About the Course

This course will cover various topics in autonomous robotics, including architectures, basic kinematics, basic controls, reactivity, planning, simulation and modeling, sensing and locomotion, and multiagent environments. The course will cover both high-level abstract topics and nuts-and-bolts of robot construction, including some simple sensor and microelectronics and programming.

The course will include labs involved in construction and programming of a "swarm" of actual camera-guided robots from basic components. Depending on class size, such construction and programming may be done by teams. A final project is likely. Programming will be done in Java and C.

This is an experimental course and requirements may and will change without notice. It will, however, be a lot of fun.

Further information will appear on the Course Web Page

Texts

There are no good general texts on autonomous robotics. They're all terrible. So we're going with the cheapest terrible option: Computational Principles of Mobile Robotics by Gregory Dudek and Michael Jenkin. Get the softcover version. We will also be using various lecture notes and chapters from other texts as necessary. This book will not be available at the GMU bookstore as a textbook, though you may be able to order it from them as a general book. Instead I recommend ordering it from Amazon. The book will be helpful at the beginning of class, but I suspect it will not be necessary until two weeks in.

Tentative Class Schedule

This class schedule is certain to change significantly. It's primary function here is to give you an idea of what the course might entail. There are fourteen weeks all told:

1. Introduction, Braitenberg Vehicles, basic concepts in autonomous agents, laboratory, robot construction
2. Microelectronics, Sensors, Effectors, Mechanics
3. Reactive Control
4. Kinematics
5. Controls and Pose Maintenance
6. Multi-agent Robotics
7. Midterm. Special lecture: Computer Vision
   (Spring Break)
8. Search, Probability
9. Localization
10. Path Planning
11. Partial-Order Planning, Multi-agent Planning, Robot Architectures
12. Reinforcement Learning
13. Neural Networks
14. (Final Project Presentations)

Grading Policies

This course will consist of homework and projects, possibly including a final project, and two exams. The breakdown will be approximately:

Homework and Projects	50% with higher weight given to harder projects (the final project may count for half of the weight or more)
Exams	25% Each

There will be no make-up tests for missed examinations. Late homework will be accepted but at a loss of 20% per day (homework later than 4 days, or after the last day of class, is worth nothing).