Data-Fusing Algorithms for Improved Localization Within V2X Network Using Low-Cost GPS
Lead Researcher(s)
Jing Sun
Michael G Parsons Collegiate Professor of Naval Architecture and Marine Engineering, Chair, Department of Naval Architecture and Marine Engineering, Professor of Naval Architecture and Marine Engineering, Professor of Electrical Engineering and Computer Science and Professor of Mechanical Engineering, College of Engineering
Ding Zhao
Project Team
Project Abstract
This project was designed to improve the positioning accuracy of non-real-time kinematic (RTK) GPS from meters to centimeters using vehicle-to-everything (V2X) communication: (1) Build an accurate base map using data collected from the U-M Mobility Transformation Center (MTC) Pillar 1 project; (2) query Safety Pilot Model Deployment and Pillar 1 data; (3) develop algorithms to improve positioning accuracy based on multiple-vehicle GPS information and the base map of the road network; (4) test the algorithms on smartphones and demonstrate their potential application to pedestrian safety at Mcity; and (5) validate the approach in the Pillar 1 living laboratory.
Project Outcome
(1) Algorithms to improve non-RTK GPS position accuracy from a few meters to a few centimeters, (2) methods and tools for near-real-time GPS signal correction, (3) algorithms to perform batch update of the GPS signal in the Ann Arbor Connected (Pillar 1) database, (4) an invisible-to-visible (I2V) and smartphone-based app to provide accurate pedestrian localization, and (5) demonstrated pedestrian-safety applications