Weather alerts for motorists

AN IN-CAR satellite navigation system offers a handy way to get around, but what if it sends you into a snowdrift or a two-hour…


AN IN-CAR satellite navigation system offers a handy way to get around, but what if it sends you into a snowdrift or a two-hour traffic tailback?, asks DICK AHLSTROM

The difficulties faced by some drivers during the recent snow and ice were not made any easier by satnav systems, proof that the shortest journey is not always the fastest or the safest. The satnav may know the roads but it doesn’t know what the traffic is up to or the condition of the roads.

The next generation satnav systems however will be smart enough to deal with these issues, says NUI Maynooth’s Prof Stewart Fotheringham. He heads the university’s National Centre for Geocomputation (NCG), set up 2004 with funding from Science Foundation Ireland.

Within five years he expects that satnav systems will combine road network data with real-time information about road conditions, traffic movement and transient risks such as snow, ice or flooding.

READ MORE

All this will be streamed to the satnav, providing the driver with a new level of information about what route to take and how to avoid problems.

Maynooth’s NCG involves the work of 35 researchers, but a much larger research collaboration has formed in the area of geocomputation and the handling of spatial data. Fotheringham also heads this group, the Strategic Research Cluster in Advanced Geotechnologies (StratAG, stratag.ie). It involves researchers from Maynooth, University College Dublin, Trinity College Dublin and the Dublin Institute of Technology, along with private-sector collaborators.

The goal is to provide some of the discoveries in mathematics, electronics and data interpretation needed to deliver next generation satnav, Fotheringham explains. “Geocomputation is all about computer algorithms for handling spatial data, data that has positional information.”

Today’s satnav systems are good at this but they can’t warn the driver about wasted time if a bridge is out or if there is a tailback behind an accident. “Five years from now we will look back at today’s satnav and think these were very crude machines.”

Satnav tells you where you are by sampling signals sent from orbiting satellites. GPS is the current US satellite cluster, with European countries collaborating on the Galileo satellite cluster ready for use in 2014.

“The satnav knows where you are and works out an optimal route based on shortest distance or optimal travel time,” Fotheringham says. “What we really want it to do is look at road conditions, for example looking for road-works, but also traffic conditions, for example traffic jams. That is what the next generation of satnav will have.”

The NCG and StratAG are conducting research to make this happen. There are two key challenges: first, how to get the raw data from sensors or other sources; and second, how to interpret the data to deliver useful information for the driver.

“We have the ability to control huge amounts of data, the challenge is in how you interpret it,” says Fotheringham. “We are working on algorithms [math- ematical processing methods] to interpret the data coming in from a lot of sensors.”

Researchers are concentrating on three ways to capture data. One is to build roadside sensors that can count cars, detect speed, measure road or air temperatures and detect rainfall, among other things. This could then be fed back and processed to deliver useful information.

Another is to build a variety of sensors into the cars themselves. “That is starting to happen,” Fotheringham says. “Cars themselves could become the sensors.”

A third approach involves using something that is ubiquitous and likely to be found in every car, a mobile phone. “Lots of work is underway to use mobile phones to estimate traffic conditions,” Fotheringham says.

This involves measuring how long it takes for a car to move between coverage from one phone mast to the next.

“We are working with O2 to work out average speed in real time based on the mast signals.”

As the volume of information increases, researchers will also have to find ways to move this large amount of data back and forth between a source and the satnav device itself.

Fotheringham believes that this will allow the current bird’s-eye-view satnav maps to be replaced by streaming video. This will actually show the driver the roads and the intersections ahead when turns are made.