. CSE350/450-011: An Introduction to Mobile Robotics TR 1:10-2:25 PM, Fall 2003

Announcements

• 10 Nov 03:  There will be no office hours this Friday, 14 Nov 03.
  
• 10 Nov 03:  I've added a reference from Dieter Fox et al on Baysian Filtering for Position Estimation
  
• 10 Nov 03:  The FAQ page is active for homework 3.
• 06 Nov 03:  Homework 3 is posted.  Due date is 13 Nov 03 before class.
• 06 Nov 03:  Syllabus has been updated to reflect the schedule for grad student talks.
• 06 Nov 03:  As requested, here's the matlab code for the in class KF exercise.  Use it at your own risk.
  
• 16 Oct 03:  Office hours tomorrow are from 9AM-1PM.  There are no office hours next week.
  
• 16 Oct 03:  Homework 2 will not be accepted after 1 PM, 17 Oct03
  
• 10 Oct 03:  Homework 2 deadline has been extended.  It is now due on 15 Oct, 1 PM.   
  
• 06 Oct 03:  I've added a FAQ page which will answer questions to the homework, etc. that I receive via e-mail.
  
• 02 Oct 03:  Lab sessions are today 11:40-2:25, and Tuesday, 7 Oct 03, 1:10-3:55  
  
• 29 Sep 03:  Homework two is posted.  Please review ASAP for clarifications.  
  
• 16 Sep 03:  A pair of matlab functions for those new to the program.  Use at your own risk:  makeRobot.m & moveRobot.m
  
• 16 Sep 03:  Late Policy added below.
  
• 04 Sep 03:  NOTE:  Homework 1 is updated to fix a typo in the lecture notes.  
  
• 04 Sep 03:  Homework 1 is posted, and due BEFORE class 16 Sep 03
  
• 29 Aug 03:  My office is open for business.  Starting next week, office hours will be in PL404A.
  
• 28 Aug 03:  Here is a Matlab function that you can use as a template for homework if you like.  It should help if you are new to the software.
  
• 28 Aug 03:  NEW, New office hours:  T 9:00-10:30, F 4:30-6:00PM and by appointment.
  
• 22 Aug 03:  I will be on travel during the week of 28-30 Oct.  We will make up the missed class time as necessary to cover the required material.
  
• 22 Aug 03:  First day of classes, 26 Aug 2003.

Course Description
In this course, we will review prevalent algorithms employed in mobile robotics for motion planning and estimation, with a bias towards the latter.  The objectives of this course are to build a toolkit of algorithms allowing students to implement common robot behaviors, as well as understanding related publications in robotics research.  Topics will include:

• Review of common sensor systems
  
• Motion planning algorithms
• Robust estimation techniques
• Bayesian estimation techniques

Prerequisites
Students will be required to implement several simulators in Matlab over the course of the semester, so a background in programming is assumed.  Knowledge of linear algebra and basic probability theory is helpful, but not required.  

Instructor
John Spletzer
Office Hours:  T 9:00-10:30, F 4:30-6PM and by appointment

Course Materials
There is no required text for this course.  We will be making use of the following online materials (and others), plus selected papers:

1. 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
2. Kalman Filters
• G. Welch & G. Bishop, An Introduction to the Kalman Filter, University of North Carolina - Chapel Hill, 2001
3. Random Sample Consensus (RANSAC)
• A one page summary from CVOnline
• The original paper (15 pages - you can only access this from the Lehigh server)
  
4. Motion Planning: 
• “Motion Planning Using Potential Fields”,  R. Beard & T. McClain, BYU, 2003
  
5. Image Features
   
• findCorners.m is a matlab function for corner detection
  
• CVOnline “Color Image Processing” Lecture Notes
• Charles Poynton's Color FAQ
  
6. “A Tutorial on Linear Algebra” by Professor C. T. Abdallah, University of New Mexico
7. Mathworld has the answer to just about any math question you might have.
   
8. J. Borenstein, H. R. Everett, and L. Feng, Where am I? Systems and Methods for Mobile Robot Positioning, Mobile Robotics Lab, University of Michigan, March 1996 
9. A. Kelly, Introduction to Mobile Robots Course Notes, Field Robotics Center, Carnegie Mellon University, 2000, with a particular emphasis on "Uncertainty 1-3"
10. Matlab Tutorial & Documentation, Mathworks web site

Grading
Grading will be based on 4 assignments, with an additional  informal paper presentation to satisfy the 400-level requirement. Students are permitted to discuss the assignment.  However, all work must be the students' own.  

Students will be given a minimum of one week notice before the assignment due date.  Late assignments will not be accepted unless previous arrangements were made with the course instructor.    

There will be no exams for this course.

The breakout of course grading  is as follows:

• Homework assignments:  80% (20% each)
• Class attendance/participation - particularly for the paper presentations (undergrad):  20%
• Paper presentation (grad):  20%

ALL ASSIGNMENTS MUST BE SUBMITTED IN ORDER TO RECEIVE A PASSING GRADE IN THIS CLASS.

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

Late Policy:  Assignments are due PRIOR to class on the due date.  You can submit late (up to the NEXT CLASS TIME) ONE TIME without penalty.  Additional late submissions will be penalized 10% of the final grade.  No assignments will be accepted if over 1 class late.  

Disabilities:  If you have a disability for which you are or may be requesting academic accommodations, please contact your professor and the Office of Academic Support Services, Room 212, University Center (610-758-4152) as early as possible in the semester.

Tentative Schedule
.

DATE	TOPICS	NOTES	ASSIGNMENT
26 Aug	A brief overview to mobile robotics	26 Aug 03
28 Aug - 9 Sep	A review of common sensor systems Position Estimation Vision Systems Range/Bearing Sensors	28 Aug 03 2 Sep 03 4 Sep 03 9 Sep 03	Development of Sensor models
11-16 Sep	Motion Planning Algorithms Potential Field Approaches A* Grassfire	11 Sep 03 16 Sep 03 18 Sep 03
18-23 Sep	Robust Estimation RANSAC	23 Sep 03	Motion Planner & Wall follower simulators
25 Sep - 30 Sep	Review of rigid transformations and more probability theory	25 Sep 03 30 Sep 03
2 Oct - 7 Oct	Computer Vision Laboratory
9 Oct	Pacing Break
14-23 Oct	Sensor Fusion - The Kalman Filter	14 Oct 03 16 Oct 03
28-30 Oct	On Travel (NO CLASS)
23 Oct - 6 Nov	The Extended Kalman Filter	23 Oct 03 06 Nov 03	Kalman Filter Simulator
11-13 Nov	Particle Filters & Monte Carlo Localization (MCL)	11 Nov 03 13 Nov 03	MCL Simulator
18-25 Nov	Paper presentations 18 Nov:  Raim & Derenick 20 Nov:  Luksenberg & Maurizio 25 Nov:  Erekson & Ormsby
27 Nov	Thanksgiving (NO CLASS)
30 Nov - 5 Dec	TBD	02 Dec 03 04 Dec 03