build details

Show: section status errors & todos local changes recent changes last change in-page changes feedback controls

Learning-based Control Exercise

Modified 2020-12-06 by Fidel Esquivel Estay

Implement a learning-based controller for the Duckiebot in the simulator.

In this exercise you will implement an LQR controller for your Duckiebot using the Duckietown simulator. First, you will collect Duckiebot data (states and control inputs) of the robot performing standard lane following. Second, you will train a machine learning model with the data collected, and lastly, you will use the model to implement an LQR controller and comment on your performance and implementation.