Boards & Bots
Menu
Home
Sign inSubmit a link

GuideBB-0058

MathWorks tech talk on an agent building a Simulink pendulum controller

MathWorks shows an agentic AI tool working with MATLAB and Simulink to control an inverted pendulum on a Quanser Qube-Servo and Raspberry Pi. In the 14-minute video the agent helps define requirements, builds the Simulink model and designs a model predictive controller. After model-, software- and processor-in-the-loop tests, the final chapter deploys the generated code to the hardware.

Open original source youtube.com · MATLAB

Shown working
Why this evidence label?

Fourteen-minute video; the final chapter deploys the generated controller to the Qube-Servo and Raspberry Pi hardware after staged MIL, SIL and PIL testing.

How we check entries
Play videoMathWorks tech talk on an agent building a Simulink pendulum controllerThe player loads from youtube-nocookie.com when you press play.
The source video, on YouTube.

Briefing

The first 4 minutes are framing: how to approach engineering problems with an agent, then getting it to work through Model-Based Design and existing workflows. The description sets this against "vibe coding" and lists reusing verified models and toolboxes as part of the workflow. Requirements start at 04:27, the Simulink model at 05:53, the model predictive controller at 07:53, MIL, SIL and PIL testing at 09:00 and deployment at 11:52.

MathWorks published the 14:26 talk on its MATLAB channel on 21 May 2026, as a demo of its own tools. The description does not name the agentic AI tool used.

Written by the Boards & Bots curator

Tags

Report a problem with this entry