The 20th and 21st joint COTA International Conference of Transportation Professionals (CICTP 2020-21)

The inertial navigation system is vital for automated driving, which has been proposed as an alternative since GPS signals are easily blocked under such driving conditions – crossing a tunnel, going through the sea, driving under trees, etc., which poses a huge safety hazard. This paper proposes an unmanned system for autonomous driving which is mainly based on inertial navigation technology. And then, the core method of the inertial navigation including the automatic driving state (velocity, displacement) update is introduced in detail. Accordingly, this paper sets up a simulation platform to test the proposed system using the tools of the Raspberry Pi, MATLAB, etc. The results show that the proposed system could provide accurate navigation information for autonomous vehicles in a short time window. However, the error of the velocity and displacement values calculated by the integral is remarkably larger than the error of the attitude angle. It indicates that the inertial navigation information must be filtered and corrected by other more accurate positioning systems if it is applied in a long time window.

CICTP