Publication

Accurate vehicle positioning is a critical enabler for safety functions in automated driving (AD) and advanced driver assistance systems (ADAS), particularly at higher levels of automation. In level 3 ADAS, where the vehicle must make autonomous decisions, precise positioning is essential to correctly interpret the environment and select the most probable path (MPP). A positioning error can result in the wrong path being chosen, potentially leading to unsafe situations.

Today’s positioning solutions typically combine GPS, IMUs, odometers, and on-board sensors such as cameras and radars. However, GPS suffers from limited availability in urban environments and tunnels, while other sensors provide only relative information. Cellular signals offer a promising complement due to their ubiquity and cost-efficiency,
and their use in positioning has been emphasized in several automotive-focused research projects.

This project investigates the use of low Earth orbit (LEO) non-terrestrial networks (NTN) as a positioning source and explores how LEO-based data can be fused with terrestrial network (TN) information to enhance positioning accuracy and reliability for heavy vehicles. LEO satellites, which orbit much closer to the Earth than GPS satellites, offer
higher signal strength, better penetration into vehicle cabins, and lower latency, all critical for safety applications in AD/ADAS. Additionally, the rapid movement of LEO satellites provides richer positioning data over time, further improving accuracy. 

The goal is to leverage combined NTN and TN positioning to support safe and reliable operation of large vehicles, especially in challenging environments where conventional positioning technologies fall short.