5.3 - Testing Acquisition of GPS / GNSS Location and Velocity to Improve Safety in Autonomous Driving
- ettc2018 - European Test and Telemetry Conference
2018-06-26 - 2018-06-28
- 5. Time-Space position technologies
- K. von Hünerbein, W. Lange - Lange-Electronic GmbH, Gernlinden (Germany)
- 106 - 113
GPS (Global Positioning) and GNSS (Global Navigation Satellite Systems) yield very accurate positioning, velocity and timing, which are crucial for many ADAS (advanced driver assistance systems) and in autonomous driving. 3D locations can be as accurate as a few decimeters, with the help of additional correction methods. In addition a wide range of different sensors, among them optical, LIDAR and RADAR are used to detect obstacles and street marks, GNSS location and trajectory data can be exchanged via wireless data links with other cars (Car2Car) or sent to infrastructure (Car2X), in order to improve early detection of possible accidents, avoid drastic breaking events on the road ahead, traffic jams and detect vehicles around the corner, which are not visible to the driver due to obscuration. This element of telemetry in ADAS and autonomous driving serves to improve safety of all traffic participants and efficiency of driving. In order to guarantee safety of ADAS and autonomous driving, millions of test km need to be driven on different roads in diverse environments. Especially in city centers and mountaineous areas, GNSS reception can be impaired by obscuration of signals by buildings, bridges, vegetation and mountains and by signals reflected on even surfaces, called multipath. Furthermore, there are many sources of interference, potentially jamming, spoofing and meaconing GPS/GNSS receivers, especially on motorways and in cities. Testing can become very time consuming and costly, especially when driving all the required kms. Thus, it is more efficient to record the GPS/GNSS and interference signal environment during test drives with advanced record and replay systems, to be able to reproduce the realistic signal environment inside the laboratory, enabling repeatable and realistic tests. Such new record and replay systems are able to record GNSS signals on all frequency bands with a high bandwidth and resolution plus many additional sensors synchronously. In this paper we will present the use cases of GPS/GNSS data in autonomous driving, Car2Car and Car2X, sources of errors and vulnerabilities for GPS/GNSS and appropriate test systems to allow for repeatable and realistic tests in the laboratory. The aim is to decrease time and money spent on testing and to improve the safety of driver assistance systems and autonomous driving.