MathWorks Adds AI Copilots to MATLAB and Simulink for Embedded Development
MathWorks released MATLAB and Simulink R2026a with new AI features designed to speed up embedded systems development without sacrificing rigor or traceability. The update introduces three copilots-Simulink Copilot, Polyspace Copilot, and MATLAB Copilot-built directly into the tools engineers already use.
What the Copilots Do
Simulink Copilot answers questions about model behavior, generates explanations of designs, and helps locate specific blocks and subsystems. It can isolate issues, suggest fixes, and guide engineers through standardized development tasks.
Polyspace Copilot interprets static analysis findings and helps resolve code issues faster. Polyspace as You Code checks C and C++ code for defects and vulnerabilities as developers write, including code generated by AI tools.
MATLAB Copilot generates test cases, including starter tests and equivalence tests drawn from command history.
New Products and Tools
R2026a includes several new capabilities across the product family:
- MATLAB Course Designer helps educators build courses and assessments using MATLAB and Simulink
- Simulink FMU Builder creates standalone Functional Mockup Units from Simulink models and C or C++ code
- Polyspace desktop application unifies configuration and results management for static analysis
- Wireless Network Toolbox models and simulates wireless communication systems
- Signal Processing Toolbox adds new Filter Designer and Filter Analyzer apps
MATLAB now supports building interactive webpages with visualizations without installing the software locally. Simulink can simulate C and C++ code directly within models.
Integration with AI Workflows
MathWorks also introduced MATLAB MCP Core Server and MATLAB Agentic Toolkit to integrate MATLAB and Simulink into agent-based workflows. This allows teams to understand designs faster and apply development practices more consistently across projects.
The copilots are grounded in users' models, team-defined processes, and MathWorks documentation rather than general-purpose language models. Avinash Nehemiah, head of product management for design automation at MathWorks, said the goal is to deliver AI tools that improve speed without compromising the discipline required for complex engineered systems.
For professionals working in embedded systems, software verification, or model-based design, these tools address a specific challenge: adopting AI capabilities while maintaining the traceability and rigor that safety-critical systems demand. Learn more about AI Coding Courses and AI for IT & Development to understand how these tools fit into broader development workflows.
Your membership also unlocks:
