Patricio Vela: Course Wiki

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
~~NOTOC~~ 


====== ECE6554: Adaptive Control ======
--------
**Course Description:**
Methods of parameter estimation and adaptive control for systems with constant or slowly varying unknown parameters. MATLAB design projects emphasizing applications to physical systems. 

--------

===== Course Resources =====

  - {{ECE6554:syllabus.pdf| Sample Syllabus}} (Not Official and certainly Not Valid)
  - [[ECE6554:Schedule | Timeline]]
  - [[ECE6554:References | Course References ]]
  -  Lecture Notes [[ECE6554:Lectures |Table of Contents ]] and Additional Notes.


===== Course Project =====
==== Project Options ====

The intent behind the project is to explore more complex nonlinear control systems that are linearly controller in some manner while requiring additional effort beyond the standard linear DMRAC approach explored in the homework.  The step up in complexity can take various forms, with the simplest being the MIMO case.  Another increment would be neuro-adaptive control, which considers a more complex function space representation for addressing nonlinear elements in the control system.  The last option for nudging up in complexity would be to explore different Lyapunov-based realizations of adaptive control. What is explored in the course is one particular strand of adaptive control based on the error dynamics arising from a reference model (MRAC).

Curated project options include:
     - [[ECE6554:Project_PlanarHeli | Planar Bi-Rotor Helicopter]]
     - [[ECE6554:Project_DuctedFan | Planar Ducted Fan ]]
     - [[ECE6554:Project_InvCartPend | Inverted Cart-Pendulum]]
     - [[ECE6554:Project_Quadcopter | Quadcopter]] (//Higher Level of Difficulty//)
     - [[ECE6554:Project_BallBeam | Ball and Beam System]] (//Under construction. Ask if interested.//)
       - Using traditional approach, this is absurdly difficult. The description makes it easier be breaking the problem down into inner and outer loop control designs. 
       - There is another way to make it way easier. If you want to explore this as an option, we can try. Involves following PWMN as baseline controller and source of adaptive update law. See the performance reference adaptive control paper (Azimi and Vela).
       - Alternatively, this problem was tackled in Spring 2022 using SOS tools in conjunction with some adaptive scheme. There is a published paper describing the approach that could be followed.

==== Project Execution ====

After the core adaptive control material has been covered in class and basic implementations in the homework, the project option will be initiated.  It will be active during the last 6 weeks of the course (more or less).  The curated projects have been broken down into digestible steps to work through.  There will be project deliverables for the early steps, but not necessarily for all of the steps.  The deliverables are to ensure that the project is started with enough lead time and to build up momentum.  It also provides an opportunity to reflect on achievements to date and consider how to successfully follow through on the remaining steps.  The deliverable dates should be seen as targeting minimally acceptable milestones that improve the likelihood of successful project completion. You are free to complete more than what is requested. Doing so would strengthen your position going into the final weeks of April. 

------
;#;
[[: | Main]]
;#;