Optimize Vehicle Design with AI and Simscape

Workflow for generating a surrogate AI model from a multibody vehicle dynamics model.

https://github.com/simscape/Optimize-Vehicle-Design-with-AI-and-Simscape

Ahora está siguiendo esta publicación

This example shows the workflow to create a surrogate AI model using training data from a multibody model of a vehicle. The resulting AI model can be used for design space exploration and for finding the optimal design parameters.
  • Early-stage physical physical design is supported by creating a reduced order model to rapidly evaluate hardpoint locations.
  • Sensitivity analysis is supported by running many simulations in parallel and analyzing the influence of design parameters on performance metrics
  • Training data for the AI model is produced using Design of Experiments to ensure the entire design spaces is covered.
  • Machine Learning and Deep Learning are both used to create surrogate models that are automatically validated against the generated data.
  • Optimization algorithms are used to identify the set of design parameters that balance the tradeoff between multiple performance metrics.
  • A MATLAB App enables exploration of the design space using responses surfaces.
Open the project file SSVT_Susp_Opt.prj to get started.
Use the "Download" button above to get files compatible with the latest release of MATLAB.
For earlier MATLAB releases, use "Version History" tab above or these links:
Vehicle Model
Simscape Multibody is used to model the vehicle. The multibody model has 94 parameters defining the front and rear suspensions which can be tuned. This includes hardpoint locations, spring stiffnesses, and damping coefficients. The parameter values can be varied without recompiling the model so that parameter sweeps can be run as efficiently as possible.
Workflow
Please visit the GitHub repository to see the steps in the workflow.
This model was extracted from Simscape Vehicle Templates, a more elaborate repository for vehicle modeling.
Learn how to use Simscape with
Product Capabilities:

Citar como

Steve Miller (2026). Optimize Vehicle Design with AI and Simscape (https://github.com/simscape/Optimize-Vehicle-Design-with-AI-and-Simscape/releases/tag/26.1.1.1), GitHub. Recuperado .

Información general

Compatibilidad con la versión de MATLAB

  • Compatible con cualquier versión desde R2024b hasta R2026a

Compatibilidad con las plataformas

  • Windows
  • macOS
  • Linux
Versión Publicado Notas de la versión Action
26.1.1.1

See release notes for this release on GitHub: https://github.com/simscape/Optimize-Vehicle-Design-with-AI-and-Simscape/releases/tag/26.1.1.1

25.2.1.1

See release notes for this release on GitHub: https://github.com/simscape/Optimize-Vehicle-Design-with-AI-and-Simscape/releases/tag/25.2.1.1

25.1.1.1

See release notes for this release on GitHub: https://github.com/simscape/Optimize-Vehicle-Design-with-AI-and-Simscape/releases/tag/25.1.1.1

24.2.1.1

See release notes for this release on GitHub: https://github.com/simscape/Optimize-Vehicle-Design-with-AI-and-Simscape/releases/tag/24.2.1.1

24.2.1.0

Para consultar o notificar algún problema sobre este complemento de GitHub, visite el repositorio de GitHub.
Para consultar o notificar algún problema sobre este complemento de GitHub, visite el repositorio de GitHub.