robocup

Uploaded from authorPOINTLite
Views:
 
Category: Education
     
 

Presentation Description

No description available.

Comments

By: mohamad.bdiwi (39 month(s) ago)

how can i download this presentation, thanks

Presentation Transcript

Introducing Simulation Middle Size: A New Soccer League to RoboCup: 

Introducing Simulation Middle Size: A New Soccer League to RoboCup Mansour Jamzad Mahdi Asadpour

Table of Contents: 

Table of Contents What is RoboCup? What is the problem? What is our solution? Conclusion

1. What is RoboCup?: 

1. What is RoboCup?

What is RoboCup? 1.1 Research goal: 

What is RoboCup? 1.1 Research goal An international research and education initiative Its goal is to foster artificial intelligence and robotics research by providing a standard problem where a wide range of technologies can be examined and integrated.

What is RoboCup? 1.2 Ultimate goal: 

What is RoboCup? 1.2 Ultimate goal "By 2050, a team of fully autonomous humanoid robot soccer players shall win a soccer game, complying with the official FIFA rules, against the winner of the most recent World Cup of Human Soccer."

What is RoboCup? 1.3 Participated teams: 

What is RoboCup? 1.3 Participated teams

What is RoboCup? 1.4 Soccer league: 

What is RoboCup? 1.4 Soccer league The main focus on competitive football RoboCup Soccer is divided into: Simulation Leagues: Coach 2D and 3D Real Robot Leagues: Small-size robot league Middle-size robot league Four-legged (AIBO) robot league Humanoid league

What is RoboCup? 1.5 Simulation league: 

What is RoboCup? 1.5 Simulation league Main Components: Soccer Server Players as clients Monitor Players do not have physical capabilities.

What is RoboCup? 1.6 Real Robot league (Middle Size): 

What is RoboCup? 1.6 Real Robot league (Middle Size) Play soccer in teams of from 4 to 6 robots An orange soccer ball Robots with different styles

What is the problem?: 

What is the problem?

What is the problem? 2.1 Fact: 

What is the problem? 2.1 Fact The final aim is real robotics. The main goal of any simulation is to facilitate the trial and evaluation of real world ideas.

What is the problem? 2.2 Difference: 

What is the problem? 2.2 Difference 1. How to recognize the environment elements and their positions according to soccer field: Real: use image processing algorithms Simulation: there is a Soccerserver which provides the field, ball and players position information.

What is the problem? 2.2 Difference, cont.: 

What is the problem? 2.2 Difference, cont. 2. Hardware or software failures Real: may damage during the game, impose serious damage to others Simulation: these failures are not considered. 3. Designing and implementing advanced algorithms Real: limits in implementation of advanced algorithms Simulation: there has been many advances based on AI methods and multiagent systems.

What is the problem? 2.3 Big gap: 

What is the problem? 2.3 Big gap A program written in the simulation league can not be easily transferred to a real robot or used it in the real robot league. So there is a gap between the simulation and real robot leagues.

3. What is our solution?: 

3. What is our solution?

What is our solution? 3.1 Preamble: 

What is our solution? 3.1 Preamble Our experience in Middle Size league; champion in RoboCup-99 Our implementation of Simulation Middle Size proves that such league is possible, and it’s not only an idea! Simulations tools: Webots Player/Stage Familiar with Webots usage which enabled us to develop in a short period of time.

What is our solution? 3.2 Webots: 

What is our solution? 3.2 Webots Cyberbotics Ltd., http://www.cyberbotics.com Provides a rapid prototyping environment for modeling, programming and simulating mobile robots. Using the libraries of this software one can transfer his control programs to several commercially available real mobile robots. Good samples and documentation

What is our solution? 3.3 Elements of suggested league: 

The field of play The ball Players The referee Automated referee Human referee What is our solution? 3.3 Elements of suggested league

What is our solution? 3.4 How to participate: 

Each team should provide: The hardware model of own robots in VRML97 format The robots controller program for that modeled robot What is our solution? 3.4 How to participate

What is our solution? 3.5 Model: 

Should be in VRML97 format: Virtual Reality Modeling Language version 97 3D Studio Max, Maya, AutoCAD include the VRML97 export feature Arvand player: Front camera Kicker Four servo motors Castor wheel Communication Arvand goalkeeper: Two castor wheels What is our solution? 3.5 Model

What is our solution? 3.6 Controller program: 

Written in C/C++ or Java To control each part of the modeled robot This simple controller enables goalkeeper to recognize the ball position so it tries to prevent the ball from passing over the goal line What is our solution? 3.6 Controller program

What is our solution? 3.7 The match routine: 

Human referee records the names of two teams Runs the simulation Automated referee Imports the robot models to the Simulation Middle Size Places the players in soccer field Kick off Saves the match events Finishes the match and shows the winner What is our solution? 3.7 The match routine

What is our solution? 3.8 Movies: 

What is our solution? 3.8 Movies Prototype:

What is our solution? 3.8 Movies, cont.: 

What is our solution? 3.8 Movies, cont. Kick:

What is our solution? 3.8 Movies, cont.: 

What is our solution? 3.8 Movies, cont. Cooperation:

What is our solution? 3.8 Movies, cont.: 

What is our solution? 3.8 Movies, cont. Flagpost:

What is our solution? 3.8 Movies, cont.: 

What is our solution? 3.8 Movies, cont. Referee:

4. Conclusion: 

4. Conclusion

Conclusion 4.1 Solution vs. problem: 

This league is much close to the real. 1. How to recognize the environment elements and their positions according to soccer field: Real: use image processing algorithms Simulation: use a Soccerserver Simulation Middle Size: use image processing algorithms but simplified and noise free Conclusion 4.1 Solution vs. problem

Conclusion 4.1 Solution vs. problem, cont.: 

2. Hardware or software failures Real: robots may damage during the game Simulation: not considered Simulation Middle Size: use fault injection mechanism Conclusion 4.1 Solution vs. problem, cont.

Conclusion 4.1 Solution vs. problem, cont.: 

3. Designing and implementing advanced algorithms Real: limits in implementation of advanced algorithms Simulation: many advances based on AI methods Simulation Middle Size: as powerful as simulation The rules that are not currently applied in Middle Size e.g. Offside, Penalty Kick, can be applied here to evaluate its practical effectiveness. Conclusion 4.1 Solution vs. problem, cont.

Conclusion 4.2 Finally: 

We suggested to start a new league called Simulation Middle Size We believe that this league can fill the big gap between simulation league and real robot league so that the programs developed for the new league can be easily ported to real robots. We believe that if the suggested league is presented as a new match in RoboCup, we will see a high improvement in the real robots league. Conclusion 4.2 Finally

Conclusion 4.3 Suggestion for future works: 

Small Size league Humanoid league Four-legged (AIBO) league Conclusion 4.3 Suggestion for future works

Thank you very much for your time: 

Thank you very much for your time Mahdi Asadpour (asadpur@ce.sharif.edu) Computer Engineering Department, Sharif University of Technology. http://ce.sharif.edu/~asadpur/robocup.html