Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Virtual Commissioning of Distributed Systems in the Industrial Internet of Things.

Sensors (Basel, Switzerland)·2023
Same author

Fatigue Life Modelling of Steel Suspension Coil Springs Based on Wavelet Vibration Features Using Neuro-Fuzzy Methods.

Materials (Basel, Switzerland)·2023
Same author

Deep Reinforcement Learning Multi-Agent System for Resource Allocation in Industrial Internet of Things.

Sensors (Basel, Switzerland)·2022
Same author

Ensuring the Reliability of Virtual Sensors Based on Artificial Intelligence within Vehicle Dynamics Control Systems.

Sensors (Basel, Switzerland)·2022

Related Experiment Video

Updated: Oct 8, 2025

A Fully Automated and Highly Versatile System for Testing Multi-cognitive Functions and Recording Neuronal Activities in Rodents
09:13

A Fully Automated and Highly Versatile System for Testing Multi-cognitive Functions and Recording Neuronal Activities in Rodents

Published on: May 3, 2012

14.5K

Hyperparameter Optimization Techniques for Designing Software Sensors Based on Artificial Neural Networks.

Sebastian Blume1, Tim Benedens1, Dieter Schramm1

  • 1Department of Mechanical and Process Engineering, Institute for Mechatronics and System Dynamics, University of Duisburg-Essen, 47057 Duisburg, Germany.

Sensors (Basel, Switzerland)
|December 28, 2021
PubMed
Summary
This summary is machine-generated.

Knowledge-based hyperparameter optimization methods, like the Genetic Algorithm, significantly improve artificial neural network performance for software sensors in vehicle development. These methods outperform random search techniques in designing accurate roll angle estimators.

Keywords:
artificial neural networkshyperparameter optimizationintelligent transportationsoftware sensors

More Related Videos

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
11:18

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Published on: March 2, 2015

10.5K
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.3K

Related Experiment Videos

Last Updated: Oct 8, 2025

A Fully Automated and Highly Versatile System for Testing Multi-cognitive Functions and Recording Neuronal Activities in Rodents
09:13

A Fully Automated and Highly Versatile System for Testing Multi-cognitive Functions and Recording Neuronal Activities in Rodents

Published on: May 3, 2012

14.5K
Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
11:18

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Published on: March 2, 2015

10.5K
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.3K

Area of Science:

  • Automotive Engineering
  • Machine Learning
  • Software Sensing

Background:

  • Software sensors are crucial in modern vehicle development, utilizing data-driven modeling without requiring prior system knowledge.
  • Hyperparameters critically influence the performance of machine learning models, affecting both architecture and training.
  • Accurate estimation of vehicle dynamics, such as roll angle, is essential for safety and control systems.

Purpose of the Study:

  • To compare the effectiveness of different hyperparameter optimization methods for designing an artificial neural network-based roll angle estimator.
  • To evaluate random search and knowledge-based optimization techniques using simulation data from standardized driving maneuvers.

Main Methods:

  • Implemented and compared four hyperparameter optimization methods: Random Search, Hyperband, Bayesian Optimization, and Genetic Algorithm.
  • Utilized a pre-generated simulation dataset based on ISO standard driving maneuvers for training and validation.
  • Employed k-fold cross-validation to ensure robust model evaluation and integrated root mean square error as the objective function.

Main Results:

  • Knowledge-based optimization methods (Bayesian Optimization and Genetic Algorithm) demonstrated superior performance compared to purely random methods.
  • The Genetic Algorithm yielded particularly promising results for the roll angle estimation task.
  • Hyperparameter optimization significantly impacts the quality and accuracy of the data-driven soft sensor model.

Conclusions:

  • Knowledge-based hyperparameter optimization is more effective than random search for designing accurate artificial neural network models in vehicle applications.
  • The Genetic Algorithm presents a highly effective approach for optimizing hyperparameters in soft sensor design for vehicle dynamics estimation.
  • Further research into advanced optimization techniques can enhance the reliability and performance of software sensors in automotive systems.