Syllabus

• Overview of mobile robotics (hardware, software architectures, sensors)
• Probabilistic models of sensing and acting
• Bayesian state estimation and filtering
  • - Kalman filters (extended, unscented)
  • - Particle filters
  • - Dynamic Bayesian networks
  • - Gaussian processes for model learning
• Robot localization
• Map building / SLAM
• Boosting and conditional random fields
• Markov decision processes (MDPs, POMDPs, reinforcement learning)

Textbook

• Probabilistic Robotics.
  S. Thrun, W. Burgard, and D. Fox.
  MIT Press, Cambridge, MA, September 2005.
  ISBN 0-262-20162-3.
  

Outline

Week	Date	Content	Readings (papers, book chapters)	Assignments
Introduction
#1	Sept 27
Probabilistic Models / State Estimation
#2	Oct 2	Probabilities, Bayes rule, Bayes filters	1, 2
	Oct 4	Motion models, Sensor models	5, 6
Bayes Filters and Robot Localization
#3	Oct 9	Kalman filters: KF, EKF	3, 7
	Oct 11	Kalman filters: UKF	3, 7	hw2 project1
#4	Oct 16	Discrete filters, particle filters	8
	Oct 18	Particle filters	8
Mapping / SLAM
#5	Oct 23	Occupancy maps, EKF-SLAM	9, 10, 11
	Oct 25	Fast-SLAM	13
#6	Oct 30	Fast-SLAM contd.	13
Further Estimation Topics
#7	Nov 6, 8	Complex tracking	GPS-street-map, Robocup-ball-tracking
	Nov 9	Gaussian Processes	Overview, GP-UKF active-learning, heteroscedastic
#8	Nov 13	GP contd., Boosting	Overview, Java Applet,
	Nov 15	Conditional Random Fields	Tutorial, GPS place labeling
#9	Nov 20	CRF applications	Indoor place labeling, Scan labeling
	Nov 22	Thanksgiving, no class
#10	Nov 27	MDP planning
	Nov 29	RL, POMDPs
#11	Dec 4	Active localization and sensing	Ch 17, active localization, sensing
	Nov 29	Summary