Resilient Multi-vehicle Positioning with Sensor Errors

iStock-670631592-small
Supervisor Name/s: 
Prof Andrew Dempster
Dr Joon Wayn Cheong
Research field/s: 
Cooperative Intelligent Transport Systems
Keywords: 
C-ITS, Intelligent Transport Systems, GNSS Navigation, Multi-sensor Integration

In the past decade, ACSER has established a strong heritage in multi-sensor integration for land navigation as evidenced by its numerous high quality journal publications in this domain of research. Triggered by recent developments in vehicular safety, driverless vehicles, etc., ACSER has focussed in enhancing the accuracy and resilience of GNSS-based navigation in vehicular applications by combining GNSS with other on-board sensors. The exploitation of channel parameters from vehicle-to-vehicle (V2V) communications has resulted in significant improvement in not only positioning accuracy but also availability, allowing resilient navigation in semi-urban environments where typical standalone GPS receivers will intermittently fail. It is further shown that expanding this idea to networked vehicles enhances the overall positioning accuracy of the entire network of vehicles is analysed. These concepts led by ACSER form the cornerstone of its research in the domain of Cooperative-Intelligent Transport Systems (C-ITS).

ACSER’s prominence in this field of research has attracted the attention of Thales Alenia Space France (TAS). who has approached ACSER to study the benefits of integrating Inertial Measurement Unit (IMU), GNSS and V2V sensors. The aim is to investigate the effectiveness of exploiting the redundancy in such multi-sensor integration to mitigate the signal distortions due to multipath and Non-Line-of-Sight by rejecting measurements from effected signals. Non-conventional estimators such as modified particle-filter algorithms may be able to reject distorted and erroneous signals.