CrashPos - Single-frequency RTK for an Advanced Driver Assistance System Test Bed

Road traffic injuries are the leading cause of death among young people. People referred to as “vulnerable road users” such as pedestrians or cyclists are especially at risk. As a result, one of the major tasks within automotive industry is to enhance Advanced Driver Assistance Systems (ADAS) which support the driver in critical situations. Examples of ADAS are pedestrian detection systems or emergency brake assistants.

The increasing importance of ADAS in the car industry also requires the development of control modules designed to facilitate the test process of these systems with a special focus on realistic tests. Such a test system usually consists of a test vehicle, one or more dummy platforms and a control server. One very important part of the test system is the accurate positioning and guidance of the vehicles involved, which is generally solved with Global Navigation Satellite Systems (GNSS).

Currently available test systems are based on high-cost positioning equipment. The scope of this project is to use reasonably priced positioning components as well as a unique dummy platform called UFO (Ultraflat Overrunnable Obstacle), which was developed by Dr. Steffan – Datentechnik Gesellschaft m.b.H.. The UFO platform can be used to simulate accident scenarios, e.g. with pedestrians. It is also flat enough to be overrun by a standard car, thus preventing vehicle damage. To meet the price and size restrictions, a single-frequency Real Time Kinematics (RTK) solution, a relative GNSS technique, will be calculated on board the test vehicle and the dummy platform(s).

The vehicle motions are highly kinematic - up to 80 km/h are expected. Special challenges of the project include mounting the flat antenna on the UFO, satellite shadowing and multipath as well as meeting the accuracy requirements with low-cost equipment.

The relative RTK solution will thus be integrated with inertial measurements, which are essential to increase the update rate and improve the position solution received from relative GNSS measurements. The final trajectory positions will be sent to a control server to enable supervising the test of the specific ADAS module in real time.

Coordinator

Graz University of Technology, Institute of Navigation - Roman Lesjak

Partners

• Alpen-Adria-Universität Klagenfurt, Institute for Smart-Systems Technologies - Vamsi Prakash Makkapati
• Dr. Steffan Datentechnik Gesellschaft m.b.H. - Andreas Moser

Graz University of Technology - Institute of Navigation
Roman Lesjak
Steyrergasse 30
A-8010 Graz
Tel.: +43 (316) 873 - 6832
E-mail: roman.lesjak@tugraz.at
Web: www.inas.tugraz.at