<![CDATA[Personal mobility vehicles (PMVs) are gaining popularity for short urban trips, reducing car reliance and urban pollution. The development of autonomous PMVs heavily relies on accurate localization, often using the global positioning system (GPS) as a primary sensor. However, standard GPS suffers from poor accuracy, which requires data fusion with supplementary sensors to improve precision. This study presents a sensor fusion approach using low-cost, consumer-grade hardware to enhance the PMV localization. The fusion system integrates data from an inertial measurement unit (IMU) and wheel odometry with GPS, fusing them via Kalman Filter (KF) and Extended Kalman Filter (EKF) methods. A field experiment was conducted along a 67-meter route at velocities ranging from 0.25 to 1.23 m/s. Comparative analysis has shown that the EKF method consistently outperforms the standard KF, improving positioning accuracy by approximately 29 % and reducing the maximum deviation to a range of 1.8 m to 2.7 m across different velocities. The results have confirmed the EKF as an effective and reliable strategy for achieving high-precision localization with affordable sensors, a key step towards scalable autonomous navigation for PMVs.]]>
Copyrights © 2025
