MathWorks Automotive Conference Online 2026

Discover how generative AI and Model-Based Design can transform the way engineering teams innovate. Learn how organizations can combine trusted modeling and simulation workflows with emerging generative AI technologies to improve quality, reduce rework, and move faster from concept to validated designs.

Model-Based Design is central to automotive and embedded software development, helping engineers build and test software models in MATLAB^® and Simulink^®. Currently, much of this work requires manual, time-consuming modeling and testing efforts that depend heavily on experienced specialists. Tata Elxsi demonstrates how integrating MATLAB MCP Server with GitHub^® Copilot accelerates development.

Explore a simple but powerful idea: What if an AI assistant could actively support engineers throughout development and testing? See how to connect GitHub Copilot to MATLAB using MATLAB MCP Server, which acts as a bridge that lets the AI operate inside MATLAB—not just provide guidance. Tata Elxsi presents an end-to-end demonstration highlighting five key capabilities:

• Architecture Generation: Automatically deriving a Simulink architecture model directly from textual requirements
• Design Modeling: Generating the detailed Simulink model informed by both the requirements and the architecture
• Requirements Linking: Establishing and maintaining traceability links between requirements and model elements to support compliance and impact analysis
• MAB Compliance: Applying MathWorks Advisory Board (MAB) modeling guideline checks to ensure standards-compliant, review-ready models
• Verification: Autogenerating structured test cases in Simulink Test^™ to accelerate validation

The measurable impact includes a significant reduction in manual modeling effort, faster requirements-to-model turnaround, improved traceability, and earlier defect detection through automated test generation.

It is important to note that AI is not always right. The AI does the heavy lifting, but Model-Based Design specialists remain essential—they review, correct, and approve everything the AI produces. Think of the AI as a fast assistant that prepares the work, while the specialist evaluates quality and correctness and makes the final decisions.

This presentation highlights the development of a scalable hardware-in-the-loop (HIL) simulation platform for next-generation Lamborghini supersport vehicles, enabling real-time testing of the complete vehicle software architecture using MATLAB^®, Simulink^®, and Simscape^™ to model longitudinal and lateral vehicle dynamics along with the full powertrain architecture.

A modular and model-based development approach integrates high-fidelity physical models with real-time execution constraints to achieve an optimal balance between accuracy and computational efficiency. The platform supports direct interfacing with ECUs and embedded controllers, which offers a safe and cost-effective environment for verifying software functions while reducing reliance on physical prototypes.

The framework has been validated against experimental data and optimized for real-time execution. Using Simscape Vehicle Templates and a fully parameterized model architecture reduced the development effort for HIL-ready models by approximately six months compared to a traditional approach. The availability of ready-to-use, detailed, and modular Simscape components accelerated model deployment and enabled significant cost savings.

The open and fully customizable model architecture allowed engineers to extend, adapt, and refine subsystems without vendor lock-in or additional licensing costs, ensuring long-term flexibility, reusability, and scalability across future vehicle platforms.

Validation time was reduced by 45–50 percent, supported by the adoption of automated test suites that increased requirement coverage above 90 percent and significantly shortened regression execution time. Early virtual testing enabled targeted stress-testing of critical edge cases that would be difficult, costly, or unsafe to reproduce on physical prototypes, reducing the overall testing effort and improving software maturity before on-road validation. 

Discover how AI-driven surrogate models can accelerate early-stage physical design. 

See how to leverage deep learning, machine learning, optimization, multibody modeling, and parallel computing to create a surrogate model. The resulting AI model calculates performance metrics nearly instantaneously,  enabling design space exploration and optimizing the design. 

To illustrate this workflow, a vehicle suspension system is used, where physical design parameters are tuned using the AI model to improve and balance competing metrics such as ride comfort and handling. The workflow is applicable to early-stage physical design for mechanical, electrical, and thermal systems.

As vehicle programs increasingly span internal combustion engine (ICE), hybrid electric vehicle (HEV), and battery electric vehicle (BEV) platforms, engineering teams face growing challenges in evaluating powertrain performance across diverse operating conditions, customer usage patterns, and global markets. Traditional desktop-based simulation approaches often constrain large-scale design exploration, fleet-level analysis, and rapid decision-making.

Learn how cloud-native simulation and analytics platforms enable engineers to execute thousands of virtual experiments in parallel, leveraging scalable computing infrastructure and automated workflows. By combining physics-based powertrain models with cloud computing, engineers can rapidly assess fuel economy, energy consumption, emissions, battery utilization, thermal performance, range, drivability, and component durability across a wide range of operating scenarios.

Discover an integrated framework that combines simulation, high-performance cloud computing, and advanced data analytics to accelerate design optimization, root-cause analysis, and virtual validation. Gain insight into how cloud-based execution reduces simulation turnaround time significantly, improves engineering productivity, and supports data-driven development of next-generation ICE, hybrid, and electric vehicles.