Better Player Position in Rough/Urban Terrain How sensor fusion of GNSS, UWB, IMU could improve the positioning in urban or rough terrain.

Abstract

This thesis investigates the development of a scalable and robust positioning system designed for use in training environments such as forests and urban terrain, where traditional Global Navigation Satellite Systems (GNSS) often suffer from limited reliability due to signal blockage and multipath interference. To overcome these limitations, a system was designed combining GNSS, Inertial Measurement Units (IMUs), and Ultra-Wideband (UWB) technologies using advanced sensor fusion algorithms. A prototype was implemented and tested in multiple environments using stationary and dynamic test scenarios. The results show that while each sensor has specific strengths and limitations, sensor fusion significantly enhances positioning accuracy and stability, particularly in environments where GNSS data alone proves unreliable. The system was evaluated using real-world data, and its performance was benchmarked against high-accuracy reference systems such as Real-Time Kinematic (RTK). The study concludes that sensor fusion provides a cost-effective and scalable way to improve positional awareness in demanding environments.