Integrity Monitoring of GNSS-based Vehicle Positioning based on Cooperative Weighting Strategy

With the rapid development of Cooperative Intelligent Transportation System (C-ITS) and the vehicle-infrastructure cooperation technology, there have been new requirements for the integrity monitoring of urban vehicle positioning based on Global Navigation Satellite System (GNSS). The existing weight evaluation solutions based on certain satellite observation information have to be enhanced for advanced GNSS-enabled applications. Under the cooperative vehicle-infrastructure scheme, this paper investigates the novel GNSS weighting method by combining the GNSS observation information from the subject vehicle and its neighborhoods in a local area. The Back-Propagation Neural Network (BPNN) is utilized to build prediction models to describe the relationship between the satellite observation information and pseudo-range residual. With the off-line models, weight of GNSS measurements can be real-timely evaluated by integrating the information of all connected vehicles within a specific region area, which makes good use of the cooperation by wireless communication. Experiments were carried out to compare the proposed cooperative weighting strategy and other related methods. The results demonstrate the capability and potentials of this proposal in integrity monitoring by over-bounding the horizontal position error with the horizontal protection level, which is significant in many safety-related GNSS applications in novel transportation systems.