Class information

Room: Science Center 252
Time: Tuesday, Thursday 1:15–2:30pm
Professor: Lisa Meeden
Office: Science Center 243
Phone: 328-8565
Office hours: I will always be available Wednesdays 3-5pm, or you can just stop by whenever my door is open.

Introduction

This seminar will explore the topic of adaptive robotics with a special focus on developmental robotics, a newly emerging paradigm of research. The goal of developmental robotics is to create intelligent robots by allowing them to go through a developmental process, rather than being directly programmed to solve a particular task. By endowing a robot with an appropriate initial control architecture and adaptive mechanisms, it can learn through continual interactions with the world, developing self-organized knowledge about itself and its environment. We will be studying the following sorts of questions: What should be innate in the robot? What adaptive mechanisms are needed? What motivates the robot to act?

Goals for the course

• Analyze and critically discuss research papers on adaptive robotics both in writing and in class.
• Understand the fundamental questions in the field of adaptive robotics and be able to analyze different solutions to these questions.
• Formulate and evaluate a hypothesis about adaptive robotics by proposing, implementing and testing a project.
• Relate your project to prior research via a review of related literature.
• Orally present a clear and accessible summary of your project.
• Write a coherent, complete paper describing and evaluating your project. You can see examples of past projects at CS81 Spring 2008.

Grading

• Class Participation and Paper Summaries: 20%
  Each Tuesday we will focus on a particular paper from the primary literature. Read the paper prior to class and bring a typed summary of the paper with you to class. You will turn this in at the end of class. The summary must include:
  • A 1-2 paragramph summary of the paper, describing its main points, and why you think this paper is important.
  • A list of questions that we can discuss during class. These may include questions of clarification, but should largely be made up of more interesting questions. These might include: how this paper fits into the field as a whole or how could the experiments be improved or expanded.
At the end of class you will also complete a self-evalutation of your particiaption in the discussion.

Your participation will be evaluated with this form.
• Labs: 20%
  Each Thursday we will focus on lab work. Initially the labs will give you a chance to learn how to control both simulated and physical robots. Later labs will allow you to experiment with the models we are reading about in the primary literature. Some lab time will be designated for making progress on the midterm and final projects.

• Midterm Project: 20%
  There will be a mid-semester project assigned on February 10 and due on March 6. The goal of this project is to give you a preview, on a smaller scale, of what will be expected on the final project.
  • Checkpoint demonstration
    
  • Paper 20%
    
• Final Project: 40%
  You will be asked to design a project related to adaptive robotics. These projects may be carried out alone or in pairs.
• Proposal 5%
  
• Checkpoint demonstration 5%
  
• Presentation 10%
  
• Paper 20%
  

Schedule

WEEK	DAY	ANNOUNCEMENTS	READING
1	Jan 20		Introduction and A Neural Network Primer by Thomas R. Schultz, Chapters 1-2 from Computational Developmental Psychology, MIT Press, 2003.
	Jan 22		Lab1: Robots learning with neural networks
2	Jan 27		Trends in Evolutionary Robotics, by Lisa Meeden and Deepak Kumar, Chapter 9 in Soft Computing for Intelligent Robotic Systems, eds. Lakhmi Jain and Toshio Fukuda, Physica-Verlag, 1998.
	Jan 29	Drop/Add ends (Jan 30)	Lab2: Controlling Aibo, Khepera, and Pioneer robots
3	Feb 03		Competitive coevolution through evolutionary complexification by Kenneth Stanley and Risto Miikkulainen, Journal of Arificial Intelligence Research, vol. 21, 2004.
	Feb 05		Lab3: Evolving robot controllers with NEAT
4	Feb 10		Evolving 3D Morphology and Behavior by Competition by Karl Sims, in Artificial Live IV Proceedings, eds. Rodney Brooks and Patti Maes, MIT Press, 1994. Automatic design and manufacture of robotics lifeforms by Hod Lipson and Jordan Pollack, Nature, Vol. 406, 2000.
	Feb 12		Midterm Project
5	Feb 17		Learning and Evolution by Stefano Nolfi and Dario Floreano, Autonomous Robots, vol. 7.1, 1999.
	Feb 19		Lab: Continue Midterm Project
6	Feb 24		Introduction to developmental robotics by Lisa Meeden and Douglas Blank, Connection Science, vol 18.2, 2006. Also read sections 1-3 of next week's paper.
	Feb 26		Midterm Project checkpoint demonstration
7	Mar 03		Developmental Robotics: A survey by Max Lungarella, Giorgio Metta, Rolf Pfeifer, and Giulio Sandini, Connection Science, vol. 15.4, 2003.
	Mar 05	No class
	Mar 10	Spring Break
	Mar 12
8	Mar 17		Intrinsic motivation systems for autonomous mental development by Peirre-Yves Oudeyer, Frederic Kaplan, Verena Hafner, IEEE Transactions on Evolutionary Computation, vol. 11.2, 2007.
	Mar 19		Lab4: Experimenting with IAC
9	Mar 24		Sensory flow segmentation using a RAVQ by Fredrik Linaker and Lars Niklasson, in Advances in Pattern Recognition, 2000.
	Mar 26	Last day to declare CR/NC or withdraw (Mar 27)	Lab5: Unsupervised categorization
10	Mar 31		Final Project
	Apr 02		Developing navagation behavior through self-organizing distinctive state abstraction by Jefferon Provost, Benjamin Kuipers, and Risto Mikkulainen, Connection Science, vol. 18.2, 2006. A growing neural gas learns topologies by Bernd Fritzke, in Advances in Neural Information Processing Systems 7, 1995.
11	Apr 07		An emergent framework for self-motivation in developmental robotics by James Marshall, Douglas Blank, and Lisa Meeden, in the proceedings of The Third International Conference on Development and Learning, 2004.
	Apr 09		Final Project checkpoint demonstration
12	Apr 14		Vist by Swat alum John Rieffel
	Apr 16		Lab: Work on Final Project
13	Apr 21		Presentations by: 1. Roby 2. Ryan
	Apr 23		Presentations by: 1. Alex, Andrew, and Raul 2. Phyo
14	Apr 28		Presentations by: 1. Derek, Erick, and Malcom 2. Amber and Anne Marie
	Apr 30	Paper due at 4pm, Seminar Party at 5pm (May 12)	Presentations by: 1. Ashley, Maria, and Rachel 2. Madeleine

Final Projects

• Housebreaking your robot: The observe-train-perform algorithm, Madeleine Abromowitz
• Evolutions of communication, Alex Bell, Andrew Pace, and Raul Santos
• Curiously adaptive growing intelligent neural gas: A hybrid approach to category-based intrinsic motivation, Maria Kelly, Rachel Lee, and Ashley Oudenne
• Extending robot soccer using NEAT, Phyo Thiha
• Maze solving by learning state topologies, Derek Tingle, Erick Ball, and Malcolm Augat
• Using NEAT to teach a crawling robot to follow a light, Roby Velez
• Curiosity as a survival technique, Amber Viescas and Anne-Marie Frassica
• Cooperative multi-step behavior in an evolved robot team, Ryan Walker