Piyush Khandelwal firstname.lastname@example.org
Office Hours: TBD (ENS32)
Jack O’Quin email@example.com
Austin Robot Technology
Jesse Vera firstname.lastname@example.org
Office Hours: TBD (Intel Lab)
This course presents an opportunity for students to help decide whether they would enjoy going on to graduate school and an eventual career as a computer science researcher. In particular, students will be required to read published research papers, participate in discussions, propose and execute a solution to a challenging open-ended problem, make presentations to the class, and write about their work. The 2005 DARPA Grand Challenge proved that autonomous vehicles are currently technologically feasible. 5 cars navigated more than 100 miles in the Mojave Desert without any human control. However in that case, the cars were given pre-specified routes, and did not need to deal extensively with each other.
The obvious next challenge is getting cars to drive in traffic. Indeed DARPA hosted the 2007 Urban Challenge with exactly that focus. This course began as an attempt to participate in the 2007 Urban challenge. The software written by students of the class for an existing autonomous vehicle placed in the top 20 teams. The challenge of this particular class will be to recreate the software necessary to support the Urban Challenge behaviors, as well as support future undergraduate and graduate-level research on the vehicle.
You can find more information about the autonomous car at http://cs.utexas.edu/~piyushk/marvin
If you have a question, please send an email to email@example.com. Everyone in the class will be subscribed to this mailing list, so others will also be able to see the email. If you only wish to contact the instructor only, please do so directly.
Attendance/Participation - 10%
Reading Responses - 15% (~20 Papers)
Assignment - 10%
Paper Presentation - 25%
Final Project - 40%
|1||8/29/12||Class Introduction [slides]|
|2||9/3/12||No class (Labor Day)|
|9/5/12||Project Discussion + ROS Recap [slides]|
|Data Fitting & Classification|
|3||9/10/12||Introduction to Probability Theory. The EM algorithm. k-means [slides]|| - *What is the expectation maximization algorithm? Do, C.B. and Batzoglou, S. Nature. (2008) [pdf]|
- Expectation Maximization: A Gentle Introduction. Blume, M (2002) [pdf]
- The Expectation Maximization Algorithm. Dellaert, F. (Tech Report 2002) [pdf]
|9/12/12||Hough Transform||- * (NOTE: Require reading is only Sections 1-3, pages 87-95) A survey of the Hough Transform. Illingworth, J. and Kittler, J. (ICVGIP 1988) [pdf]|
|4||9/17/12||RANSAC + Applications|| -* (Chapter 3 only) RANSAC for Dummies. Marco Zuliani (2012) [pdf]|
- Overview of the RANSAC Algorithm. Konstantinos G. Derpanis (2010) [pdf]
|9/19/12||Nearest Neighbors, SVMs|| - A Tutorial on Support Vector Machines for Pattern Recognition. Christopher J.C. Burges (1998) [pdf] [slides]|
- Support Vector Machines. Michael Quinlan [pdf]
|Vision + Detection|
|5||9/24/12||Intro to Vision - I [AAAI Tutorial] [slides]||- *Object recognition from local scale-invariant features. Lowe, D.G. (ICCV 1999) [pdf] [binaries] [slides]|
|9/26/12||Intro to Vision - II|| - *Rapid object detection using a boosted cascade of simple features. Viola, P. and Jones, M. (CVPR 2001) [pdf]|
- Histograms of oriented gradients for human detection. Dalal, N. and Triggs, B. (CVPR 2005) [pdf]
|6||10/1/12||*”Bag of words”||- *Video Google: A text retrieval approach to object matching in videos. Sivic, J. and Zisserman, A. (ICCV 2003) [pdf]|
|10/3/12||*Complex Models|| - *Object class recognition by unsupervised scale-invariant learning. Fergus, R. and Perona, P. and Zisserman, A. (CVPR 2003) [pdf]|
- A discriminatively trained, multiscale, deformable part model. Felzenszwalb, P. and McAllester, D. and Ramanan, D. (CVPR 2008) [pdf]
|Mechanics + Motion Control|
|7||10/8/12||Basic Mechanics Tutorial (TBD)||Assignment 1|
|10/10/12||Controls||- *Control Tutorial. Kuipers, B. (2004) [pdf]|
|8||10/15/12||Braitenberg's Vehicles||- *Vehicles: Experiments in synthetic psychology. Braitenberg, V. (1986) [pdf]|
|Localization + SLAM|
|10/17/12||Localization - I||- *An introduction to the Kalman filter. Welch, G. and Bishop, G. (1995) [pdf]||Literary Review|
|9||10/22/12||Localization - II||- *Monte Carlo Localization: Efﬁcient Position Estimation for Mobile Robots. Dieter Fox, Wolfram Burgard, Frank Dellaert, Sebastian Thrun. (AAAI 1999) [pdf]|
|10/24/12||Intro to SLAM||- *Simultaneous localisation and mapping (SLAM): Part I the essential algorithms. Durrant-Whyte, H. and Bailey, T. (RAM 2006) [pdf]|
|10||10/29/12||*SLAM - II||- *DP-SLAM: Fast, Robust Simultaneous Localization and Mapping Without Predetermined Landmarks. Eliazar, A and Parr, R. (IJCAI 2003) [pdf]|
|Path Planning + Navigation|
|10/31/12||Intro to Navigation + AStar Search Algorithm|
|11||11/5/12||*DStar Lite||- *Improved fast replanning for robot navigation in unknown terrain. Koenig, S. and Likhachev, M. (ICRA 2002) [pdf]|
|11/7/12||*RRTs||- *Rapidly-exploring random trees: Progress and prospects. LaValle, S.M. and Kuffner Jr, J.J. (2000) [pdf] (access this through a cs machine)|
|12||11/12/12||*Potential Fields (sort of)||- *Elastic Bands: Connecting Path Planning and Control. Quinlan, S and Khatib, O. (ICRA 1993) [pdf]|
|11/14/12||*RGB-D Mapping||- *RGB-D mapping: Using depth cameras for dense 3D modeling of indoor environments. Henry, P. and Krainin, M. and Herbst, E. and Ren, X. and Fox, D. (ISER 2010) [pdf]|
|13||11/19/12||Jazz Improvization Robot (discussion led by Elad Liebman)||- *Gesture-based human-robot Jazz improvisation. Hoffman, G. and Weinberg, G. (ICRA 2012) [pdf]|
|11/21/12||No class (Pre-Thanksgiving)|
|14||11/26/12||*Parking Lot Navigation||- *Apprenticeship learning for motion planning with application to parking lot navigation. Abbeel, P. and Dolgov, D. and Ng, A.Y. and Thrun, S. (IROS 2008) [pdf]|
|11/28/12||*Robotic Grasping||- *Robotic grasping of novel objects using vision. Saxena, A. and Driemeyer, J. and Ng, A.Y. (IJRR 2008) [pdf]|
|15||12/3/12||Final Project Presentations|
|12/5/12||Final Project Presentations|
- Topics marked with a * are open for student presentations
- Papers marked with a * are required reading. Responses need to be submitted for these papers.
No final exam.
CS378 Autonomous Vehicles in Traffic I. This requirement can be completed by submitting solutions to the first 3 assignments of the course.
Use any of machines in the ENS basement (ENS 1, ENS 2, Intel Lab - ENS31NR).
- Do not copy code from anywhere without citing where you got the code from. We encourage the use of code snippets available from the internet, as long as you understand them. However, you need to ensure that your code is written independently of your classmates.
- No part of the text written in your reviews or assignments must be copied verbatim from a source. You may rewrite the text and use a citation.
- If there are circumstances which are preventing you from completing your assignment on time, please let me know in advance.
Late submissions will not be accepted without prior consent from the teaching staff. We are fairly lenient, so do ask.
2 weeks before your presentation, read through your paper and setup a meeting with me. We'll go over any details that were not clear, and discuss what background material you should prepare in your slides as well.
1 week before your presentation, prepare a rough draft of your slides (including the background material), and setup a meeting with me. We'll go through your slides and change them if necessary to make them more suitable for the class.
While reading the paper, you might be unfamiliar with some concepts. Try searching for these topics online. All you really need is a high level overview of these concepts.
Based on your reading of the paper, you have to write and turn in a 2-3 paragraph response by the due date (typically 10PM the night before when we plan to discuss the paper). The responses should be free form. Credit will be based on evidence that you have done the readings carefully. Acceptable responses include (but are not limited to):
- Insightful questions;
- Clarification questions about ambiguities (please specify exactly where in the text);
- Thoughts on what you would like to learn about in more detail;
- Possible extensions or related studies;
- Thoughts on the paper's importance; and
- Summaries of the most important things you learned.
To submit your response, please mail it directly in plain-text (no attachments, no rich formatting) to the instructor at firstname.lastname@example.org. Please make the subject of the email ”[cs378] reading response due <month>/<date>”.
We will help you choose a suitable research project. Grading for research projects will be divided into 3 parts:
Literary Review - It is necessary to review existing literature before you start work on the project. You will need to summarize relevant papers and compare your research project against these approaches.
Final Presentation - This will be a formal 5-10 minute presentation given to your fellow classmates. You will need to summarize various aspects of your project with the results.
Final Report - Needs to be submitted on the last Friday of the semester.