Increasing
urbanisation means that city traffic networks are reaching their capacity
limits ever more frequently. Driving in urban areas means having to deal with
complex traffic situations involving many different road users, and there is a
growing risk of congestion and accidents in our cities. Traffic systems need to
be able to handle the strain, however, if the future of unrestricted mobility
is to be preserved.
Specialists from
the BMW Group are therefore also involved in the research initiative entitled
UR:BAN (German acronym standing for Urban Space: User-oriented assistance
systems and network management) and working on the development of cooperative
and intelligent driver assistance and traffic management systems.
“This will enable us to further increase safety,
efficiency and comfort in urban areas to significant effect,” explains Dr
Christoph Grote, Managing Director of BMW Forschung und Technik GmbH. One of
the main focal points is humans and the many different roles they play in the
traffic system – be it as a motorist, motorcyclist, cyclist, pedestrian or
traffic planner. For this, the research initiative has brought together 31
partners from the automotive industry and its suppliers, the fields of
electronics and software, as well as from research institutes and cities.
Running from 2012
until 2016 and funded by the German Federal Ministry for Economic Affairs and
Energy (BMWi), the initiative comprises three separate projects: “Cognitive
Assistance”, “Networked Traffic System” and “Human Factors in Traffic”. The BMW
Group is making key contributions to all three projects.
The interim results were presented at the UR:BAN midterm
event held at the German Aerospace Centre (DLR) in Braunschweig.
Today’s driver assistance systems are designed first and
foremost to make driving on motorways safer and more comfortable. There is a
whole new set of challenges to be tackled in city traffic, however, where
motorists also have to watch out for cyclists and pedestrians. The methods for
detecting situations and dangers and for aiding drivers’ braking and steering
reactions that already work successfully on motorways need to be adapted
accordingly.
As part of the sub-project “Protection of Vulnerable Road
Users”, BMW Forschung und Technik GmbH is developing a driver assistance system
to help protect pedestrians: the system analyses the situation and the
pedestrian’s behaviour to assess whether there is a risk of collision with the
vehicle. Accidents with pedestrians can be avoided by braking, steering or a combination
of the two. In a BMW 5 Series research vehicle it is already possible to
recognise detailed features of a pedestrian – i.e. the head and upper part of
the body – and to classify the direction in which the pedestrian is moving.
If driver assistance systems that effectively prevent
accidents are to work in an urban environment, they need a reliable and
complete “picture” of their surroundings and must also be able to correctly
interpret complex situations involving many different protagonists and boundary
conditions. BMW Forschung und Technik GmbH is therefore working on the
development of powerful assessment algorithms for fusing data and evaluating
situations in a cross-disciplinary sub-project designated “Measurement and
Modeling of the Environment”.
At the midterm event, object detection was demonstrated
along with free space or generic object detection using grids. The aim is to
develop a system of 360° environment modelling for urban scenarios to be used
by multiple driver assistance systems.
Source: BMW