ComS 610 AS: Developmental Robotics
Fall 2005, 2:10 - 3:30 p.m. (Tuesday and Thursday) Coover Hall, room 1219 Instructor: Alexander Stoytchev |
This class provides an introduction to the emerging interdisciplinary field of Developmental Robotics, which crosses the boundaries between robotics, artificial intelligence, and developmental psychology. The main goal of this field is to create autonomous robots that are more intelligent, more adaptable, and more useful than the robots of today which can only function in very limited domains and situations. The basic research hypothesis of developmental robotics is that truly intelligent robot behavior cannot be achieved in the absence of a prolonged interaction with a physical or a social environment. In other words, robots must undergo a developmental period similar to those of humans and animals.
The class will focus on representations and algorithms for robot learning that facilitate and/or benefit from a developmental period. At the end of this class you will have an understanding of the current state of the art in developmental robotics and will be able to conduct original research. In addition to that, you may also gain a deeper understanding of the development of human intelligence.
The class will cover a range of topics, including the following: Overview of robotics, robot architectures, behavior-based robotics, robot simulators. Theories of development and their application to robotics. The nature v.s. nurture debate. The principle of embodiment - or why should intelligent agents have bodies. The sense of self, body schema representation, and self-recognition in robots. Sensing and sensory self-organization. Sensory substitution. Learning about objects and their affordances. Computational representation of affordances. Learning to use objects as tools. Learning about space and spatial frames of reference - body-centric v.s. allocentric coordinates. Social learning - scaffolding, imitation, theory of mind, and language acquisition.
There is no required textbook for this class. The lectures will be based on a number of sources most of which are available for download from the Internet (links will be provided on the class web page). Reading material that is not available on-line will be placed on reserve in the library. The reading list will be posted on the class web page.
Advanced undergraduates and non-CS majors are welcome!
This is a joint graduate and advanced undergraduate class. Knowledge of C/C++ is required for the programming assignments in this class. An undergraduate/graduate class in Artificial Intelligence is recommended but is not required. The most important prerequisite of all, however, is your interest in the course, motivation, and commitment to learning. If you are not sure whether this class is for you, please talk to the instructor.
For non-CS students only (e.g., psychology majors): no programming experience is required. You will have the option of submitting a term paper instead of a term project.
Homework Assignments:
There will be three homework assignments. You will have two weeks to complete each one of them. These assignments will be used to emphasize and clarify important concepts discussed in the lectures.
A midterm exam will be given in the 8th week of classes. The exam will cover the topics discussed in the first half of the class. There will be a review session before the exam.
The final project must be a small research or design project that is related to the topics covered in class. You may choose to work individually or in small groups (2-3 members each).
You are encouraged to select a topic for your final project as soon as possible. A written project proposal (1-2 pages) will be due on October 20. The final project report (10-15 pages) will be due on December 1. Each team will be required to present the results of their final project during the last week of the semester.
Non-CS students have the option of writing a term paper instead of submiting a final project.
You are encouraged to form study groups and discuss the reading materials assigned for this class. You are allowed to discuss the homework assignments with your colleagues. However, each student will be expected to write his own solutions/code. Sharing of code is not allowed. No collaboration will be allowed during the in-class midterm exam.
You are expected to attend every class and participate in the class discussions. If you miss a class, it is your responsibility to find out what we talked about, including any announcements that were made in class.
Class participation: 10% Homeworks (3x 10%): 30% Midterm Exam: 25% Final Project: 35%