Real-Time Integrity Monitoring for Civil Aviation with Improved Navigation Performance

Abstract

With the worldwide development in the past decades, multi-constellation Global Navigation Satellite System (GNSS) are able to provide consistent and reliable navigation services today, which are expected to bring significant performance improvement to civil aviation in the future. For the GNSS-based aircraft navigation, meeting the integrity and continuity requirements is of the most importance. In the currently proposed baseline Advanced Receiver Autonomous Integrity Monitoring (ARAIM) user algorithm, the integrity risk is evaluated using a conservative upper bound. Despite its computational efficiency, this bound is not tight enough, which may lead to overly conservative results. Operationally, the system may incorrectly alert the user, which severely impacts navigation continuity. Therefore, in this work, we develop a new method to tightly bound the integrity risk and establish a multi-constellation ARAIM test platform to validate the theory. The new approach takes advantage of the independence between position estimation error and detection test statistics and expresses the integrity risk evaluation as a convex optimization problem. It is shown that the global maximum of the objective function is a tight bound on integrity risk, and it can be efficiently computed using an numerical method. Other than the theoretical derivations, another major contribution of this work is prototyping the ARAIM user segment in the Guidance, Navigation, and Control (GNC) laboratory at Shanghai Jiao Tong University. Both of the ARAIM Multiple Hypothesis Solution Separation (MHSS) algorithm and the new approach are incorporated into the prototype, and the real-time integrity monitoring results are visually displayed in terms of horizontal and vertical protection levels, effective monitoring threshold, integrity risk, etc. As compared to the existing MHSS theory, the results suggest that the navigation service availability can be noticeably improved using the proposed method, especially when the constellations are subject to larger ranging errors.