Geomagnetic Vector Pattern Recognition Navigation Method Based on Probabilistic Neural Network

Chen, Zhuo; Liu, Kunjia; Zhang, Qi; Liu, Zhongyan; Chen, Dixiang; Pan, Mengchun; Hu, Jiafei; Xu, Yujing

doi:10.1109/tgrs.2023.3273552

Cited by 5 publications

(7 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The TA-LSTM penalizes the weights for abnormal elements in an input time-series in case of anomalies, thus providing a robust heading angle prediction. To determine the existence of anomalies, the predicted heading angle by TA-LSTM is compared with the calculated heading angle by (7). The final heading angle to be leveraged by the navigation is determined based on the calibration between the predicted and calculated heading angle.…”

Section: The Proposed Underwater Navigation With Anomaly Resistancementioning

confidence: 99%

“…The ground-truth for the TA-LSTM output during the traing is also provided by the INS. After the training, the TA-LSTM takes input of the declination/inclination and latitude/longitude calculated from ( 6) and (7), respectively, to predict the heading angle toward the destination. Below we detail the proposed approach regarding the description of a Long Short-Term Memory (LSTM) network, the encoding and decoding of geomagnetic information in the TA-LSTM, and the anomaly resistance for the navigation with the TA-LSTM.…”

Section: The Proposed Underwater Navigation With Anomaly Resistancementioning

confidence: 99%

“…Geophysical matching navigation approaches, including terrain matching [2], gravity matching [5], and geomagnetic matching [6], can provide strong autonomy for underwater navigation and they can work under diverse terrain and time conditions. Such approaches refers to the navigation based on comparing the similarity between the navigation trajectory and a reference map [7]. The terrain contour matching (TERCOM) algorithm [8] receives continuous navigation corrections throughout the mission, and it can reserve the positioning accuracy in cases of large initial position error and altimeter measurement noise.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Data-Driven Long-Landing Event Detection and Interpretability Analysis in Civil Aviation

Yang

Ren

et al. 2022

IEEE Access

View full text Add to dashboard Cite

Various animals exhibit accurate navigation using environment cues. The Earth's magnetic field has been proved a reliable information source in long-distance fauna migration. Inspired by animal navigation, this work proposes a bionic and data-driven approach for long-distance underwater navigation. The proposed approach uses measured geomagnetic data for the navigation, and requires no GPS systems or geographical maps. Particularly, we construct and train a Temporal Attentionbased Long Short-Term Memory (TA-LSTM) network to predict the heading angle during the navigation. To mitigate the impact of geomagnetic anomalies, we develop the mechanism to detect and quantify the anomalies based on Maximum Likelihood Estimation. We integrate the developed mechanism with the TA-LSTM, and calibrate the predicted heading angles to gain resistance against geomagnetic anomalies. Using the retrieved data from the WMM model, we conduct numerical simulations with diversified navigation conditions to test our approach. The simulation results demonstrate a resilience navigation against geomagnetic anomalies by our approach, along with precision and stability of the underwater navigation in single and multiple destination missions.

show abstract

Section: The Proposed Underwater Navigation With Anomaly Resistancementioning

confidence: 99%

Section: The Proposed Underwater Navigation With Anomaly Resistancementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Data-Driven Long-Landing Event Detection and Interpretability Analysis in Civil Aviation

Yang

Ren

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…ANNs can establish highly accurate geomagnetic databases by predicting and supplementing the existing geomagnetic information [22][23][24]. In addition, they have been proven to improve the positioning accuracy of geomagnetic matching navigation [25][26][27].…”

Section: Introductionmentioning

confidence: 99%

Application of bioinspired geomagnetic sensor measurements and geomagnetic map modeling based on neural networks in simulated navigation

Shi,

Tang,

Wang

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

A geomagnetic field is a vector field in which the strength and direction are related to geographical location. Geomagnetic navigation technology, which uses collected geomagnetic field information to achieve positioning and navigation, has the advantages of reliability, stability, accuracy, and concealment. With the deepening research on geomagnetic navigation, bioinspired geomagnetic navigation technology has also been developed, which mainly studies and imitates the magnetic sensing mechanism and navigation behavior of animals, providing new research ideas for geomagnetic navigation technology. The magnetic particle hypothesis and free radical pair hypothesis are two mainstream mechanisms of biological sensing using the geomagnetic field, and studies have shown that these two mechanisms may be coupled within organisms. In this study, we propose a bioinspired weak magnetic vector (BWMV) sensor based on the joint sensing mechanism of magnetic particles and free radicals. It consists of a magnetic rod made of soft magnetic material and a tunnel magnetoresistance (TMR) sensor array. A magnetic rod was used to simulate magnetic particles to convert magnetic field angle information into magnetic field intensity distribution information, and the TMR sensor array was used to simulate the perception of the magnetic field distribution by free radicals. In addition, artificial neural networks (ANNs) were used for BWMV sensors to obtain the mapping relationship between the magnetic field distribution and parameters, which can be used for geomagnetic navigation. To verify the navigation effect of the BWMV sensor in the laboratory, a simulated geomagnetic navigation device was built, and the high-precision mapping relationship from geomagnetic parameters to latitude and longitude information of the selected navigation area was obtained through another ANN. Finally, the effectiveness of the BWMV sensor based on ANNs for geomagnetic navigation is verified using simulated navigation experiments.

show abstract

“…So far, the commonly used measurement elements for measuring the roll angle of high-speed flight carriers (HSFCs) are the combination of global navigation satellite system (GNSS) [1][2][3], MEMS inertial navigation systems [4][5][6] (MEMS-INSs), and geomagnetic sensors [7,8]. Among them, using geomagnetic sensors to measure the roll angle has the advantages of high reliability, low cost, no accumulated error, and passive autonomy, and can be used normally under satellite rejection conditions, it has been widely used in the field of navigation [9][10][11][12][13].…”

Section: Introductionmentioning

confidence: 99%

A novel method for measuring roll angle

Zhang,

Gao,

Wang

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

The precise measurement of the spin speed of a high-speed autonomous unmanned aerial vehicle (HAS-UAV) is a key element in mastering flight stability, and the measurement of roll angle is the key to determining the accuracy of navigation control systems. We put forward a novel method for measuring roll angle. This method starts from the time-frequency domain analysis (TFDA) of the output signal of the geomagnetic sensor, extracts the time-frequency ridge of the time-frequency matrix (TFM) to obtain the spin speed of the HSA-UAV, and reconstructs the output signal of the geomagnetic sensor. It can calculate the roll angle when the calibration parameters of the geomagnetic sensor are unknown and have good engineering practical value. In addition, we also propose an improved nonlinear short-time Fourier transform (INLSTFT) with high-frequency resolution and a forward penalty dynamic path ridge-extraction method (FPDPRM) with frequency jump suppression to extract instantaneous frequency (IF) in the TFM.

show abstract

Geomagnetic Vector Pattern Recognition Navigation Method Based on Probabilistic Neural Network

Cited by 5 publications

References 16 publications

Data-Driven Long-Landing Event Detection and Interpretability Analysis in Civil Aviation

Data-Driven Long-Landing Event Detection and Interpretability Analysis in Civil Aviation

Application of bioinspired geomagnetic sensor measurements and geomagnetic map modeling based on neural networks in simulated navigation

A novel method for measuring roll angle

Contact Info

Product

Resources

About