Lehigh CSE398/498 Advanced Topics in Mobile Robotics: DARPA Urban Challenge

.
CSE398-012/498-012
Advanced Topics in Mobile Robotics:
The DARPA Urban Challenge
7:55-9:10 PM, Spring 2007

Announcements

26 Apr 07: Your final assignment on optical flow is posted. Start early!
17 Apr 07: GP Stereo homework assignment is posted. This is due on 24 Apr 07 BEFORE CLASS. Note that late submissions will NOT be accepted.
16 Apr 07: I've posted Chao's presentation on vehicle detection.
22 Mar 07: I've linked notes on RANSAC below.
15 Mar 07: I will be on travel during lecture today. Jason Derenick (the TA) will be lecturing in class instead.
13 Mar 07: I've linked the lecture and and action items for development on the Urban wiki.
05 - 09 Mar 07: We will be testing all week at Goodman with Ben and the UPenn team. If you are interested in coming out, please let me know.
15 Feb 07: Testing is canceled for the near-term until we shed this snow on the ground...
08 Feb 07: We've collected our first data sets at Goodman and general roads. These will be available in the near-term for lane tracking
24 Jan 07: I've posted the lecture notes regarding stereo vision. If you expect to be working with the Bumblebee cameras, you should review these.
22 Jan 07: There will be no class on 01 Feb as I will be on travel.
18 Jan 07: The course wiki is available. You should begin documenting all of you work, as well as submitting progress reports. The latter will be used to support you final course grade. No progress reports = F in the course. You have been warned.
16 Jan 07: First day of classes. Welcome!

Course Description

The Defense Advanced Research Projects Agency (DARPA) is sponsoring a 60 mile race of autonomous vehicles in an urban environment know as the DARPA Urban Challenge (DUC). Lehigh will field an entry into the DUC in collaboration with the GRASP Laboratory at the University of Pennsylvania and Lockheed Martin Advanced Technologies Laboratory (ATL). The goal of this course is to implement estimation algorithms in software (C/C++) that will be used in support of this effort. These will be evaluated in a test vehicle in terrain representative of the DUC.

Possible topics to be investigated will include:

LIDAR
RADAR
Stereo and Monocular Vision
Vision-based lane tracking
Localization and Mapping

Prerequisites

CSE 397/497 Introduction to Mobile Robotics
CSE 397/497 Real-time Image Processing
CSE 109 Systems Programming or equivalent programming experience in C/C++
Instructor approval required

Instructor
John Spletzer
Office Hours: TR 9:15-11:00 and by appointment

Teaching Assistant
Jason Derenick
Office Hours: VADER Laboratory, PA450, by appointment

Course Location
Lectures: PA208
Laboratory Location: VADER Laboratory, PA450

Course Materials

DUC Home Page
DUC Rules (required reading for ALL)
The Ben Franklin Racing Team Home Page
Course Wiki Page
Particle Filters

D. Fox et al, Bayesian Filtering for Location Estimation, University of Washington, 2003
S. Thrun, Probabilistic Algorithms in Robotics, Carnegie Mellon University, 2000

Kalman Filters

G. Welch & G. Bishop, An Introduction to the Kalman Filter, University of North Carolina - Chapel Hill, 2001

Random Sample Consensus (RANSAC)

A one page summary from CVOnline
The original paper (15 pages - you can only access this from the Lehigh server)

Grading
This course is intended to be very applied with student work supporting the Lehigh entry into the DARPA Urban Challenge. Lectures will be a mixture of working group meetings and presentations by the instructor and students.

There will be no exams for this course. Your grade will be based upon three different components: progress reports, lecture/lab attendance, and peer evaluation. You are expected to submit weekly progree reports AT A MINIMUM. When you fail to document your work, you fail to justify a passing grade for the course. Lecture/Laboratory attendance is required, and will be graded as such. Finally, your peers will have input to your course grade.

The preliminary breakout of course grading is as follows:

Progress Reports: 40%
Lecture Attendance: 30%
Peer Evaluation: 30%

Grades will be assigned acording to the following distribution:
.

A	92-100	C	72-76
A-	90-91	C-	70-71
B+	87-89	D+	67-69
B	82-86	D	62-66
B-	80-81	D-	60-61
C+	77-79	F	00-59

University Policy on Disabilities: If you have a disability for which you are or may be requesting accommodations, please contact your professor and the Office of Academic Services, Room 212, University Center or call (610-758-4152) as early as possible in the semester. University policy states that you must notify your professor seven (7) days prior to the exam.

Plagiarism: I will submit any case of suspected copying to the University Committee on Discipline. Often, when the discipline committee finds a student guilty of copying, the penalty is a WF in the course. If the Committee on Discipline finds your guilty of the charge of copying and does not assign a WF, I reserve the right to assign the grade of F. NO COPYING OF ASSIGNMENTS WILL BE TOLERATED! Ensure that you give proper credit in citing/presenting the work of others.

Tentative Schedule
.

DATE	TOPICS	REFERENCES
16 Jan 07	Course overview (pdf)
13 Jan 07	Stereo Vision Review (pdf)
25 Jan 07	DARPA Grand Challenge Presentation (pdf)	TBA
30 Jan 07	Lane Tracking Review (Dilsizian)	pdf1 pdf2
01 Feb 07	Class canceled
06 Feb 07	Lane Tracking Review (Gao)	pdf
08 Feb 07	Lane Tracking Review (Boinot)	pdf
13 Feb 07	Particle Filtering	pdf
15 Feb 07	Stanley's vision during the DGC II	pdf
20 Feb 07	Phased Array RADAR (Dzaba)	pdf
22 Feb 07	Horizon / Sky Detection (Schaner)	pdf
13 Mar 07	Testing results	See the Wiki site
22 Mar 07	RANSAC / Hough Transforms	pdf
12 Apr 07	Vehicle Detection (Gao)	pdf
17 Apr 07	Ground Plane Stereo for Obstacle Detection (pdf)	hwk .m file left image right image left pedestrian image right pedestrian image
24 Apr 07	Optical Flow (pdf)	hwk .m file image files