Tightly Coupled GNSS/INS Integration via Factor Graph and Aided by Fish-Eye Camera

Abstract

GNSS/INS integrated solution has been extensively studied over the past decades. However, its performance relies heavily on environmental conditions and sensor cost. The GNSS positioning can obtain satisfactory performance in the open area. Unfortunately, its accuracy can be severely degraded in a highly urbanized area, due to the notorious multipath effects and none-line-of-sight (NLOS) receptions. As a result, excessive GNSS outliers occur, which causes huge error in GNSS/INS integration. This paper proposes to apply a fish-eye camera to capture the sky view image to further classify the NLOS and line-of-sight (LOS) measurements. In addition, the raw INS and GNSS measurements are tightly integrated using a state-of-the-art probabilistic factor graph model. Instead of excluding the NLOS receptions, this paper makes use of both the NLOS and LOS measurements by treating them with different weightings. Experiments conducted in typical urban canyons of Hong Kong showed that the proposed method could effectively mitigate the effects of GNSS outliers, and an improved accuracy of GNSS/INS integration was obtained, when compared with the conventional GNSS/INS integration.

A Review of Environmental Context Detection for Navigation Based on Multiple Sensors

“…For the same objective of NLOS confirmation, we can propose another indicator of the distance between a projected satellite and the sky, also by taking into account the uncertainties in the sky extraction, as well as in the position estimation. Wen et al [76] proposed to weight NLOS and LOS satellites differently instead of a simple filtering of NLOS satellites. The method showed an improvement in terms of positioning, but the authors did not compare their work with other papers.…”

Robust Visual-Inertial Integrated Navigation System Aided by Online Sensor Model Adaption for Autonomous Ground Vehicles in Urban Areas

“…The IMU measurements can be employed to constrain the motion between two epochs using the standard IMU mechanism [44], which can work efficiently in the filtering-based sensor fusion, such as the extended Kalman filter (EKF) [45]. However, the standard IMU mechanism [44] can cause a high computation load in sensor fusion using FGO [46], due to the high frequency of IMU measurement. We employ the state-of-the-art IMU pre-integration technique [47,48] to integrate the IMU measurements, which can effectively alleviate the high computation load in FGO and the accuracy is guaranteed, by integrating multiple IMU measurements into a single factor in FGO.…”

“…Exclusion all NLOS satellites will severely distort the geometry distribution of satellites in deep urban areas, and even cause a lack of satellites for further positioning. In our latest research [40] in GNSS positioning that makes use of both the NLOS and line-of-sight (LOS) measurements by giving them with different weightings and improved positioning performance is obtained. Therefore, we believe that remodel the outlier measurement is preferable.…”

“…Moreover, our recent work [41] shows that the excessive exclusion of DFP can distort the geometry of feature distribution, which can degrade the performance of VINS. Inspired by the work in [33,34,40] and our findings in [10,41], this paper proposes to estimate the sensor model of visual measurement (tracked feature) online based on the quality of feature tracking. Firstly, a self-tuning covariance estimation approach is proposed to model the uncertainty of each feature measurements by integrating two parts: 1) the geometry of feature distribution (GFD), 2) the quality of feature tracking.…”

Multi-Agent Collaborative GNSS/Camera/INS Integration Aided by Inter-Ranging for Vehicular Navigation in Urban Areas

“…Whereas, the performance of GNSS solutions can be significantly degraded in urban canyons, due to the severe multipath effects and non-line-of-sight (NLOS) receptions caused by the high-rising building reflections and blockage [5]. To mitigate the impacts of the errors caused by multipath and NLOS receptions, numerous researches [6][7][8][9] are proposed to exclude [8,10], correct [9] or re-model [11] the outlier GNSS raw measurements to further improve the GNSS solution in urban canyons. However, these methods rely on the availability of additional costly 3D LiDAR sensors or 3D building model information.…”