IN Brief:
- MathWorks R2026a introduces Simulink Copilot and Polyspace Copilot for embedded engineering workflows.
- The release adds Polyspace as You Code, Simulink FMU Builder, STM32 and Raspberry Pi support, and Wireless Network Toolbox.
- Engineering AI tools are moving into verification, traceability, and model-based workflows rather than remaining simple code-generation assistants.
MathWorks has released R2026a of the MATLAB and Simulink product families, adding AI capabilities for embedded systems development, model-based design, and software verification.
The release introduces Simulink Copilot for model-based design and Polyspace Copilot for embedded software code analysis. MathWorks has also added Polyspace as You Code, enabling developers to check C and C++ coding rules and identify coding defects and vulnerabilities as code is written, including code produced with AI-assisted tools.
Simulink Copilot operates inside the modelling environment. It can generate model explanations, answer questions about model behaviour, help locate relevant blocks and subsystems, isolate issues, suggest remedies, and guide next steps. It can also execute standardised tasks that support consistent development and verification practices.
Polyspace Copilot works from Polyspace analysis results to help engineers interpret static-analysis findings, understand issues, and resolve them more efficiently. R2026a also brings a new Polyspace desktop application for unified configuration and results management, extensions to Polyspace Bug Finder with custom checkers and coding standards, and software-sanitising capabilities in Polyspace Test for dynamic analysis of runtime errors.
Avinash Nehemiah, head of product management and marketing, design automation at MathWorks, said: “Engineering teams now have access to capabilities enabled by generative AI, and leaders need confidence that these translate into tangible engineering and business benefits. In engineering design and software verification, productivity improvements cannot come at the expense of rigor, traceability, or trust.”
The release also includes MATLAB Course Designer, Simulink FMU Builder, Wireless Network Toolbox, Raspberry Pi Blockset, STM32 Microcontroller Blockset, and updates across MATLAB, Simulink, MATLAB Test, Mapping Toolbox, Signal Processing Toolbox, RF PCB Toolbox, RF Toolbox, Polyspace Bug Finder, and Polyspace Test. Simulink FMU Builder creates standalone Functional Mockup Units from Simulink models and C or C++ code, supporting model exchange and system integration workflows.
The embedded development value sits in the placement of AI assistance inside established engineering environments. Model-based design and static analysis depend on traceability, repeatability, and evidence. Automotive, aerospace, medical, industrial, and other regulated electronics sectors can use productivity tools only when they preserve the audit trail and keep design decisions within controlled workflows.
Grounding Simulink Copilot in models, defined processes, and MathWorks documentation keeps AI support inside the engineering context that teams already use. That boundary is becoming more relevant as development groups adopt generative tools while still working under coding standards, safety processes, cybersecurity requirements, and verification gates.
Polyspace as You Code extends the same shift-left approach into embedded software quality. Finding defects, vulnerabilities, or coding-rule violations while code is written reduces late-stage clean-up and avoids pushing static analysis into the end of the development cycle. AI-assisted coding can increase code output, but embedded systems still need predictable behaviour under constrained memory, timing, safety, and power conditions.
The wider R2026a updates show how embedded workflows are broadening. Wireless Network Toolbox supports end-to-end wireless network modelling and simulation, while STM32 and Raspberry Pi support strengthen routes from simulation into physical hardware. Simulink FMU Builder supports cross-domain model exchange as electronics, controls, mechanical systems, and software teams work through more integrated development cycles.
R2026a moves MathWorks’ AI direction into model interpretation, code analysis, software quality, and verification support. The release gives engineering teams a route to reduce repetitive navigation and analysis work while retaining the structure needed for complex embedded systems development.
