Course Project for MAE 280A Linear Systems Theory, taken during Fall 2019.
This project aims to model and control the MiP (Mobile Inverted Pendulum) Robot, developed by WowWee Robotics, in collaboration
with UCSD Coordinated Robotics Lab headed by Professor Tom Bewley.
The task is to design a discrete linear state-estimate feedback controller for MiP that performs decently.
The system has 4 states θ , ϕ , θdot , ϕdot . Our task is to design a controller for this system such that
we get a linear quadratic gaussian controller consisting of a linear feedback controller coupled with a state observer.
The problem requires us to experiment with the eigen values of the feedback and the observer system,
and to arrive at the optimal controller parameters. We are trying to control the reduced-order system,
as θ plays no role in the the system's model.
MIPParameters.m -- Initialises a list of parameters in the workspace.
nLMIp2.slx -- Simulink model containing th non-linear MiP model.
VenkateshPrasadVReport.pdf -- Detailed report of the experiments conducted to arrive at the optimum eigen values for the controller.