ABSTRACT
ABSTRACT: This paper demonstrates the potential of integrated systems to provide the necessary outputs of position, attitude and visualisation to support augmented reality (AR) applications. AR systems have been identified in many areas as holding enormous promise to enhance human management of complex systems, such as power plant maintenance procedures (Klinker et al. 2001) and cardiac surgery (Devernay et al. 2001). Key to the effectiveness of AR systems is the performance of the integrated positioning system, as this establishes the accuracy of the visualisation component, i.e. how well virtual objects can be aligned with the real world. This paper presents a prototype AR system that combines a multi-antenna array of dual frequency GPS receivers, a fibre optic gyro and vehicle odometer as an integrated positioning system, with real-time imagery containing augmented objects. It describes the calibration procedures for the visualisation component as well as the Kalman filter operating as the central processing engine for the integrated positioning system. The approaches adopted to tune the filter for operation in both high and low dynamics and reduce inherent sensor noise are also presented. A case study undertaken within the land mobile environment is used to demonstrate the performance of the AR prototype as a means of improving a driver’s ability to “see” the road and surrounding vehicles despite poor visibility. The AR prototype designed, the testing procedures adopted and results obtained in this research are fully described in this paper.
