Artificial Intelligence (AI)

Bring the Benefits of AI to Engineers and Scientists

Engineers and scientists use MATLAB to build impactful AI-driven products and services spanning multiple industries, from aerospace and automotive to biotech, energy production, financial services, medical devices, and railway systems.

Panel Navigation

Korea Institute of Energy Research

Panel Navigation

Poclain Hydraulics

Reliable Tools for AI-Driven Systems

AI is an emerging and rapidly evolving technology. MATLAB empowers engineers and scientists to use AI in their domain and enables collaboration across teams and organizations.

With MATLAB, you can:

  • Create AI models with a few lines of code or use pretrained models
  • Use domain-specific tools and low-code apps to build complete and scalable AI workflows
  • Combine AI techniques with system-level simulation to reduce errors in production
  • Deploy AI models to high-performance systems, such as edge devices and the cloud
  • Exchange AI models and design functionality between MATLAB and Python
Deep Learning

Deep Learning

Design, simulate, and deploy systems with deep neural networks.

Machine Learning

Machine Learning

Train models, tune parameters, and deploy to production or the edge.

Reinforcement Learning

Reinforcement Learning

Define, train, and deploy reinforcement learning policies.

Featured Tools

Interactive Apps

Use low-code apps to label and process data, build and train deep AI models, and manage AI experiments.

Pretrained Models

Get a pretrained model from the MATLAB Model Hub, TensorFlow™, or PyTorch® and adapt it to your task.

Robust Modeling

Visualize and interpret the AI model’s predictions and verify the model’s robustness properties.

 

Apply AI to Your Domain

Whether you are new to AI or developing your skills, MATLAB lets you integrate AI into various application workflows such as robotics, predictive maintenance, and many more.

Featured Application: Visual Inspection

Use computer vision to detect anomalies in images automatically. AI applications such as visual inspection require a systematic approach to:

  • Improve the quality of the training data with automatic labeling, data cleaning, and synthetic data generation
  • Achieve prediction accuracy with the AI model that is suitable for production deployment
  • Test the integration of the AI model with other parts of the system

AI with Model-Based Design

Engineers combine AI with Model-Based Design to accelerate and enhance the design of complex systems.

  • Create AI models of complex nonlinear dynamics to complement first-principles models.
  • Use AI to develop embedded algorithms that are difficult or impossible to implement with other methods.
  • Validate and verify AI-driven systems through requirements linking, simulation, and testing.
  • Train reinforcement learning agents via interactions with a simulated environment.
  • Generate synthetic data for training AI models by simulating a model of the physical system.
Panel Navigation

Integrating AI into System-Level Design

Panel Navigation

Coca-Cola Develops Virtual Sensor with Machine Learning

Panel Navigation

Virtual Sensors with AI and Model-Based Design

 

Engage with the MATLAB AI Team and Community

Discover the latest AI news and insights, explore tutorials and examples on AI workflows and applications, and share ideas, knowledge, and code.