The integration of Global Navigation Satellite Systems (GNSS) carrier phases with Inertial Navigation System (INS) measurements is essential to provide accurate and continuous position, velocity and attitude information, however it is necessary to fix ambiguities rapidly and reliably to obtain high accuracy navigation solutions. In this paper, we present the notion of combining the Global Positioning System (GPS), the BeiDou Navigation Satellite System (BDS) and low-cost micro-electro-mechanical sensors (MEMS) inertial systems for reliable navigation. An adaptive multipath factor-based tightly-coupled (TC) GPS/BDS/INS integration algorithm is presented and the overall performance of the integrated system is illustrated. A twenty seven states TC GPS/BDS/INS model is adopted with an extended Kalman filter (EKF), which is carried out by directly fusing ambiguity fixed double-difference (DD) carrier phase measurements with the INS predicted pseudoranges to estimate the error states. The INS-aided integer ambiguity resolution (AR) strategy is developed by using a dynamic model, a two-step estimation procedure is applied with adaptively estimated covariance matrix to further improve the AR performance. A field vehicular test was carried out to demonstrate the positioning performance of the combined system. The results show the TC GPS/BDS/INS system significantly improves the single-epoch AR reliability as compared to that of GPS/BDS-only or single satellite navigation system integrated strategy, especially for high cut-off elevations. The AR performance is also significantly improved for the combined system with adaptive covariance matrix in the presence of low elevation multipath related to the GNSS-only case. A total of fifteen simulated outage tests also show that the time to relock of the GPS/BDS signals is shortened, which improves the system availability. The results also indicate that TC integration system achieves a few centimeters accuracy in positioning based on the comparison analysis and covariance analysis, even in harsh environments (e.g., in urban canyons), thus we can see the advantage of positioning at high cut-off elevations that the combined GPS/BDS brings.

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.

This paper propose a scheme for indoor pedestrian location, based on UWB (Ultra Wideband)/PDR (Pedestrian Dead Reckoning) and Floor Map data. Firstly, a robust algorithm that uses Tukey weight factor and a pathological parameter for UWB positioning is proposed. The ill-conditioned position problem is solved for a scene where UWB anchors are placed on the same elevation of a narrow corridor. Secondly, a heading angle-computed strategy of PDR is put forward. According to the UWB positioning results, the location of pedestrians is mapped to the Floor Map, and 16 possible azimuth directions with 22.5° interval in this position are designed virtually. Compared to the heading angle of PDR, the center direction of the nearest interval is adopted as the heading. However, if the difference between the head angles of PDR and the nearest map direction is less than five degrees, the heading angle of PDR is regarded as the moving heading. Thirdly, an EKF (Extended Kalman Filter) algorithm is suggested for UWB/PDR/Floor Map fusion. By utilizing the positioning results of UWB, PDR, and the possible heading angle of Floor Map, high precision positioning results are acquired. Finally, two experimental scenarios are designed in a narrow corridor and computer room at a university. The accuracy of pedestrian positioning when all the data are available is verified in the first scenario; the positioning accuracy of a situation where part of UWB is unlock is verified in the second scenario. The results show that the proposed scheme can reliably achieve decimeter-level positioning.

This paper proposes a method for determining a pedestrian’s indoor location based on an UWB (ultra-wideband) and vison fusion algorithm. Firstly, an UWB localization algorithm based on EKF (extended Kalman filter) is proposed, which can achieve indoor positioning accuracy of 0.3 m. Secondly, a method to solve scale ambiguity and repositioning of the monocular ORB-SLAM (oriented fast and rotated brief-simultaneous localization and mapping) algorithm based on EKF is proposed, which can calculate the ambiguity in real time and can quickly reposition when the vision track fails. Lastly, two experiments were carried out, one in a corridor with sparse texture and the other with the light brightness changing frequently. The results show that the proposed scheme can reliably achieve positioning accuracy on the order of 0.2 m; with the combination of algorithms, the scale ambiguity of monocular ORB-Slam can be solved, with the failed vision trace repositioned by UWB, and the positioning accuracy of UWB can be improved, making it suitable for pedestrian location in indoor environments with sparse texture and frequent light brightness changes.