INS/GNSS integration using recurrent fuzzy wavelet neural networks
TL;DR: A comparison of test results shows that the proposed NN algorithm could efficiently provide high-accuracy corrections on the INS velocity and position information during GNSS outages.
Abstract: In recent years, aided navigation systems through combining inertial navigation system (INS) with global navigation satellite system (GNSS) have been widely applied to enhance the position, velocity, and attitude information of autonomous vehicles. In order to gain the accuracy of the aided INS/GNSS in GNSS gap intervals, a heuristic neural network structure based on the recurrent fuzzy wavelet neural network (RFWNN) is applicable for INS velocity and position error compensation purpose. During frequent access to GNSS data, the RFWNN should be trained as a highly precise prediction model equipped with the Kalman filter algorithm. Therefore, the INS velocity and position error data are obtainable along with the lost intervals of GNSS signals. For performance assessment of the proposed RFWNN-aided INS/GNSS, real flight test data of a small commercial unmanned aerial vehicle (UAV) were conducted. A comparison of test results shows that the proposed NN algorithm could efficiently provide high-accuracy corrections on the INS velocity and position information during GNSS outages.
Citations
More filters
TL;DR: A novel position estimation system based on learning to the prediction model to address the above challenges and it is observed that the proposed Kalman filter with learning module performed better than the traditionalKalman filter algorithm in terms of root mean square error metric.
Abstract: Internet of Things is advancing, and the augmented role of smart navigation in automating processes is at its vanguard. Smart navigation and location tracking systems are finding increasing use in the area of the mission-critical indoor scenario, logistics, medicine, and security. A demanding emerging area is an Indoor Localization due to the increased fascination towards location-based services. Numerous inertial assessments unit-based indoor localization mechanisms have been suggested in this regard. However, these methods have many shortcomings pertaining to accuracy and consistency. In this study, we propose a novel position estimation system based on learning to the prediction model to address the above challenges. The designed system consists of two modules; learning to prediction module and position estimation using sensor fusion in an indoor environment. The prediction algorithm is attached to the learning module. Moreover, the learning module continuously controls, observes, and enhances the efficiency of the prediction algorithm by evaluating the output and taking into account the exogenous factors that may have an impact on its outcome. On top of that, we reckon a situation where the prediction algorithm can be applied to anticipate the accurate gyroscope and accelerometer reading from the noisy sensor readings. In the designed system, we consider a scenario where the learning module, based on Artificial Neural Network, and Kalman filter are used as a prediction algorithm to predict the actual accelerometer and gyroscope reading from the noisy sensor reading. Moreover, to acquire data, we use the next-generation inertial measurement unit, which contains a 3-axis accelerometer and gyroscope data. Finally, for the performance and accuracy of the proposed system, we carried out numbers of experiments, and we observed that the proposed Kalman filter with learning module performed better than the traditional Kalman filter algorithm in terms of root mean square error metric.
47 citations
Cites background from "INS/GNSS integration using recurren..."
...From the last several decades, ’ many indoor studies have been introduced, which uses the machine learning approaches to predict and track the location of the object in an indoor environment [38]....
[...]
TL;DR: In this article, a fuzzy logic controller is designed for the strong tracking cubature Kalman filter (STCKF), which aims at strengthening the filter's ability to identify and respond to the dynamics.
Abstract: To further enhance the positioning accuracy and stability of the INS/GNSS integrated navigation system, we present a new fuzzy strong tracking cubature Kalman filter (FSTCKF) algorithm for data fusion. A fuzzy logic controller is designed for the strong tracking cubature Kalman filter (STCKF), which aims at strengthening the filter’s ability to identify and respond to the dynamics. Chi-square tests are separately conducted on the innovation vector in order to reveal the dynamic properties inside the velocity and position states. Thereafter, parallel fuzzy inferences are conducted to generate a time-varying smoothing factor matrix, which helps the STCKF obtain the multiple fading factors distributed to each state variable. Numerical simulations and real data testing results demonstrate the superiority and robustness of the proposed FSTCKF algorithm. Not only can the proposed algorithm maintain the accuracy and stability in steady conditions, but further increase the dynamic tracking ability as well. Finally, the positioning performances of the INS/GNSS can be improved.
10 citations
TL;DR: It is concluded that the adaptive estimation theory is an effective complement to neural network-aided navigation, as the GRU- aided AKF reduced the horizontal error of GRU -aided KF by 31.71% and 16.12% after 180 and 120s of GNSS outage, respectively.
Abstract: The integrated navigation system consisting of an inertial navigation system (INS) and Global Navigation Satellite System (GNSS) provides continuous high-accuracy positioning whereas the navigation accuracy during a GNSS outage inevitably degrades owing to INS error divergence. To reduce such degradation, a gated recurrent unit (GRU) and adaptive Kalman filter (AKF)-based hybrid algorithm is proposed. The GRU network, which has advantages of high accuracy and efficiency, is constructed to predict the position variations during GNSS outage. Furthermore, this paper takes the GRU-predicted error accumulation into consideration, and introduces AKF as a supplementary methodology to improve the navigation performance. The proposed hybrid algorithm is trained and tested by practical road datasets and compared with four algorithms, including the standard KF, Multi-Layer Perceptron (MLP)-aided KF, Long Short Time Memory (LSTM) aided KF, and GRU-aided KF. Periods of 180 and 120 s GNSS outage are employed to test the performance of the proposed algorithm in different time scales. The comparison result between the standard KF and neural network-aided KF indicates that the neural network is an effective methodology for bridging GNSS outages. The performance comparison between three kinds of neural networks demonstrate that both recurrent neural networks surpass the MLP in prediction position variation, and the GRU transcends the LSTM in prediction accuracy and training efficiency. Furthermore, it is concluded that the adaptive estimation theory is an effective complement to neural network-aided navigation, as the GRU-aided AKF reduced the horizontal error of GRU-aided KF by 31.71% and 16.12% after 180 and 120s of GNSS outage, respectively.
8 citations
TL;DR: A novel hybrid algorithm based on Gated Recurrent Unit (GRU) and interacting multiple model adaptive robust cubature Kalman filter (IMM-ARCKF) is proposed to solve the uncertainty of system model and measurement noise statics in the application of INS/GPS on the road.
Abstract: In order to ensure that Inertial Navigation System/Global Positioning System integrated navigation system (INS/GPS) can still provide high precision positioning results when GPS outages, a novel hybrid algorithm based on Gated Recurrent Unit (GRU) and interacting multiple model adaptive robust cubature Kalman filter (IMM-ARCKF) is proposed. Firstly, the IMM-ARCKF algorithm is proposed to solve the uncertainty of system model and measurement noise statics in the application of INS/GPS on the road. Then, GRU neural network is introduced into INS/GPS system which includes two modes of training and prediction. When GPS signal can be received, the GRU neural network works in the training mode. When GPS outages, the GRU neural network predicts the GPS position increment. Finally, the effectiveness of the algorithm is evaluated by the experiment and analysis. From the data of the experiment, the proposed algorithm can improve the positioning accuracy during GPS outages.
8 citations
Cites methods from "INS/GNSS integration using recurren..."
...In reference [26], to improve the navigation accuracy, the regression fuzzy wavelet neural network (RFWNN) was added to the INS/GPS system to compensate the speed and position errors of INS/GPS....
[...]
TL;DR: A novel data-driven modelling methodology for lateral stability description of articulated steering vehicles and the experimental results show that the output of the RNN model can accurately quantify lateral stability of the vehicle, i.e., the risk of rollover, when it is turning and crossing uneven surfaces or obstacles.
Abstract: Articulated steering vehicles have outstanding capability operating but suffer from frequent rollover accidents due to their complicated structure. It is necessary to accurately detect their rollover risk for drivers to take action in time. Their variable structure and the variable center of mass exhibit nonlinear time-variant behavior and increase the difficulty of dynamic modelling and lateral stability description. This paper proposes a novel data-driven modelling methodology for lateral stability description of articulated steering vehicles. The running data is first collected based on the typical operations that prone to rollover and then classified into two types: Safety and danger. The data quality is further improved by wavelet transformation. Finally, an RNN model is built on the data. The experimental results show that the output of the RNN model can accurately quantify lateral stability of the vehicle, i.e., the risk of rollover, when it is turning and crossing uneven surfaces or obstacles.
7 citations
References
More filters
Book•
29 Dec 2000
TL;DR: The authors explore the various subtleties, common failures, and inherent limitations of the theory as it applies to real-world situations, and provide numerous detailed application examples and practice problems, including GNSS-aided INS, modeling of gyros and accelerometers, and SBAS and GBAS.
Abstract: An updated guide to GNSS and INS, and solutions to real-world GPS/INS problems with Kalman filtering Written by recognized authorities in the field, this second edition of a landmark work provides engineers, computer scientists, and others with a working familiarity with the theory and contemporary applications of Global Navigation Satellite Systems (GNSS), Inertial Navigational Systems (INS), and Kalman filters. Throughout, the focus is on solving real-world problems, with an emphasis on the effective use of state-of-the-art integration techniques for those systems, especially the application of Kalman filtering. To that end, the authors explore the various subtleties, common failures, and inherent limitations of the theory as it applies to real-world situations, and provide numerous detailed application examples and practice problems, including GNSS-aided INS, modeling of gyros and accelerometers, and SBAS and GBAS. Drawing upon their many years of experience with GNSS, INS, and the Kalman filter, the authors present numerous design and implementation techniques not found in other professional references. This Second Edition has been updated to include: GNSS signal integrity with SBAS Mitigation of multipath, including results Ionospheric delay estimation with Kalman filters New MATLAB programs for satellite position determination using almanac and ephemeris data and ionospheric delay calculations from single and dual frequency data New algorithms for GEO with L1 /L5 frequencies and clock steering Implementation of mechanization equations in numerically stable algorithms To enhance comprehension of the subjects covered, the authors have included software in MATLAB, demonstrating the working of the GNSS, INS, and filter algorithms. In addition to showing the Kalman filter in action, the software also demonstrates various practical aspects of finite word length arithmetic and the need for alternative algorithms to preserve result accuracy.
1,650 citations
"INS/GNSS integration using recurren..." refers background in this paper
...Because of the high update rate of inertial sensors measurement and its autonomous data production, the INS accelerometer bias and gyroscopes drift should lead to the accumulation of navigation output errors over time (Mohinder et al. 2001)....
[...]
TL;DR: Algorithms for wavelet network construction are proposed for the purpose of nonparametric regression estimation and particular attentions are paid to sparse training data so that problems of large dimension can be better handled.
Abstract: Wavelet networks are a class of neural networks consisting of wavelets. In this paper, algorithms for wavelet network construction are proposed for the purpose of nonparametric regression estimation. Particular attentions are paid to sparse training data so that problems of large dimension can be better handled. A numerical example on nonlinear system identification is presented for illustration.
760 citations
"INS/GNSS integration using recurren..." refers methods in this paper
...These activation functions are obtained from a Mexican hat mother wavelet (x) , with a translation of t and a dilation of d (Zhang 1997): where ‖x‖2 = xTx ....
[...]
Book•
17 Jan 2021
TL;DR: This text/CD-ROM presents elements of basic mathematics, kinematics, equations describing navigation systems and their error models, and Kalman filtering.
Abstract: Intended for those directly involved with the design, integration, and test and evaluation of navigation systems, this text/CD-ROM presents elements of basic mathematics, kinematics, equations describing navigation systems and their error models, and Kalman filtering. Detailed derivations are presen
599 citations
"INS/GNSS integration using recurren..." refers background in this paper
...Therefore, the governing dynamics of the inertial position, velocity, and attitude coordinates are representable as (Rogers 2007):...
[...]
...According to Poisson’s equation, the estimation of orientation matrix Ĉ b is related to the true orientation matrix C b , and the misalignment angles (Rogers 2007):...
[...]
...RN and RE are the meridian radii of curvature and transverse (Rogers 2007)....
[...]
TL;DR: This study suggests the use of Input-Delayed Neural Networks (IDNN) to model both the INS position and velocity errors based on current and some past samples of INS location and velocity, respectively, which results in a more reliable positioning solution during long GPS outages.
208 citations
TL;DR: This paper focuses on the design of a dynamic Petri recurrent fuzzy neural network (DPRFNN), and this network structure is applied to the path-tracking control of a nonholonomic mobile robot for verifying its validity.
Abstract: This paper focuses on the design of a dynamic Petri recurrent fuzzy neural network (DPRFNN), and this network structure is applied to the path-tracking control of a nonholonomic mobile robot for verifying its validity. In the DPRFNN, the concept of a Petri net and the recurrent frame of internal-feedback loops are incorporated into a traditional FNN to alleviate the computation burden of parameter learning and to enhance the dynamic mapping of network ability. Moreover, the supervised gradient-descent method is used to develop the online-training algorithm for the DPRFNN control. In order to guarantee the convergence of path-tracking errors, analytical methods based on a discrete-type Lyapunov function are proposed to determine varied learning rates for DPRFNN. In addition, the effectiveness of the proposed DPRFNN control scheme under different moving paths is verified by experimental results, and its superiority is indicated in comparison with FNN, RFNN, Petri FNN, and PRFNN control systems.
147 citations