Applied Control Systems 2: autonomous cars (360 tracking)
About Course
How do you make autonomous cars track a general trajectory on a 2D plane and how do you make sure that the velocities, accelerations and steering wheel angles of the autonomous cars stay within their realistic minimum and maximum values?
My name is Mark. I’m an Aerospace & Robotics Engineer and in this course, I will give you intuition, Mathematics and Python implementation for all that.
This course is a direct continuation to the course “Applied Control Systems 1: autonomous cars: Math + PID + MPC. In the previous course, the Model Predictive Control (MPC) algorithm only allowed the autonomous cars to change lanes on a straight road. We applied a small angle approximation to convert our nonlinear model to linear time invariant (LTI). It made our lives easier but it also restricted our Model Predictive Control algorithm.
In this course however, we will remove that simplification and I will show you how you can apply a linear MPC controller to a nonlinear system by putting it in a Linear Parameter Varying form first. With this highly popular technique, your car will be able to track a general 2D trajectory.
In addition, you will learn how to use quadratic solvers such as qpsolvers & quadprog to apply MPC constraints to autonomous cars. In most control problems, you have to consider constraints in order to keep your system within reasonable values.
The knowledge that you get from this course is universal and can be applied to so many systems in control systems engineering.
Take a look at some of my free preview videos and if you like what you see, then ENROLL NOW, and let’s get started.
Hope to see you inside!
Course Content
Revision
-
Introduction & general instructions
06:24 -
Equations of motion formulation in the lateral direction
06:53 -
Going from equations of motion to the state-space equations
06:33 -
Controller limitations from the small yaw angle assumption 1
05:02 -
Controller limitations from the small yaw angle assumption 2
08:21 -
Model Based Control VS PID
05:50 -
Revision of the plant box equations
04:58 -
Revision of the MPC basic principle
05:26 -
Revision of the MPC basic principle
05:26 -
Revision of the MPC basic principle
05:26 -
Revision of making the LTI system discrete
11:38 -
Revising how MPC predicts future states
07:12 -
Revision of augmenting the LTI state space model – exercise
05:43 -
Revision of augmenting the LTI state space model – solution
03:32 -
Revision of MPC cost function reformulation
11:06 -
Course Material Download Link
00:00
State space equations for the plant
LPV – MPC – no constraints (Intro to Linear Parameter Varying method)
LPV – MPC controller – with constraints + cubic polynomials
The Python code implementation of the LPV-MPC controller with the constraints
Earn a certificate
Add this certificate to your resume to demonstrate your skills & increase your chances of getting noticed.