A GRU and AKF-Based Hybrid Algorithm for Improving INS/GNSS Navigation Accuracy during GNSS Outage

Tang, Yanan; Jiang, Jinguang; Liu, Jianghua; Yan, Peihui; Tao, Yifeng; Liu, Jingnan

doi:10.3390/rs14030752

Cited by 30 publications

(14 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tang et al proposed a gated recurrent unit (GRU) and an adaptive Kalman filter-based hybrid algorithm to predict position variations during GNSS outages, which have the advantages of high efficiency and accuracy. The above studies verified that CNN can extract complex observation information, and that LSTM/GRU can selectively memorize and predict time-series information [40]. Tao et al proposed a CNN-LSTM based real-time multipath mitigation strategy, which shook off the formation mechanism of multipaths and realized the effective mitigation of GPS multipaths [41].…”

Section: Introductionmentioning

confidence: 80%

ConvGRU-MHM: a CNN GRU-enhanced MHM for mitigating GNSS multipath

Tong,

Liu,

Tao

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

In high-precision global navigation satellite system (GNSS) short-baseline positioning, multipath is the main source of errors. If the station environment is quasi-static, repeat periods of satellites can be utilized to generate time- or space-dependent multipath models to mitigate multipaths. However, two general problems are associated with the multipath models constructed based on satellite mechanics: (1) an accuracy decrease occurs when the above models are applied to multipath mitigation over a long time-span; (2) when constructing the spatial and temporal grids of the satellite-based spatially dependent multipath model, it is challenging to balance computational efficiency and spatial resolution. We propose a convolutional neural network-gated recurrent unit enhanced multipath hemispherical map (ConvGRU-MHM) in the observational domain to address these problems. The proposed method directly mines the deep features of elevation, azimuth angle, and multipath and the mapping relationship between these to establish a real-time prediction model. The predicted multipath is obtained and returned to the observation equation for multipath mitigation when the real-time position of the satellite is placed in the pre-trained model. We compared the multipath mitigation performance of sidereal filtering (SF) and a multipath hemispherical map (MHM) with that of the ConvGRU-MHM to demonstrate the advantages of the proposed method. The experimental results are as follows: (1) in the short time-span (first 20 days), the mean accuracy improvements of the ConvGRU-MHM in the E/N/U direction performed better than those of the SF and MHM; and (2) in the long-term time (after 50 days), the mean accuracy improvements of the ConvGRU-MHM in the E/N/U direction are higher than that of the SF and MHM by 10–20%. As a lightweight model, the ConvGRU-MHM can effectively improve the measurement accuracy of GNSS real-time monitoring in fields, such as deformation monitoring and seismic research.

show abstract

Section: Introductionmentioning

confidence: 80%

ConvGRU-MHM: a CNN GRU-enhanced MHM for mitigating GNSS multipath

Tong,

Liu,

Tao

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Inertial navigation systems can provide rich navigation information and do not rely on external equipment, but they also have some shortcomings such as error divergence, and cannot provide time. Satellite navigation has the advantages of stable precision output and providing time references, but the signal is vulnerable to environmental interference and dynamic response lag [ 24 , 25 , 26 ]. Inertial/satellite tight integration can be used to estimate the attitude parameters of the carrier in a specific environment [ 27 , 28 ].…”

Section: Mems/gnss Combined Attitudementioning

confidence: 99%

An Adaptive Fusion Attitude and Heading Measurement Method of MEMS/GNSS Based on Covariance Matching

Sun

2022

Micromachines

View full text Add to dashboard Cite

Aimed at the problem of filter divergence caused by unknown noise statistical characteristics or variable noise characteristics in an MEMS/GNSS integrated navigation system in a dynamic environment, on the basis of revealing the parameter adjustment logic of covariance matching adaptive technology, a fusion adaptive filtering scheme combining innovation-based adaptive estimation (IAE) and the adaptive fading Kalman filter (AFKF) is proposed. By setting two system tuning parameters, for the process noise covariance adaptation loop and the measurement noise covariance adaptation loop, covariance matching is sped up and achieves an effective suppression of filter divergence. The vehicle-mounted experimental results show that the mean square error of the combined attitude error obtained based on the fusion filtering method proposed in this paper is better than 0.5°, and the mean square error of the heading error is better than 1.5°. The results can provide technical support for the continuous extraction of low-cost attitude information from mobile platforms.

show abstract

“…These networks are also suitable for fitting and prediction of the time series. By considering the error accumulation of GRU network prediction, a hybrid algorithm based on the GRU and an adaptive Kalman filter was proposed in [28] to improve the navigation performance. To improve the position and velocity accuracy of the navigation system during GNSS outages, a new method was proposed in [29] that combined an unscented Kalman filter with the NARX, and the performance of the proposed method was validated experimentally using a real-world dataset.…”

Section: Introductionmentioning

confidence: 99%

Seamless Micro-Electro-Mechanical System-Inertial Navigation System/Polarization Compass Navigation Method with Data and Model Dual-Driven Approach

Zhao,

Shen,

Cao

et al. 2024

Micromachines

View full text Add to dashboard Cite

The integration of micro-electro-mechanical system–inertial navigation systems (MEMS-INSs) with other autonomous navigation sensors, such as polarization compasses (PCs) and geomagnetic compasses, has been widely used to improve the navigation accuracy and reliability of vehicles in Internet of Things (IoT) applications. However, a MEMS-INS/PC integrated navigation system suffers from cumulative errors and time-varying measurement noise covariance in unknown, complex occlusion, and dynamic environments. To overcome these problems and improve the integrated navigation system’s performance, a dual data- and model-driven MEMS-INS/PC seamless navigation method is proposed. This system uses a nonlinear autoregressive neural network (NARX) based on the Gauss–Newton Bayesian regularization training algorithm to model the relationship between the MEMS-INS outputs composed of the specific force and angular velocity data and the PC heading’s angular increment, and to fit the integrated navigation system’s dynamic characteristics, thus realizing data-driven operation. In the model-driven part, a nonlinear MEMS-INS/PC loosely coupled navigation model is established, the variational Bayesian method is used to estimate the time-varying measurement noise covariance, and the cubature Kalman filter method is then used to solve the nonlinear problem in the model. The robustness and effectiveness of the proposed method are verified experimentally. The experimental results show that the proposed method can provide high-precision heading information stably in complex, occluded, and dynamic environments.

show abstract

A GRU and AKF-Based Hybrid Algorithm for Improving INS/GNSS Navigation Accuracy during GNSS Outage

Cited by 30 publications

References 32 publications

ConvGRU-MHM: a CNN GRU-enhanced MHM for mitigating GNSS multipath

ConvGRU-MHM: a CNN GRU-enhanced MHM for mitigating GNSS multipath

An Adaptive Fusion Attitude and Heading Measurement Method of MEMS/GNSS Based on Covariance Matching

Seamless Micro-Electro-Mechanical System-Inertial Navigation System/Polarization Compass Navigation Method with Data and Model Dual-Driven Approach

Contact Info

Product

Resources

About