Intelligent Autonomous Robotics: A Robot Soccer Case Study

✍ Scribed by Peter Stone

• Publisher: Springer
• Year: 2007
• Tongue: English
• Leaves: 150
• Series: Synthesis Lectures on Artificial Intelligence and Machine Learning
• Edition: 1
• Category: Library

✦ Synopsis

Robotics technology has recently advanced to the point of being widely accessible for relatively low-budget research, as well as for graduate, undergraduate, and even secondary and primary school education. This lecture provides an example of how to productively use a cutting-edge advanced robotics platform for education and research by providing a detailed case study with the Sony AIBO robot, a vision-based legged robot. The case study used for this lecture is the UT Austin Villa RoboCup Four-Legged Team. This lecture describes both the development process and the technical details of its end result. The main contributions of this lecture are (i) a roadmap for new classes and research groups interested in intelligent autonomous robotics who are starting from scratch with a new robot, and (ii) documentation of the algorithms behind our own approach on the AIBOs with the goal of making them accessible for use on other vision-based and/or legged robot platforms.

✦ Table of Contents

Cover
Copyright Page
Title Page
Contents
Introduction
The Class
Initial Behaviors
Vision
Camera Settings
Color Segmentation
Region Building and Merging
Object Recognition with Bounding Boxes
Position and Bearing of Objects
Visual Opponent Modeling
Movement
Walking
Basics
Forward Kinematics
Inverse Kinematics
General Walking Structure
Omnidirectional Control
Tilting the Body Forward
Tuning the Parameters
Odometry Calibration
GENERAL MOVEMENT
Movement Module
Movement Interface
High-Level Control
Learning Movement Tasks
Forward Gait
Ball Acquisition
Fall Detection
Kicking
Creating the Critical Action
Integrating the Critical Action into the Walk
Localization
Background
Basic Monte Carlo Localization
MCL for Vision-Based Legged Robots
Enhancements to the Basic Approach
Landmark Histories
Distance-Based Updates
Extended Motion Model
Experimental Setup and Results
Simulator
Experimental Methodology
Test for Accuracy and Time
Test for Stability
Extended Motion Model
Recovery
Localization Summary
Communication
Initial Robot-to-Robot Communication
Message Types
Knowing Which Robots Are Communicating
Determining When A Teammate Is ``Dead''
Practical Results
General Architecture
Global Map
Maintaining Location Data
Information from Teammates
Providing a High-Level Interface
Behaviors
Goal Scoring
Initial Solution
Incorporating Localization
A Finite State Machine
Goalie
Coordination
Dibs
Relevant Data
Thrashing
Stabilization
Taking the Average
Aging
Calling the Ball
Support Distance
Phasing out Dibs
Final Strategy
Roles
Supporter Behavior
Defender Behavior
Dynamic Role Assignment
Simulator
Basic Architecture
Server Messages
Sensor Model
Motion Model
Graphical Interface
UT Assist
General Architecture
Debugging Data
Visual Output
Localization Output
Miscellaneous Output
Vision Calibration
Conclusion
Heuristics for the Vision Module
Region Merging and Pruning Parameters
Tilt-Angle Test
Circle Method
Beacon Parameters
Goal Parameters
Ball Parameters
Opponent Detection Parameters
Opponent Blob Likelihood Calculation
Coordinate Transforms
Walking Parameters
Kicks
Initial Kick
Head Kick
Chest-Push Kick
Arms Together Kick
Fall-Forward Kick
Back Kick
TCPGateway
Extension to World State in 2004
Simulator Message Grammar
Client Action Messages
Client Info Messages
Simulated Sensation Messages
Simulated Observation Messages
Competition Results
American Open 2003
RoboCup 2003
Challenge Events 2003
U.S. Open 2004
RoboCup 2004
U.S. Open 2005
RoboCup 2005
References
Biography