King Fahd University of Petroleum & Minerals
College of Computer Sciences and Engineering
Computer Engineering Department

 

AUTONOMOUS ROBOTICS

COE 499  

    FALL 2008  

                                                   Watch the Robocup 2007 Germany-Japan Game

RoboCup is an international joint project to foster AI and intelligent robotics research by providing a standard problem. The ultimate goal of RoboCup is to develop by 2050 a team of fully autonomous humanoid robots that can win against the human world champion team in soccer. This idea will promote advancing human knowledge about autonomous robotics and will culminate by an intelligent robot that will assist human in many aspect of life. The educational approach is estimated to be one of the most effective because it teaches sophisticated robotic intelligence design through the programming of a robotic game, e.g. teaching by playing.

     

       http://www.ccse.kfupm.edu.sa/~mayez/ps-coe499/coe499-syllabus.html

Course Description:

Taxonomy of robots, intelligent humanoid robots. RoboCup: goal, application, and technologies. Robcup Leagues. Control algorithms, machine vision, sensing and localization, real-time distributed computing, real-time ad-hoc networking, mechanical design, Machine learning, and autonomous multiagent systems. Robot mechanisms and control, leg movements, forward and inverse geometric model of articulated systems, redundancy, singularities, and multiple solutions. Sensing and perception, humanoid vision, scan methods and optimization, color analysis, and algorithms for the detection of the ball, goal, beacons, and field lines. Monte-Carlo Self-localization. Autonomous robotics, control and planning architectures, behavior-based control programming, the reactive, subsumptive, and deliberative control behaviors. Design of intelligent behavior for soccer games and programming. Integration of motion, sensing, and behavior.

Course Outline:  

1.        Introduction (4 lectures): History of robotics, Control algorithms, machine vision, sensing and localization, real-time distributed computing, real-time ad-hoc networking, mechanical design, Machine learning, and autonomous multiagent systems.

2.      Motion coordination (10 lecures): Linear algebra, the forward geometric model, inverse model, examples, redundancy, singularities, multiple solutions, and robot programming. Robot mechanisms and control, leg movements.

3.       Sensing and perception (8 lectures), humanoid vision, scan methods and optimization for real-time applications, color analysis, and algorithms for the detection of the ball, goal, beacons, and field lines. Monte-Carlo Self-localization.

4.        Self-localization(8 lectures), Monte-Carlo approach to localization, optimization, regeneration, and convergence.

5.      Autonomous robotics (15 lectures): model of autonomous robotics, control and planning architectures, behavior-based control programming. Design of intelligent behavior for soccer games.  

Proposed Projects:

  1. Programming the Kondo KHR (Download Demo 1 (6.5MB), Demo 2 (22 MB), and Demo3 (11 MB)) humanoid robot for Soccer Gaming (Robocup):  programming the motion, vision, self-localization, control and behavior-based control for soccer game. To work towards the Robocup Goal, annual robot competitions are held. The Humanoid League (http://www.humanoidsoccer.org/index.html) competition is one option in the International RoboCup Competition. This competition is attracting student participation from all around the world specifically from Universities in North America, Europe, Japan, Corea, China, and Iran.

Kondo KHR-1 Humanoid Robot Kit

Kondo KHR-1 Humanoid Robot Kit

In the Humanoid League, autonomous robots with a human-like body plan and human-like senses play soccer against each other. In addition to soccer games, penalty kick competitions and technical challenges will take place. The robots are divided into two size classes: KidSize (30-60cm height) and TeenSize (65-130cm height). Dynamic walking, running, and kicking the ball while maintaining balance, visual perception of the ball, other players, and the field, self-localization, and team play are among the many research issues investigated in the Humanoid League.

  1. Distributed robotic intelligence using the Stargate Sensor Network System. Inter-robot communication is supported using INTEL Xscale Stargate (Crossbow) operating as a distributed embedded processor node each has wireless communication capabilities. Distributed auction for multi-robot task allocation in a power aware system. Eight Stargate Sensor Network nodes are available.

  2. Wireless Sensor Network for relaying sensor data to a client using the MICA2/DOT Professional KIT for robot sensor network (MOTE-KIT 5x4x) which is also available.

  3. Free project of your choice

Course Outcomes

 

Course Learning

Outcomes

 

Outcome Indicators and Details

 

O1. Ability to apply knowledge of mathematics and science in robotics.

         Linear algebra, frame, linear operators, translation, rotation, and properties.

         General methods for deriving the forward and inverse geometric model for robot arms.

         Discuss multiple solutions and singularities.

O2. Ability to design robot motion coordination and autonomous robot.

         Design of forward geometric model, inverse model, for 3 or 6 dof robot arms and solve for redundancy, singularities, and multiple solutions.

         Design the robot effector representation and tool motion coordination.

         Design of real-time client-server

         Design of simple distributed intelligence autonomous robots.

          

O3. Ability to identify, formulate, and solve robotics problems.

         Formulate Internet telerobot as  a client-server using advanced real-time control and data remoting schemes (MSF .NET).

         Formulate distributed robotics as a distributed component software

         Formulate distributed intelligence using subsumptive programming schemes.

O4. Ability to use software development tools.

 

 

         Use of  Stargate wireless embedded computing system

         Program client-server application on stargate

         Program wireless communication among distributed modules.

O5. Ability to engage in self-learning.

 

         Demonstrates reading, writing, listening and speaking skills.

         Identifying, retrieving, and organizing information.

         Following a learning plan.

         Demonstrate critical thinking skills such as applying the facts, formulas, theories, etc. to everyday situations.