Estimation of Vehicle Yaw Rate Using a Virtual Sensor with a Speed Scheduled Observer

Kahraman, Kerim; Emirler, Mümin Tolga; Centürk, Mutlu; Güvenç, Bilin Aksun; Güvenç, Levent; Efendioglu, Baris

doi:10.3182/20100712-3-de-2013.00043

Cited by 5 publications

(3 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Virtual sensing is intensively used for automotive applications. For example, for passenger thermal comfort, tire pressure monitoring system, powertrain applications, sprung mass state estimation [15,16] and others. With AD development, the role of VS will only increase as power-efficient embedded computing devices will be available.…”

Section: Introductionmentioning

confidence: 99%

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

Šabanovič¹,

Kojis²,

Šukevičius³

et al. 2021

Preprint

View full text Add to dashboard Cite

With the automotive industry moving towards automated driving, sensing is becoming an increasingly important part of 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 (BiLSMT) 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 was 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 for the estimation of vehicle unsprung mass relative velocity.

show abstract

Section: Introductionmentioning

confidence: 99%

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

Šabanovič¹,

Kojis²,

Šukevičius³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Linea midsized sedan. Three Fiat Linea midsized sedans were used by the authors and their colleagues in lateral dynamics testing [12], semiautonomous driving in a platoon [2], and autonomous path following experiments [13], respectively. The authors, therefore, have considerable experimental experience with this vehicle.…”

Section: Experimental Vehicle the Experimental Vehicle Is A Fiatmentioning

confidence: 99%

Robust PID Steering Control in Parameter Space for Highly Automated Driving

Emirler

Uygan

Güvenç

et al. 2014

International Journal of Vehicular Technology

Self Cite

View full text Add to dashboard Cite

This paper is on the design of a parameter space based robust PID steering controller. This controller is used for automated steering in automated path following of a midsized sedan. Linear and nonlinear models of this midsized sedan are presented in the paper. Experimental results are used to validate the longitudinal and lateral dynamic models of this vehicle. This paper is on automated steering control and concentrates on the lateral direction of motion. The linear model is used to design a PID steering controller in parameter space that satisfies D-stability. The PID steering controller that is designed is used in a simulation study to illustrate the effectiveness of the proposed method. Simulation results for a circular trajectory and for a curved trajectory are presented and discussed in detail. This study is part of a larger research effort aimed at implementing highly automated driving in a midsized sedan.

show abstract

“…Virtual sensing is intensively used for automotive applications, for example, for passenger thermal comfort, the tire pressure monitoring system, powertrain applications, sprung mass state estimation [19,20] and others. VS may be a key technology of advanced control algorithms, enabling customisation of vehicle-human interaction.…”

Section: Introductionmentioning

confidence: 99%

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

Šabanovič

Kojis

Šukevičius

et al. 2021

Sensors

View full text Add to dashboard Cite

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.

show abstract

Estimation of Vehicle Yaw Rate Using a Virtual Sensor with a Speed Scheduled Observer

Cited by 5 publications

References 8 publications

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

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

Robust PID Steering Control in Parameter Space for Highly Automated Driving

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

Contact Info

Product

Resources

About