AUSTRALIA – US – Studying rats and mice finding their way around might seem an unlikely topic for someone whose discoveries could revolutionise truck driving but many in the road haulage community will watch with interest the developments in the work of Dr Michael Milford from Queensland University of Technology's (QUT) Science and Engineering Faculty following his presentation of his paper SeqSLAM: Visual Route-Based Navigation for Sunny Summer Days and Stormy Winter Nights at the International Conference on Robotics and Automation (ICRA)in Minnesota in May this year.
Dr Milford has taken on the challenge of making the ubiquitous ‘Sat Nav’ GPS units obsolete, a feat which just a few years ago would have seemed ridiculous given the recent pedigree of such satellite technology. GPS systems however have a fundamental weakness in that they are of course dependent on a totally reliable network of space based hardware extremely vulnerable to any major disturbances such as increased solar activity.
The work which the doctor is undertaking is a completely new field built upon visual navigation algorithms, dubbed SeqSLAM (Sequence Simultaneous Localisation and Mapping), using local best match and sequence recognition components to lock in locations. As opposed to the three satellite signals needed as a minimum to triangulate an exact position the doctor is using low resolution cameras which ‘study’ their surroundings and recognise where they are.
Such kit would have sounded ludicrous before we witnessed the advent of systems such as Google Street View but suddenly the research undertaken by Dr Milford, which has been funded for three years by a grant of A$375,000 from the Australian Research Council, begins to sound logical. The doctor explains:
"My core background is based on how small mammals manage incredible feats of navigation despite their eyesight being quite poor. As we develop more and more sophisticated navigation systems they depend on more and more maths and more powerful computers. At the moment you need three satellites in order to get a decent GPS signal and even then it can take a minute or more to get a lock on your location. There are some places geographically, where you just can't get satellite signals and even in big cities we have issues with signals being scrambled because of tall buildings or losing them altogether in tunnels.
"But no one's actually stepped back and thought 'do we actually need all this stuff or can we use a very simple set of algorithms which don't require expensive cameras or satellites or big computers to achieve the same outcome?’ For example if I am in a kitchen in an office block, the algorithm makes the assumption I'm in the office block, looks around and identifies signs that match a kitchen. Then if I stepped out into the corridor it would test to see if the corridor matches the corridor in the existing data of the office block lay out.
"If you keep moving around and repeat the sequence for long enough you are able to uniquely identify where in the world you are using those images and simple mathematical algorithms."
The system obviously has a lot of work needed to make it a practical, and relatively inexpensive, navigation system, but the inherent dangers of relying on global positioning systems may mean that soon the navigation system in your cab may have to rely on a clean windscreen so it can deduce where in the world you are.
Dr Milford’s paper will be presented at the ICRA, theme ‘Robots and Automation: Innovation for Tomorrow's Needs’ which will be held in Saint Paul, Minnesota between the 14th and 18th May 2012.
Photo:- Dr Michael Milford