Feasibility of a Neural Network-Based Virtual Sensor for Vehicle Unsprung Mass Relative Velocity Estimation

With the automotive industry moving towards automated driving, sensing is increasingly important in enabling technology. The virtual sensors allow data fusion from various vehicle sensors and provide a prediction for measurement that is hard or too expensive to measure in another way or in the case...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Eldar Šabanovič, Paulius Kojis, Šarūnas Šukevičius, Barys Shyrokau, Valentin Ivanov, Miguel Dhaens, Viktor Skrickij
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
Acceso en línea:https://doaj.org/article/457fd2d432384d908f258e36fc565764
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:457fd2d432384d908f258e36fc565764
record_format dspace
spelling oai:doaj.org-article:457fd2d432384d908f258e36fc5657642021-11-11T19:08:18ZFeasibility of a Neural Network-Based Virtual Sensor for Vehicle Unsprung Mass Relative Velocity Estimation10.3390/s212171391424-8220https://doaj.org/article/457fd2d432384d908f258e36fc5657642021-10-01T00:00:00Zhttps://www.mdpi.com/1424-8220/21/21/7139https://doaj.org/toc/1424-8220With the automotive industry moving towards automated driving, sensing is increasingly important in enabling technology. The virtual sensors allow data fusion from various vehicle sensors and provide a prediction for measurement that is hard or too expensive to measure in another way or in the case of demand on continuous detection. In this paper, virtual sensing is discussed for the case of vehicle suspension control, where information about the relative velocity of the unsprung mass for each vehicle corner is required. The corresponding goal can be identified as a regression task with multi-input sequence input. The hypothesis is that the state-of-art method of Bidirectional Long–Short Term Memory (BiLSTM) can solve it. In this paper, a virtual sensor has been proposed and developed by training a neural network model. The simulations have been performed using an experimentally validated full vehicle model in IPG Carmaker. Simulations provided the reference data which were used for Neural Network (NN) training. The extensive dataset covering 26 scenarios has been used to obtain training, validation and testing data. The Bayesian Search was used to select the best neural network structure using root mean square error as a metric. The best network is made of 167 BiLSTM, 256 fully connected hidden units and 4 output units. Error histograms and spectral analysis of the predicted signal compared to the reference signal are presented. The results demonstrate the good applicability of neural network-based virtual sensors to estimate vehicle unsprung mass relative velocity.Eldar ŠabanovičPaulius KojisŠarūnas ŠukevičiusBarys ShyrokauValentin IvanovMiguel DhaensViktor SkrickijMDPI AGarticlevirtual sensorautomotive controlactive suspensionvehicle state estimationneural networksdeep learningChemical technologyTP1-1185ENSensors, Vol 21, Iss 7139, p 7139 (2021)
institution DOAJ
collection DOAJ
language EN
topic virtual sensor
automotive control
active suspension
vehicle state estimation
neural networks
deep learning
Chemical technology
TP1-1185
spellingShingle virtual sensor
automotive control
active suspension
vehicle state estimation
neural networks
deep learning
Chemical technology
TP1-1185
Eldar Šabanovič
Paulius Kojis
Šarūnas Šukevičius
Barys Shyrokau
Valentin Ivanov
Miguel Dhaens
Viktor Skrickij
Feasibility of a Neural Network-Based Virtual Sensor for Vehicle Unsprung Mass Relative Velocity Estimation
description With the automotive industry moving towards automated driving, sensing is increasingly important in enabling technology. The virtual sensors allow data fusion from various vehicle sensors and provide a prediction for measurement that is hard or too expensive to measure in another way or in the case of demand on continuous detection. In this paper, virtual sensing is discussed for the case of vehicle suspension control, where information about the relative velocity of the unsprung mass for each vehicle corner is required. The corresponding goal can be identified as a regression task with multi-input sequence input. The hypothesis is that the state-of-art method of Bidirectional Long–Short Term Memory (BiLSTM) can solve it. In this paper, a virtual sensor has been proposed and developed by training a neural network model. The simulations have been performed using an experimentally validated full vehicle model in IPG Carmaker. Simulations provided the reference data which were used for Neural Network (NN) training. The extensive dataset covering 26 scenarios has been used to obtain training, validation and testing data. The Bayesian Search was used to select the best neural network structure using root mean square error as a metric. The best network is made of 167 BiLSTM, 256 fully connected hidden units and 4 output units. Error histograms and spectral analysis of the predicted signal compared to the reference signal are presented. The results demonstrate the good applicability of neural network-based virtual sensors to estimate vehicle unsprung mass relative velocity.
format article
author Eldar Šabanovič
Paulius Kojis
Šarūnas Šukevičius
Barys Shyrokau
Valentin Ivanov
Miguel Dhaens
Viktor Skrickij
author_facet Eldar Šabanovič
Paulius Kojis
Šarūnas Šukevičius
Barys Shyrokau
Valentin Ivanov
Miguel Dhaens
Viktor Skrickij
author_sort Eldar Šabanovič
title Feasibility of a Neural Network-Based Virtual Sensor for Vehicle Unsprung Mass Relative Velocity Estimation
title_short Feasibility of a Neural Network-Based Virtual Sensor for Vehicle Unsprung Mass Relative Velocity Estimation
title_full Feasibility of a Neural Network-Based Virtual Sensor for Vehicle Unsprung Mass Relative Velocity Estimation
title_fullStr Feasibility of a Neural Network-Based Virtual Sensor for Vehicle Unsprung Mass Relative Velocity Estimation
title_full_unstemmed Feasibility of a Neural Network-Based Virtual Sensor for Vehicle Unsprung Mass Relative Velocity Estimation
title_sort feasibility of a neural network-based virtual sensor for vehicle unsprung mass relative velocity estimation
publisher MDPI AG
publishDate 2021
url https://doaj.org/article/457fd2d432384d908f258e36fc565764
work_keys_str_mv AT eldarsabanovic feasibilityofaneuralnetworkbasedvirtualsensorforvehicleunsprungmassrelativevelocityestimation
AT pauliuskojis feasibilityofaneuralnetworkbasedvirtualsensorforvehicleunsprungmassrelativevelocityestimation
AT sarunassukevicius feasibilityofaneuralnetworkbasedvirtualsensorforvehicleunsprungmassrelativevelocityestimation
AT barysshyrokau feasibilityofaneuralnetworkbasedvirtualsensorforvehicleunsprungmassrelativevelocityestimation
AT valentinivanov feasibilityofaneuralnetworkbasedvirtualsensorforvehicleunsprungmassrelativevelocityestimation
AT migueldhaens feasibilityofaneuralnetworkbasedvirtualsensorforvehicleunsprungmassrelativevelocityestimation
AT viktorskrickij feasibilityofaneuralnetworkbasedvirtualsensorforvehicleunsprungmassrelativevelocityestimation
_version_ 1718431580178874368