Abstract
Inertial Navigation System (INS) is often combined with Global Navigation Satellite System (GNSS) to increase the positioning accuracy and continuity. In complex urban environments, GNSS/INS integrated systems suffer not only from dynamical model errors but also GNSS observation gross errors. However, it is hard to distinguish dynamical model errors from observation gross errors because the observation residuals are affected by both of them in a loosely-coupled integrated navigation system. In this research, an optimal Radial Basis Function (RBF) neural network-enhanced adaptive robust Kalman filter (KF) method is proposed to isolate and mitigate the influence of the two types of errors. In the proposed method, firstly a test statistic based on Mahalanobis distance is treated as judging index to achieve fault detection. Then, an optimal RBF neural network strategy is trained on-line by the optimality principle. The network’s output will bring benefits in recognizing the above two kinds of filtering fault and the system is able to choose a robust or adaptive Kalman filtering method autonomously. A field vehicle test in urban areas with a low-cost GNSS/INS integrated system indicates that two types of errors simulated in complex urban areas have been detected, distinguished and eliminated with the proposed scheme, success rate reached up to 92%. In particular, we also find that the novel neural network strategy can improve the overall position accuracy during GNSS signal short-term outages.
Highlights
Global Navigation Satellite System (GNSS) is a commonly used technology in location-based services (LBS)
A field vehicle test in urban areas with a low-cost GNSS/Inertial Navigation System (INS) integrated system indicates that two types of errors simulated in complex urban areas have been detected, distinguished and eliminated with the proposed scheme, success rate reached up to 92%
A typical land vehicle navigation system (LVNS) based on GNSS has to operate in dense urban areas where GNSS signals are either blocked or severely degraded by phenomena such as cycle slips or multipath effects, which limit its capability to achieve satisfactory accuracy and positioning reliability [1]
Summary
GNSS is a commonly used technology in location-based services (LBS). A typical land vehicle navigation system (LVNS) based on GNSS has to operate in dense urban areas where GNSS signals are either blocked or severely degraded by phenomena such as cycle slips or multipath effects, which limit its capability to achieve satisfactory accuracy and positioning reliability [1]. Jiang et al [20] proposed an adaptively-robust strategy for GPS/INS integrated navigation systems to resist model deviations and outliers, but it applied only to tiny state perturbations and treated model deviations and outliers uniformly. Multi-Layer Perceptron (MLP) network to predict and estimate a pseudo-GPS position when the GPS signal is unavailable and demonstrated that proposed model can effectively provide corrections to standalone INS during 300 s GPS outages. A fault detection method based on Mahalanobis distance is put forward, whereafter robust and adaptive filtering algorithms, are proposed to reduce observation and dynamic model errors. The difference in position between GNSS measurements and INS measurements in the n-frame is regarded as measurements, so the integrated navigation observation model can be written as:. Where Kk is the Kalman gain matrix; the symbols “ˆ” and “~” above a variable represent an estimate and a measurement, while the superscripts “−” and “+” represent the a priori and a posteriori estimates, respectively
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
