Performance comparison of HONNs and FFNNs in GPS and INS integration for vehicular navigation

Malleswaran, M.; Vaidehi, V.; Manjula, S.; Deborah, S. Angel

doi:10.1109/icrtit.2011.5972273

Cited by 8 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The method of neural network can ignore specific physical parameters of the process and system and the complex nonlinear mapping of input and output can be realized through studying the training samples. The method, with good generalization ability, is widely used in the modeling of nonlinear systems [17,18]. The RBF neural network, a forward net with good capability and global best approaching performance as well as a fast and easy training method, does not have the problem of local optima [19].…”

Section: Modeling Methodsmentioning

confidence: 99%

Dynamic angular velocity modeling and error compensation of one-fiber fiber optic gyroscope (OFFOG) in the whole temperature range

Zhang¹,

Wang²,

Yang³

et al. 2012

Meas. Sci. Technol.

View full text Add to dashboard Cite

Dynamic angular velocity modeling and error compensation of VG095M in the whole temperature range, based on a radial basis function (RBF) neural network, is presented in this paper. With gyro output voltage and environmental temperature as the input and angular velocity as the output, an RBF neural network model is established. The model is trained and validated by the experiment data. The fitting error of the model is 4.3818 × 10−6 deg s−1, which shows that the model has high precision. The experiment data except the data used for modeling were processed with this model. The results show that the maximum, minimum and mean square error of the angular velocity were reduced to 4.6%, 4.3% and 4.7% respectively after compensation.

show abstract

Section: Modeling Methodsmentioning

confidence: 99%

Dynamic angular velocity modeling and error compensation of one-fiber fiber optic gyroscope (OFFOG) in the whole temperature range

Zhang¹,

Wang²,

Yang³

et al. 2012

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…GPS and INS data integration has been performed using Radial Basis Function Neural Network, Back Propagation Neural Network and Fuzzy system. Radial Basis Function Neural Network (RBF-NN) generally has simpler architecture and faster training procedure than multi-layer perceptron neural networks ( [1], [2], [6], [8], [10]). Though it has simple architecture and faster training procedure, it only has fixed topology, so it lacks dynamicity.…”

Section: Existing Ins/gps Data Fusion Techniquesmentioning

confidence: 99%

Analysis of Different Associative Memory Neural Network for GPS/INS Data Fusion

Deborah

2012

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

Abstract. Aircraft navigation relies mainly on Global Positioning System (GPS) to provide accurate position values consistently. However, GPS receivers may encounter frequent GPS outages within urban areas where satellite signals are blocked. To overcome this drawback generally GPS is integrated with inertial sensors mounted inside the vehicle to provide a reliable navigation solution. Inertial Navigation System (INS) and GPS are commonly integrated using a Kalman filter (KF) to provide a robust navigation solution, overcoming situations of GPS satellite signals blockage. This work presents New Position Update Architecture (NPUA) for GPS and INS data integration. The NPUA has an Artificial Neural Network (ANN) block that uses Associative memoy Neural

show abstract

“…Deep learning is a data-driven approach that can learn complex nonlinear properties and uncertainty from data. Researchers have proposed deep learning-based techniques to discover the error drift characteristics of IMUs over time for better GPS/INS combined navigation solutions [17][18][19][20]. Several deep learning techniques have also been applied to wheel odometry: an LSTM model was proposed to learn the uncertainty of wheel odometry in [21], and a WhONet model was proposed in [22], which further improved the performance compared to the LSTM model.…”

Section: Introductionmentioning

confidence: 99%

Wheel Odometry with Deep Learning-Based Error Prediction Model for Vehicle Localization

Ding

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Wheel odometry is a simple and low-cost localization technique that can be used for localization in GNSS-deprived environments; however, its measurement accuracy is affected by many factors, such as wheel slip, wear, and tire pressure changes, resulting in unpredictable and variable errors, which in turn affect positioning performance. To improve the localization performance of wheel odometry, this study developed a wheel odometry error prediction model based on a transformer neural network to learn the measurement uncertainty of wheel odometry and accurately predict the odometry error. Driving condition characteristics including features describing road types, road conditions, and vehicle driving operations were considered, and models both with and without driving condition characteristics were compared and analyzed. Tests were performed on a public dataset and an experimental vehicle. The experimental results demonstrate that the proposed model can predict the odometry error with higher accuracy, stability, and reliability than the LSTM and WhONet models under multiple challenging and longer GNSS outage driving conditions. At the same time, the transformer model’s overall performance can be improved in longer GNSS outage driving conditions by considering the driving condition characteristics. Tests on the experimental vehicle demonstrate the model’s generalization capability and the improved positioning performance of dead reckoning when using the proposed model. This study explored the possibility of applying a transformer model to wheel odometry and provides a new solution for using deep learning in localization.

show abstract

Performance comparison of HONNs and FFNNs in GPS and INS integration for vehicular navigation

Cited by 8 publications

References 6 publications

Dynamic angular velocity modeling and error compensation of one-fiber fiber optic gyroscope (OFFOG) in the whole temperature range

Dynamic angular velocity modeling and error compensation of one-fiber fiber optic gyroscope (OFFOG) in the whole temperature range

Analysis of Different Associative Memory Neural Network for GPS/INS Data Fusion

Wheel Odometry with Deep Learning-Based Error Prediction Model for Vehicle Localization

Contact Info

Product

Resources

About