Robust, secure and precise vehicle navigation system for harsh GNSS signal conditions

摘要

GNSS has deeply entered the consumer market. It is already widely used in car navigation systems, cell phones, and various other applications of daily life. More sophisticated applications for professional and reliability critical applications like road tolling, anti-theft systems and tracking of the transport of hazardous material have already been implemented or are under development. The wide range of GNSS based applications nevertheless also attracted the appearance of jamming devices, which showed the inherent weakness of current satellite navigation systems: the extremely low signal power. 'Privacy Protection Devices', small mobile jammers, haveappeared on the market and also the possibility of spoofing has already been demonstrated, e.g. in [7]. It is probably only a matter of time until also low-cost spoofing devices will appear on the market. To sustain the navigation capabilities even under harsh signal environments, as well as intentional interference, it is necessary to detect jamming and spoofing in order to counteract them, but also to maintain basic positioning capabilities during those phases when GNSS is not available or not trustworthy. It has been widely discussed that additional sensors could be suitable for this task. The type of sensors to be used is nevertheless rated controversially.This paper focuses on a robust and precise car navigation system which can be used in harsh signal environments, as well as jamming and spoofing conditions. The system is based on and designed for so called 'On Board Units' (OBU) as they are already used for professional road toll applications. The system has the capability to detect jamming and spoofing, but also to maintain navigation capabilities under jamming and spoofing conditions, as well as during GNSS outages, e.g. in parking garages or tunnels, and in urban canyons by using additional sensors. The sensors discussed in this publication are a COTS GPS receiver, the built-in vehicle wheel speed sensors, and a MEMS (Micro-Electro-Mechanical-Systems) type Inertial Measurement Unit (IMU). The used algorithms cover dead reckoning (DR) using differential odometers, dead reckoning (DR) using the MEMS IMU and the odometer, navigation sensor fusion filter, and the jamming and spoofing detection module. The publication analyzes and evaluates the performance of the individual systems, and focuses on the question, how the use of a MEMS IMU and the advantageous combination of sensors can improve the dead reckoning capabilities of the integrated navigation system, and how the low accuracy MEMS system can cooperate with the odometric DR system, which has been developed for the past PUMA project, funded by the European Union in the 7th Framework program. The PUMA project had its focus on spoofing detection by comparing the GNSS trajectory to a DR trajectory produced by a differential odometric system. The developed system has the capability of real-time operation and postprocessing; a hardware prototype has been successfully built for demonstration and further research. The system consists of the OBU, a COTS GPS/SBAS receiver and uses the built-in wheel speed sensors of the test vehicles. The PUMA system is now extended with a MEMS IMU to improve and extend the dead reckoning capabilities of the technology.