Professor Ankur Mehta

Fall 2017:

EE209AS: Robotics: Design, Manufacture, and Control

Lecture: MW 6-8pm, 2444 Boelter
Professor Office hours: T 4-5pm, W 2-3pm, 6730D Boelter
TA Office hours: R 3-4pm, EE Grad Lounge

Lecture notes (draft)

[[ November 16, 2016 ]]

1) Final presentations (Week 10):

25 minutes talk --- everyone on the team must speak!
- Briefly restate problem statement, motivation, and scope
- Defend changes from project proposal if necessary
- Describe design challenges and explain decision process and outcomes
- Present key experiments and results (think about effective ways to demonstrate results)
- Outline remaining work and set expectations for final demo
- Draw conclusions --- what can you teach the class?

10 minutes questions
- Learn from your classmates, both by asking and answering!
- Pay attention to what's interesting, confusing, or otherwise noteworthy for more detailed treatment your report

Again, you will be graded on the clarity and completeness of your talk, as well as your engagement during your and others' talks.

2) Demo day:

Monday Dec. 5, 2-4pm (Tesla room): Practice / dry run
Thursday Dec. 8, 2-5pm (Tesla room): Demo day
2-3pm setup, 3-5pm demos, 5pm pizza and cleanup
Open to the public --- invite friends, professors, colleagues

3) Anatomy of an academic paper / lab report:

Abstract: Summarize your paper in one paragraph. Briefly describe the problem statement, methods, and key results.
Introduction: Clearly motivate, define, and scope the problem that you are addressing.
Background: Describe any necessary mathematical or academic foundations.
Methods: Explain what you did and why.
Results: Present data, graphs, pictures.
Conclusions: What was learned, and what broader implications does this work have?
References: Cite your sources

[[ November 7, 2016 ]]

1) Grades for Lab 1, Lab 2, and the proposal presentations have been posted on eeweb. Note that the bulk of your final grade will come from your final project presentation and paper, so there will be ample opportunity for your grade in the class to change (in either direction).

2) We will have a "demo day" during finals week to show off your team projects to each other and any other visitors you'd like to invite. Please fill out the following poll to indicate your availability:
Please indicate all times that you can possibly attend by this Wednesday 11/9.

3) We are nearing the end of the term. After today's lecture, I will focus on current active areas of robotics systems research. We will only have time to cover 3-4 such topics, so if you have a preference for topics, fill out the following poll by this Wednesday 11/9:

4) Do you need custom parts for your final projects? The UCLA libraries are running a service where you can get parts 3D printed for free. For more information, see:

[[ October 25, 2016 ]]

1) Overview of your proposal presentations tomorrow:

10 minutes talk --- everyone on the team must speak!
- Clearly define problem statement
- Present motivation and scope
- Provide context / related work
- Identify particular challenges
- Define team milestones and goals

5 minutes questions
- Opportunity to get helpful feedback
- Clarify motivation, technical challenges, implementation
- Ask questions to your classmates!

In short, you must convince your audience (and be convinced by the other presenters) that the project goal as well as the process by which that goal is achieved is interesting, relevant, and doable.

This will be graded on the clarity and completeness of your talk, as well as your engagement during your and others' talks.

2) A full class's worth of material on the topic of control and controllers can be found through MIT OpenCoureWare here:
- 16.30 lecture notes

3) A detailed textbook on the subject of motion planning is available:
- here.

[[ October 14, 2016 ]]

Lab 2 / final project teams are posted here.

Lab 2 is due a week from today. While working on lab 2, you should take some time to discuss potential final projects within your team. For inspiration, I've compiled some thoughts here. I'm available during office hours if you'd like to discuss your project ideas with me.

You will be required to present your project proposal in class on October 26. I'll go over more details about this in class next week.

If you'd like to learn the math behind Kalman filters, the discrete time Kalman Filter is derived in the following lecture notes:
- Lecture 5,
- Lecture 6;
from the MIT OpenCoureWare class 2.160:
- here,
which also includes details of a continuous time Kalman Filter and the Extended Kalman Filter.

And as a reminder, our TA for the class is Ray Zhang (rayucla at ucla dot edu), who holds office hours Thursdays at 3pm in the EE grad lounge on the 5th floor of EIV.

[[ October 10, 2016 ]]

Lab 2 is posted. Due by 6pm Friday, Oct. 21.

[[ October 9, 2016 ]]

The due date of Lab 1 will be extended to Wednesday 10/12 at 6pm instead. However, no late submissions will be accepted.

Lab 2 will still be released tomorrow evening, Monday 10/10. It involves the use of microcontrollers, which I will be handing out in class.

In order to allow more time for the final projects, Lab 3 has been subsumed into Lab 2, which means there will be no further labs after the one given out tomorrow.

Lab 2 will be done in your final project groups. The groups will be assigned on Wednesday, which gives you some time to play with the hardware on your own. As I mentioned in the last class, if you have a specific project group + project idea, have one member of the team email me by Tuesday with the subject line "EE209AS Project teams" with the list of team members and a description of your project idea.

[[ October 9, 2016 ]]

The lecture notes are still a work in progress, and are being edited live. That being said, you can see them as they evolve here:

Also, in case you missed the announcement in class, we do have a TA, Tianrui Zhang, who holds office hours Thursdays at 3pm in the EE grad lounge, on the 5th floor of Engineering IV.

[[ October 3, 2016 ]]

Slides from lecture 1

Problem set 1 solutions
Lab 1 Due in class Monday, Oct. 10

A detailed treatment of the material covered in lectures 2-4 can be found in chapter 2 of the textbook "Robotics: Modelling, Planning and Control" by Bruno Siciliano et al. It uses slightly different notation than what I used in class, but hopefully it's still understandable.

The UCLA library carries the e-book -- from campus you can download the PDFs for free from this link:
- Robotics: Modelling, Planning and Control

Lab 1 deals with forward and inverse kinematics. We didn't get to this in lecture today, so we will cover it in class on Wednesday. If you want a head start, this will include material in the book from section 2.8 onwards. Note that the book uses the standard (classic) D-H convention, whereas in class we will use the modified D-H convention. The Wikipedia page on Denavit–Hartenberg parameters explains the difference: "The coordinates of O_i (origin of reference frame i) is put on the axis i, not the axis i+1 in classic DH convention."

[[ September 26, 2016 ]]

Course syllabus
Problem set 1 Due in class Monday, Oct. 3
Background questionnaire Due in class Monday, Oct. 3

[[ August 17, 2016 ]]

Welcome (back) to UCLA!

Please note the following regarding this class:

This class is limited enrollment and has been listed as waiting-list only; students will be given PTEs during the second week. You must attend the first two lectures and submit the first assignment to be eligible for a spot in the class.