Material Detail

This educational module introduces undergraduate biomedical engineering students to applied machine learning in feedback control through a bounded single-neuron stochastic-gradient-descent-style adaptive controller. The controller uses proportional, integral, and derivative error features as neuron inputs and updates the corresponding adaptive weights online during simulation.   Students apply the controller to an incubator temperature-control model with actuator saturation and a cold-load disturbance representing the addition of cold eggs. A classical PID benchmark response is provided so that students can compare disturbance recovery, overshoot, steady-state error, control effort, actuator saturation, interpretability, and implementation complexity.   The module emphasizes critical evaluation rather than assuming that machine learning automatically outperforms classical control. Students are asked to determine whether online adaptation provides enough benefit to justify the added complexity compared with a simpler classical PID benchmark.   The module was developed for use in an undergraduate biomedical control systems course and is intended for instructional use in control systems, biomedical engineering, adaptive control, or applied machine learning courses.

This will delete the comment from the database. This operation is not reversible. Are you sure you want to do it?

Thank you for reporting a broken "Go to Material" link in MERLOT to help us maintain a collection of valuable learning materials.

If you feel this material is inappropriate for the MERLOT Collection, please click SEND REPORT, and the MERLOT Team will investigate. Thank you!

You are being taken to the material on another site. This will open a new window.

Do not show me this again