here (.pdf reader required).
Just got back from a gala dinner at the Carlton Crest ballroom for the 2006 Queensland Young Achiever Awards, where I was lucky enough to get the Science and Technology Young Achiever Award. Thanks must go to my supervisor Dr Gordon Wyeth for taking me on in the first place for my PhD as well as for being one of my referees. Thanks also to Dean Lane who was my other referee and Terese for nominating me in the first place. And thank you ITEE, UQ, and the ARC for providing me with the resources and support to really get stuck into my research over the past three and a half years.
After almost exactly 3.5 years, I've finally submitted my PhD thesis for examination. Yay!
My two superiors Dr Gordon Wyeth and Prof Janet Wiles (along with a number of other chief investigators) are awarded the Australian Research Council Special Research Initiative Application "Thinking Systems: Navigating Through Real and Conceptual Spaces" Grant, worth $3.3 million over 5 years. Here's the brief description of the project:
The project will provide fundamental insights into physical and conceptual spaces and develop applications in robotics and information systems. The project brings together national and international researchers to study how humans and other animals navigate: how trajectories through space are used to build maps, the neural bases of these mapping processes and how to use maps to achieve goals. The project will develop a new generation of robots that can learn about the physical spaces they work in, and create concept-mapping systems that can map and navigate information spaces. Also, it will provide new insights into the mechanisms regulating human cognition and mental dysfunctions.
The thesis is almost complete. Final draft has been given to everyone to read, will do corrections based on their comments and then submit. If you're interested in reading any of it, tell me!
I finally got around to putting up my list of massively overquoted trivia. You know those bits of trivia friends always drop into the conversation one too many times - this is a list of them. Go have a look here.
Story in the UQ News! Click here to read the full article.
Paper accepted to the International Conference on Intelligent Robots and Systems, titled "RatSLAM on the Edge: Revealing a Coherent Representation from an Overloaded Rat Brain". This one contains a heap of information on my most recent work, including the effects of shrinking our artificial rat brain, and also an algorithm that enables the robot to intelligently adapt to changes in an environment.
What a PhD student gets up to after finishing their first draft and not feeling like getting back onto it straightaway...
First thesis draft finished and scanned by supervisor. Yay! Only 224 pages and 60,000 words later...
Thesis writeup chugging along. Updated progress graph below:
Assuming linear progress (yeah right) I will hit 1000 pages in about seven and a half months :)
Rescue robot for operations in disaster zones
A soccer match between robotic dogs
Thesis writeup started with an online progress chart!
The "thesis writeup satisfaction" meter
This page count will probably oscillate slightly and won't always be going forwards.
Paper accepted into the 2005 Australian Conference on Robotics and Automation.
Am now the proud owner of a real live roboraptor, courtesy of an early birthday / Christmas present from the folks. Once I've had a good play with it I will post more photos and perhaps some video of what it can do.
Working on a journal paper for Autonomous Robots.
Finally got around to putting up some screenshots and descriptions of some games I programmed back in high school, you can see them here.
Ran my first ever full marathon at the Gold Coast on 3rd July, here are the pictures and video.
Our outdoor tractor paper got accepted into the International Conference on Field and Service Robotics - Yay!
Here's a piccie of the little red tractor:
Am in the midst of updating this entire webpage, looking at it the other day I realised it's quite out of date
Paper accepted into ICRA 2005 (International Conference on Robotics and Automation) - yay!
My PhD Project
We are trying to use the extensive research that has been carried out on the rodent hippocampus to come up with a biologically inspired system that can be used on a real mobile robot navigating around a large, unmodified indoor environment performing useful tasks. The possible applications of this work are in the short term in the area of mobile robotics whether that be service robots or robots wandering around on the surfaces of other planets. In the long term by furthering our understanding of the true nature of intelligence this work is leading towards truly autonomous robots that have genuine thinking abilities, and in the ultra long term to sentient systems.
The RatSLAM project can be found here. Below are a couple of pictures to give you an idea of what we're doing, one of the Pioneer2 DXe robot we're using and the other of the software interface for our RatSLAM system. We have also done experiments on an outdoor tractor robot from CSIRO.