Real-time Localisation and Tracking System for Navigation Based on Mobile Multi-sensor Fusion

Siim Plangi
With the rise of the smartphone, new research opportunities have emerged. With a wide array of sensors that are available in today’s smartphones, the research possibilities are endless. In this work, we present a new algorithm that can track and localise a vehicle in real-time using the GPS, accelerometer and gyroscope data streams from an Android OS smartphone. The resulting algorithm can respond to speed changes, and the car turns in real-time without any info from the GPS. This means that the algorithm can estimate the vehicle position if the GPS data stream is unavailable for unknown amounts of time. Results are promising and show that the algorithm performs well both in accuracy and
real-time responsiveness. Even without 30 seconds of GPS info, the algorithm is able to estimate the vehicle location with an average accuracy of 25 meters.
Graduation Thesis language
Graduation Thesis type
Master - Computer Science
Amnir Hadachi, PhD; Artjom Lind, MSc
Defence year