In life-critical applications, the real-time detection of faults is very important in Global Positioning System/Inertial Navigation System (GPS/INS) integrated navigation systems. A new fault detection method for soft fault detection is developed in this paper with the purpose of improving real-time performance. In general, the innovation information obtained from a Kalman filter is used for test statistic calculations in Autonomous Integrity Monitored Extrapolation (AIME). However, the innovation of the Kalman filter is degraded by error tracking and closed-loop correction effects, leading to time delays in soft fault detection. Therefore, the key issue of improving real-time performance is providing accurate innovation to AIME. In this paper, the proposed algorithm incorporates Least Squares-Support Vector Machine (LS-SVM) regression theory into AIME. Because the LS-SVM has a good regression and prediction performance, the proposed method provides replaced innovation obtained from the LS-SVM driven by real-time observation data. Based on the replaced innovation, the test statistics can follow fault amplitudes more accurately; finally, the real-time performance of soft fault detection can be improved. Theoretical analysis and physical simulations demonstrate that the proposed method can effectively improve the detection instantaneity.

A dynamic state realization for tightly coupling Global Positioning System (GPS) measurements with an Inertial Navigation System (INS) is described. The realization, based on the direct fusion of GPS and INS systems through Kalman filter state dynamics, explicitly accounts for temporal and spatial decorrelation of GPS measurement errors (such as tropospheric, ionospheric, and multipath errors) through state augmentation, thereby ensuring Kalman filter integrity under fault-free error conditions. Predicted system performance for a Local Area Augmentation System (LAAS) aircraft precision approach application is evaluated using covariance analysis and validated with flight data.

Built-in fault detection mechanisms based on the Kalman filter innovations are also evaluated to help provide integrity under certain fault conditions. It is shown that an algorithm based on the integral of Kalman filter innovations outperforms other existing GPS fault detection methods in the detection of slowly developing ranging errors, such as those caused by ionospheric and tropospheric anomalies.

Integration of Global Positioning System (GPS) and Inertial Navigation System (INS) technologies can overcome the drawbacks of the individual systems. One of the advantages is that the integrated solution can provide continuous navigation capability even during GPS outages. However, bridging the GPS outages is still a challenge when Micro-Electro-Mechanical System (MEMS) inertial sensors are used. Methods being currently explored by the research community include applying vehicle motion constraints, optimal smoother, and artificial intelligence (AI) techniques. In the research area of AI, the neural network (NN) approach has been extensively utilised up to the present. In an NN-based integrated system, a Kalman filter (KF) estimates position, velocity and attitude errors, as well as the inertial sensor errors, to output navigation solutions while GPS signals are available. At the same time, an NN is trained to map the vehicle dynamics with corresponding KF states, and to correct INS measurements when GPS measurements are unavailable. To achieve good performance it is critical to select suitable quality and an optimal number of samples for the NN. This is sometimes too rigorous a requirement which limits real world application of NN-based methods.

The support vector machine (SVM) approach is based on the structural risk minimisation principle, instead of the minimised empirical error principle that is commonly implemented in an NN. The SVM can avoid local minimisation and over-fitting problems in an NN, and therefore potentially can achieve a higher level of global performance. This paper focuses on the least squares support vector machine (LS-SVM), which can solve highly nonlinear and noisy black-box modelling problems. This paper explores the application of the LS-SVM to aid the GPS/INS integrated system, especially during GPS outages. The paper describes the principles of the LS-SVM and of the KF hybrid method, and introduces the LS-SVM regression algorithm. Field test data is processed to evaluate the performance of the proposed approach.

To ensure high accuracy results from an integrated GPS/INS system, the carrier phase observables have to be used to update the filter's states. As a prerequisite the integer ambiguities must be resolved before using carrier phase measurements. However, a cycle slip that remains undetected (and uncorrected) will significantly degrade the filter's performance. In this paper, an algorithm that can effectively detect and identify any type of cycle slip is presented. The algorithm uses additional information provided by the INS, and applies a statistical technique known as the cumulative-sum (CUSUM) test. In this approach, cycle slip decision values can be computed from the INS-predicted position (due to the fact that its short-term accuracy is very high), and the CUSUM test used to detect cycle slips (as it is very sensitive to abrupt changes of mean values). Test results are presented to demonstrate the effectiveness of the proposed algorithm.