| A future where cars talk to each other
is being taken seriously. James Griffiths looks at the
steps needed to go from concept to reality and the issues along the way
The road ahead seems clear, but the prototype Willwarn system in the
Mercedes lets out a soft warning ping. “A lot of cars have been
braking harshly up ahead,” explains the driver. As we approach,
he also applies the brakes sharply and on-screen the system performs a
rapid update. It is adding his reaction to its database. When we pass
the next car our information is transferred across, so it will also know
to warn its driver that something in the road is causing others to brake
sharply.
As it turns out, the braking was simply part of a prototype demonstration
for Mercedes-Benz’s car to car communication system on a very specific
test track. Several cars were fitted with the system which provided information
to drivers about upcoming hazards – this could be a patch of slippery
road or road-works.
 |
Should I stop or
go on? Now the car can tell you what to do |
The benefits of such a system are clear: the information being passed
on is recent and relevant as both cars are moving in opposite directions
and, more importantly, drivers are only presented with what they need
to know, when they need to know it. An event several kilometres down the
road may have been logged, but the driver will not be given the information
until just before it will affect him.
But there are concerns: the information must be gathered and the system
relies on a large number of vehicles all co-operating. With few other
cars on the market with which to communicate there is little incentive
for early adopters of the technology.
Research engineer for DaimlerChrysler Andreas Hiller has some of the answers,
having worked on the Willwarn project and Car 2 Car Communication Consortium.
“The cars can learn in two main ways,” says Hiller. “When
it runs over a patch of ice or some leaves, the ABS or ESP system can
detect the change in longitude, or the drift and angle of the vehicle.”
The system then logs the information and transmits it the next time the
car passes another similarly equipped vehicle. The system can also incorporate
the actions of the driver into its learning, logging and communicating
unexplained braking events.
The other way the system gathers information is by tuning into specially
placed roadside units. Some of these are already in use today, giving
warnings to motorists by alerting them to upcoming motorway queues, or
changes in the weather. These can be adapted to communicate with compatible
vehicles. A second type of mobile unit could be placed at temporary hazards
such as road-works.
“A mobile unit would not need any algorithms inside,” says
Hiller. “It would have a button that the worker who set it up could
press to provide a GPS function saying there were road-works. At an accident
scene, police could press a button to do the same.”
These roadside units would be valuable for the systems’ introduction,
helping to solve the problem of having too few users to form a viable
base. “Drivers would not need many other vehicles for theirs to
talk to. There would be a benefit for the first five or even one per cent
of drivers,” says Hiller.
The cars could also do more with the information than simply provide audible
warnings. There are three different actions that might occur, of which
a warning would only be the first, and the most likely application for
early systems. As they develop, Hiller says cars could take control of
the situation autonomously.
“The safety systems could start to adapt themselves according to
the information received,” he says. “It might be that the
stability control system has a sports setting that allows for some drift,
or perhaps the normal setting allows you to brake more strongly. The car
could change its own settings, adjusting for any slippery areas.”
The car could take almost full control of a dangerous situation and brake
for itself if it calculates it is approaching a hazard too quickly. This
is still a long way off, however.
Cars able to communicate in this manner would also be at an advantage
in the event of a collision. The precrash systems could be activated earlier
and the vehicles could have more information on the speed, weight and
type of the other vehicle, and in what manner they will impact.
“If you know you’re going to crash into a relatively soft
passenger car, then the timing of the airbag deployment could be different
from if you were about to hit a hard barrier, such as a lorry,”
says Hilller.
It’s not just the car companies that are interested in such systems:
computer technology company IBM recently announced a research initiative
called Collaborative Driving. It looks at how such a system could be could
be used to help traffic flow in cities to improve the efficiency of traffic
systems, vehicles’ fuel economy and overall safety.
“Our research is about how traffic will look if and when these communications
technologies are deployed,” says IBM researcher Olaf Goldschmidt.
“We want to address the question of whether or not communication
between vehicles passing through an intersection can make traffic more
dynamic and efficient.”
Goldschmidt believes the systems could allow cars to drive closer together
and reduce the amount of time spent waiting at intersections.
“Right now when you come to an intersection with a traffic light
you have to slow down, stop and then pick up speed again,” he says.
“But if all the vehicles are aware of what is happening and where
the other vehicles are coming from, you may be able to organise traffic
flow at the crossroads better.”
The research was approved and given funding last year. IBM is not looking
to develop a competing system, but rather explore further ways in which
traffic can be organised.
“If you can organise traffic in a smoother way where people brake
and accelerate less and drive with a constant speed, the savings in terms
of fuel and emissions are huge,” says Goldschmidt. “But that’s
nothing compared to the damage that traffic accidents cause in terms of
loss of human life.”
Nissan has also been working on a similar system to the Car 2 Car concept,
which it calls Skyproject, and is currently trialling it on a small selection
of vehicles in real-life conditions in Japan.
Customers buying a car in Yokohama are given the option of having the
software installed, which will operate with roadside beacons similar to
those discussed by Hiller, but are funded by the government.
 Nissan
is working on a separate system that will allow vehicles to detect pedestrians
via the GPS in their mobile phones. “This system has great potential,”
says Minoru Shinohara, head of Nissan’s technology division. “Detecting
pedestrians is a very important issue. We can use cameras which will detect
a pedestrian walking directly in front of your vehicle, but in the future,
mobile phones will be much more precise.”
For the time being, the system only works to issue a general area warning
when there are many pedestrians in an area. The most immediate application
for this, says Shinohara, is for areas like schools where, depending on
the time of day, the number of pedestrians can fluctuate greatly, and
drivers may think the danger is less than it really is.
Phones with GPS are becoming more common in Europe already. When used
in conjunction with systems such as Skyproject or Willwarn, getting information
into cars about their surrounding could become easier. It could make active
safety for pedestrians much more viable in the future too. |
|
