Adaptive Strategies in RoboCup Soccer University of Hamburg Faculty - - PowerPoint PPT Presentation

▶
adaptive strategies in robocup soccer
SMART_READER_LITE
LIVE PREVIEW

Adaptive Strategies in RoboCup Soccer University of Hamburg Faculty - - PowerPoint PPT Presentation

Sep 30, 2023 120 likes •372 views

MIN Faculty Department of Informatics Adaptive Strategies in RoboCup Soccer University of Hamburg Faculty of Mathematics, Informatics and Natural Science Department of Informatics Technical Aspects of Multimodal Systems v 12.11.18 | L UCA K

slide-1
SLIDE 1

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

University of Hamburg Faculty of Mathematics, Informatics and Natural Science Department of Informatics Technical Aspects of Multimodal Systems

Adaptive Strategies in RoboCup Soccer

slide-2
SLIDE 2

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 1

Goals of todays presentation

Introduction of RoboCup Soccer and its leagues Introduction of different attacking and defending approaches and delimitation of each other Evaluation of attacking and defending approaches

slide-3
SLIDE 3

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 2

Content

RoboCup - Soccer Attacking and defending approaches Conclusion Future Competition 01 02 03 04

slide-4
SLIDE 4

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 3

01 RoboCup - Soccer

Established 1995 Topic Description Idea

  • Playing football by combine robotics and artificial intelligence
  • Team of fast-moving robots under a dynamic environment
  • Soccer can be simplified and is reducible in labs

Location Hosting a competitive tournament in different locations world wide Alternative forms

  • f AI & robotics

Rescue | @Home | Logistics | Junior Vision Beating the FIFA World Soccer Team by 2050 with fast-moving robots

slide-5
SLIDE 5

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 4

Teams compete in different leagues

01 Simulation

  • Focus on artificial

intelligence and simulation

  • Teams play on a virtual

field

  • 2D or 3D

Source: www.robocup.org

02 Small Size

  • Use of intelligent multi-

robot/agent coordination and control in a highly dynamic environment

Source: www.robocup.org

03 Middle Size

  • Fully autonomous robots
  • Defined measures and

weight for robots

  • Focus on mechatronics

design, control and multi-agent cooperation at plan and perception levels

Source: www.robocup.org

04 Standard Platform

  • Fully autonomous robots

with individual decision making process

  • Usage of standardized

robots (NAO robot from SoftBank)

Source: www.robocup.org

05 Humanoid

  • Fully autonomous robots

with human like body plan

  • Subdivision in KidSize,

TeenSize and AdultSize

Source: www.robocup.org

Focus of presentation

Source: https:/www.robocup.com

slide-6
SLIDE 6

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 5

02 Small Size League (SSL)

Characteristics Description

Playing field

Carpeted playing field with dimensions of 12m length and 9m width

Robots

  • Specification: F180 regulation (fit in an 180mm diameter circle and maximum height
  • f 15cm)
  • robotic equipment is fully autonomous
  • intelligent multi-robot/agent coordination and control in a highly dynamic

environment with a hybrid centralized system

Positioning

  • Centralized localization solution (information from two cameras above playing field)
  • Identification of each individual robot through patterns by camera (All robots of a

team are controlled by central control unit (CCU) / Orange ball)

Communication

central program “SSL-Vision” distributed a local network connected to vision servers and the two team | Off-field computer to communicate referee commands and position information to robots

Decision Making

Individual decision making by robots

Source: https://tigers-mannheim.de/download/papers/2011-A%20Simulation%20for%20the%20RoboCup%20Small-Size-League%20I-Leinemann.pdf

slide-7
SLIDE 7

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 6

02 Small Size League (SSL) – Exemplary game field

1

Robots

2

Physical world

3

Positioning system

4

Referee software

5

Central control unit/software

Source: https://tigers-mannheim.de/download/papers/2011-A%20Simulation%20for%20the%20RoboCup%20Small-Size-League%20I-Leinemann.pdf

slide-8
SLIDE 8

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 7

03 Middle Size League (MSL)

Characteristics Description

Playing field

Carpeted playing field with dimensions of 18m length and 12m width

Robots

  • Specification: 50 x 50 x 80 cm maximal size
  • 40 kg limit
  • All robots are completely autonomous

Positioning

  • Self localization

Communication

  • Wireless networking between each robot, base station and referee box
  • Only human interaction is the referee through a referee box – controlled by an

assistant during the game

Decision Making

  • All robots are completely autonomous
  • Individual decision making by robots
  • Playing cooperative together

Source: https://www.researchgate.net/publication/275040614_RoboCup_MSL_-_History_Accomplishments_Current_Status_and_Challenges_Ahead

slide-9
SLIDE 9

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 8

03 Middle Size League (MSL)

Source: http://robocupdutchopen.org/content/images/dutch%20open%20site/Robocup/MSL.jpg

Description

1 Omnivision unit 2 Front camera 3 PC 4 Ball handling mechanism 5 Kicker 6 Omniwheel platform 1 2 3 4 4 5 6

slide-10
SLIDE 10

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 9

Content

RoboCup - Soccer Attacking and defending approaches Conclusion Future Competition 01 02 03 04

slide-11
SLIDE 11

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 10

Source: https:/www.menschundtechnik.com/bremer-team-b-human-ist-zum-fuenften-mal-weltmeister-im-robocup/

Approaches

Attacking Approach by ZJUNlict

Attacking

01 02

Defending

Defending Approaches by Tech United Eindhoven

slide-12
SLIDE 12

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 12

01 Attacking approaches by ZJUNlict

ZJUNlict

Procedure 1. Identification if ball runs to enemy or not (by

unique motion model) and by what velocity

  • 2. Estimation of defending robots direction &

speed/ velocity to define interception point

  • 3. Calculation/ Identification best player to reach

& play ball

University Zhejiang University (China) Approach Optimized movement by predicting the collision of

robot with the ball Important to predict velocity & direction of ball after impact (bouncing back of obstacle) Improvement

  • to predict the movement of the ball

SSL

Source: https://www.robocup2017.org/file/symposium/soccer_sml_size/Robocupssl2017-final21.pdf

Picture a) Picture b)

slide-13
SLIDE 13

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 17

Evaluation of attacking approaches

ZJUNlict

  • Supports team play
  • Faster able to shoot again
  • Movement of the robots will

be improved

  • Improve task match of each

robot

  • Active movement

Errors appear:

  • Wireless connection

problems

  • Prediction needs to be exact
  • Velocity prediction
  • use same as they shot
slide-14
SLIDE 14

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 19

02 Defending approaches by Tech United

Tech United Eindhoven

Use 1.

Compute the covered angle of the attacker to score a goal 2. Needs two players and goalkeeper to cover the

  • pen angle

University University of Eindhoven Approach Player

Block the direct way to the goal with two robots (a) Only 2 robots are allowed in a scum Compute space around the attacker (b) Second defender is waiting at the edge Ideal both defenders are waiting at the edge Closest defender to the goal covers the middle of the goal Once the set of positions is computed, they need to be assigned to defender

MSL

Source: https://www.techunited.nl/media/images/Champion2016.pdf; http://www.cs.cmu.edu/~mmv/papers/14aamas-cmdragons.pdf

slide-15
SLIDE 15

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 20

02 Defending approaches by Tech United

Tech United Eindhoven

Use Caught all penalties

Till all robots need to rotate it is usable

University University of Eindhoven Approach Goal Keeper

Problem: accuracy and velocity of the shot unable the Goalkeeper to save the ball Identification of shooting direction and allocate the position on goal line (a) Switches the mode to penalty situation Takes a look at the defender it self and not only to the ball (b) Through rotate to the predicted direction possible to catch the ball

MSL

Source: https://www.techunited.nl/media/images/Champion2016.pdf

slide-16
SLIDE 16

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 21

Evaluation of defending approaches

Two defenders Penalty

  • Covers more defending space
  • Minimize the space to pass/shoot
  • f the attacker
  • Good against slow building up

teams

  • Uses two defenders on one

attacker

  • Assign role to attacker
  • Needs to move back (time)
  • Exact computation
  • Saved all penalties
  • Enough time to move to right

positions

  • Needs exact localization of

itself/opponent

  • Based on rotation of the shooting

robot

slide-17
SLIDE 17

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 23

Content

RoboCup - Soccer Attacking and defending approaches Conclusion Future Competition 01 02 03 04

slide-18
SLIDE 18

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 24

Source: http://prolite-lamps.co.uk/wp-content/uploads/2016/06/lights-light-bulb-idea.jpg

Conclusion what to take from those strategies?

slide-19
SLIDE 19

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 25

Conclusion

Every league has their own ideas and their not simple to transfer Strategies are crucial to improve the attack with different approaches There are dynamic parts (two defenders) and more statically parts (predict penalty) Defending strategy(MSL) could be used in SSL / Attacking approach(SSL) could be used in MSL Crucial: exact localization of robots/ball – leagues use different ways

There is not one perfect strategy - the combination of different approaches will lead to win

slide-20
SLIDE 20

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 26

Content

RoboCup - Soccer Attacking and defending approaches Conclusion Future Competition 01 02 03 04

slide-21
SLIDE 21

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 27

Preview

SSL – localization apart from SSL Vision control More reward for passing the ball to team member Future drop in competition – focus on even more communication - dynamical team play Furthers ad hoc team organization à focus more on learning agents

Source: https://www.cs.utexas.edu/~AustinVilla/papers/JAAMAS16-katie.pdf

slide-22
SLIDE 22

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 28

Middle Size League (MSL) – old gameplay

Source: https://commons.wikimedia.org/wiki/File:RoboCupSoccer_Robot_Football_at_2009_German_Open.ogv

slide-23
SLIDE 23

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 29

2018 Tech United vs Humans

Source:https://www.youtube.com/watch?v=KT2e4Z0u2C8

slide-24
SLIDE 24

v

12.11.18 | LUCA KNOBLOCH | 8KNOBLOC@INFORMATIK

MIN Faculty Department of Informatics

ADAPTIVE STRATEGIES PAGE 30

Source: https://www.magdeburg.de/index.php?NavID=37.664&object=tx%7C37.17128.1&La=1

Thank you for listening