Optimal Data Fusion Algorithm for Navigation Using Triple Integration of PPP-GNSS, INS, and Terrestrial Ranging System

Abstract

A good port management system must be able to perform safe, predictable and efficient execution of transport processes. In order to improve the quality of the port management, a robust navigation system is required, which enables to provide the positions of vessels 24/7 on either open or impeded environment. This paper describes a robust georeferencing system that could satisfy centimetre-level accuracy requirements in port environments. The design is based on the loosely-coupled integration of Global Navigation Satellite System (GNSS) technology, a terrestrial radio frequency ranging system known as "Locata", and an inertial navigation system (INS). GNSS observations are processed using the precise point positioning (PPP) approach instead of the conventional differential approach. To satisfy both accuracy and reliability requirements, three integration algorithms -centralised Kalman filtering (CKF), federated Kalman filtering (FKF) and global optimal filtering (GOF) -are investigated and implemented into a "triple-integrated" PPP-GNSS/Locata/INS system. A preliminary performance assessment, which is based on the analysis of real data, concludes that all the three integration algorithms are able to provide centimetre-level positioning solutions. The results show that the FKF and CKF algorithms have similar performance, whereas the GOF solution has higher accuracy. Moreover, outlier simulation is conducted and the result verifies the outlier fault-tolerant capability of the GOF based PPP-GNSS/Locata/INS integrated system.

Multiple Faulty GNSS Measurement Exclusion Based on Consistency Check in Urban Canyons

“…To compensate the effect of GPS outage caused by multipath and NLOS, GPS integrates with other sensors or additional information such as low-cost MEMS level inertial navigation system (INS) [16][17][18][19][20][21], magnetometer and other sensors [22,23] and 3D digital maps [24][25][26]. This paper aims to enhance GPS performance before its integration with other sensors.…”

Background and Recent Advances in the Locata Terrestrial Positioning and Timing Technology

“…Classifying according to measurement models used, there are loosely-coupled, cascaded-coupled, and tightly-coupled integration architectures [44]. Classifying by filter structures, the Centralised KF (CKF), Decentralised KF (DKF), Parallel KF (PKF), Federated KF (FKF), and Global Optimal Filtering (GOF), have been investigated for marine, aviation, and land vehicle navigation applications [45,46,47].…”

“…Sydney Harbour by Locata Corporation, UNSW Sydney, and several New South Wales government agencies conducted a field trial in October 2012 to demonstrate Locata’s positioning capability in harbour areas. Experiment analyses that were focused on Locata carrier phase processing and Locata/INS loosely-coupled integration systems [33], on GPS/Locata precise point positing [23], and on different data fusion algorithms applied to triple-integrated GPS/Locata/INS systems [45]. However, the publications that focused on evaluating the positioning performance at the central harbour areas where the Locata signals were relatively stable and there was high availability.…”

Random vector space approach applied in integrated navigation information fusion of UAVs

“…All of those vectors above can be used to construct the bases vectors in RVS by orthogonalization theorem [13]. And the global optimal data fusion algorithm through RVS are given below [14]: (11) (12) …”