1. Research on Robot Soccer and what PSU can do

2. The perceptions n Three perceptions: see, hear and sense_body. n All related to one system of perception. n Representation of perceptions: (see 1 (ball …) …)

3. Visual Perception n see. n List of objects recognized: –type, –direction, –distance, –speed, –number. n Lines, gates, boundaries, ball.

4. sense_body sense_body n Force detection n Energy available. n Principal actions: –dash, –kick, –turn –say. Actions

5. –Role level : Determines the roles of each robot. (defender, attacker and goal keeper) (defender, attacker and goal keeper) –Action level : Selects actions of each robot. (shooting, blocking, dribbling, etc) (shooting, blocking, dribbling, etc) –Behavior level : Move and obstacle avoidance –Execution level : Motor control Role level Variants of Control structure

6. Vision-based system Vision-based system –Remote brainless system –Brain-on-board system Robot-based system Robot-based system Selection guidelines Selection guidelines –Developer’s interests –Computational capabilities of host computer and vision system –Capabilities of the robots –Cost The system can be classified using the location of intelligence Classification of Robot Soccer Systems Classification of Robot Soccer Systems

7. Centralized system Centralized system Simple and inexpensive Easy to develop the robot Simple and inexpensive Easy to develop the robot No local sensors. No local sensors. Fast computing time Fast computing time and sampling time and sampling time Easy to debug and upgrade the program Easy to debug and upgrade the program Remote-Brainless System Remote-Brainless System High cost vision system and host computer A type of vision-based system A type of vision-based system Intelligent part is implemented in the host computer. Intelligent part is implemented in the host computer.

8. Robots Robots The robots consist of driving mechanism, communication part, and computational part for velocity and for processing the data received from a host computer Host computer Host computer All the calculations for vision data processing, strategies, position control of robots and so on, are done in the host computer which controls robots like radio -controlled car Remote- Brainless System Remote- Brainless System

9. Brain-on-board system Brain-on-board system Intermediate level between the centralized and the distributed systems / between the remote-brainless and the robot based systems. Intermediate level between the centralized and the distributed systems / between the remote-brainless and the robot based systems. Robots can use local sensors to move to the goal and to avoid the opponent. Robots can use local sensors to move to the goal and to avoid the opponent. Can decompose the system into high level (host computer) Can decompose the system into high level (host computer) and low level (robot systems). and low level (robot systems). Easy to make Easy to make the system in modular form the system in modular form A type of vision-based system A type of vision-based system Intelligence is partially implemented in the host computer and robots. Intelligence is partially implemented in the host computer and robots.

10. Robots Robots The robots have functions such as velocity control, position control, obstacle avoidance, etc. Host computer Host computer The host computer processes vision data and calculates next behaviors of robots according to strategies and sends commands to the robots using RF modem. Brain- on- board system Brain- on- board system

11. Robot-based system Robot-based system Suitable when the large number of agents exist Suitable when the large number of agents exist Complex and expensive Complex and expensive Need communication among robots Need communication among robots Distributed system Distributed system Intelligent part is implemented in the robots. Intelligent part is implemented in the robots.

12. Robots Robots The robots decide their own behavior autonomously using the received vision data, own sensor data and strategies. Host computer Host computer The host computer processes only vision data can be considered as a kind of sensor. Robot- based system Robot- based system

13. Main PC Serial Port Serial Port –Select the serial communication port Home Goal Home Goal –Select the home side on the screen Find Objects Find Objects –Check the box of which you like to find on the field Initial Position: tell the vision system the initial position Initial Position: tell the vision system the initial position of each object of each object –E.g.) for the ball i) turn on the radio button of ‘Ball’ i) turn on the radio button of ‘Ball’ ii) place the mouse on the ball and press the left button ii) place the mouse on the ball and press the left button –Repeat above procedure for another object EXAMPLE

14. Main PC Select Situation Select Situation –The situation in which the game is about to start Command Command –Click ‘Ready’: the vision system starts finding the objects on the field on the field –Click ‘Start’ : the vision system starts sending commands to the robots to the robots –Click ‘Stop’ : the vision system stops finding objects and sending commands and sending commands EXAMPLE

15. Communication (Infra-red) Communication (Infra-red) Infra-red Communication Infra-red Communication –Four transmitters are used to cover the whole field EXAMPLE

16. Communication Both teams share the same transmitter via a mediator Both teams share the same transmitter via a mediator Communication Packet Communication Packet –Three 0xFFs: the start of a packet –0x0F (0xF0): Team A (Team B) –V Li, V Ri : left and right wheel velocity of robot i –0xAA: end of velocity data of each robot EXAMPLE

17. Tough, practical problems to be solved n Real-Time image processing n Sensor fusion (sonar, touch, vision, light, other). n Motors, batteries. n Sophisticated control strategies n Radio communication n Agent communication

18. Problems Problems n Players do not know absolute locations. n Points of reference are: boundaries, lines and gates. n Conflicts of players n Limited visibility n Limited communication

19. Software n Real-Time Problem, rule-based, agent behavior. n Dynamic planning and execution of plans in real-time. n Cooperation and competition. n No precise information n Non-deterministic behavior: –results of actions are uncertain.

20. Software potentials (cont.) n Voice comunication of players n Various players have different rules and behaviors, different strategies and implementations - rule-based, neural nets, fuzzy logic, etc. n multi-agents: competition versus collaboration. n on-line versus off-line, n individual versus group behaviors.

21. FIRA and the ECE 478/ECE 479 class at PSU n Class project and in future Capstone Project (if a company will sponsor) n High-School project n Portland Cyber Theatre n Experience in Visual C++, Basic, Lisp and Prolog programming. n A lot of fun. n Travel to Japan or Korea?

22. Does PSU team has a chance? n 1. Professor Kim from KAIST in Korea invited us to create a new league of walking robots n 2. We have experience with walking robots n 3. We will try to propose new competition ideas and have a leverage of knowing them earlier. –Besides, our students are smart…..? Let us look to our robots…...

23. What we propose Robot soccer system Robot soccer system –Intelligent control system –Multi-agent system Composition of robot soccer system Composition of robot soccer system –Mobile robots –Host computer –Vision system –Communication module Sports commentator Walking robots

24. Team One n Complex robots n 8 to 12 servos

25. Our walking quadrupeds and hexapods Collaborations: ATR, Japan Technical University of Warsaw, Poland Technical University of Gliwice, Poland KAIST, Korea Multi-national team

26. Quadruped

27. Basic Radio-Controlled Spider Hexapod with Gripper

28. Spider with a camera

30. New soccer-specialized hexapod

31. Hexapod’s Soccer Kicker

32. Team Two n Simple robots - hexapods n 2 servos

33. SoccerBot

34. Final design of a quadruped walker

35. Block Diagram of the Simple Robot System Block Diagram of the Simple Robot System Logic Power PWM Right PWM Left Motor Driver Motor Driver Motor Voltage Regulator Motor Power Micro-Controller Battery Communication Signal Communication Module Motor Part

36. Sport Commentators Bug

37. Virginia Woolf

38. Jonas

39. Marvin the Crazy Robot

40. Class Projects - Winter/Spring 2002 n 1. Robot position, orientation and conflict detection using top-mounted camera n 2. Ball recognition and vision interface to the server n 3. Server and user interface n 4. Egoistic soccer-playing strategy n 5. Collaborative soccer-playing strategy n 6. Recovery from body conflicts such as leg entanglements of robots

41. Class Projects - Winter/Spring 2002 n 7. Design of robo-head sport commentator: speech and natural language generation. n 8. Mechanical improvements to robots n 9. Design of a special soccer player hexapod.

42. Image Processing: find position, orientation and conflicts of a walker

43. Filtering, histogramming, Hough transform, equations Localization, orientation, conflicts (leg entanglements)

44. Robo Soccer - Why is it so cool? n Is Robot Soccer useful? n Is the result of Robot Soccer useful? n Is Robot Soccer the worlds largest playground? n Is there money to be made with robots to play sports?

45. Why are Robot Sports so cool? n Captures the essence of soccer. (That’s why there can be a lot of people cheering on the teams) n ”Hacker-party” more than a dull academic conference….that is, it is a lot of activity. n Interesting problem in co-ordinating 11 processes to achieve a common goal. n The width of the field, from Robotic-nerds to Sociologists.

46. n In the simulator league, you are forced to learn many concepts: – for example network communication, multi-threading, agents, hardware, etc. n Testbed for AI-algorithms, sociological theories. n Fairly simple environment. n A limited set of rules (also predefined), but always close to real world problems. Why are Robot Sports so cool?

47. What are the problems with Robot Competitions ? n Too much work before reaching an interesting research level n Too much focus on low level implementation (still) n Still focus on competing instead of comparing of strategies.

48. Research n We do research in several areas: –Machine learning (constructive induction) –Decision theory –Social agents –walker’s gaits evolution –+more n In all of the above we plan to use robot sports as a testbed, or applied area.

49. Conclusion n Robot sports areTHE new standard problem within AI. n 1500 researchers world-wide. n Focuses on interdisciplinary co-operative work between the researchers as well as co-operation between the agents. n New experience for PSU

50. Current class n 4 ME students n 3 CS students n 3 EE students Using existing robots CS and EE students develop software ME students develop 2 new robot prototypes and kits with good documentation CS and EE students build new robots from kits ME students adopt software to new robots and learn programming 6 robots of two types build and tested in July

51. Current class n 4 ME students n 3 CS students n 3 EE students Using existing robots CS and EE students develop software ME students develop 2 new robot prototypes and kits with good documentation CS and EE students build new robots from kits ME students adopt software to new robots and learn programming 6 robots of two types build and tested in July

52. Technical Publications n Hiroaki Kitano, Masahiro Fujita, Stephane Zrehen, and Koji Kageyama, "Sony Legged Robot for RoboCup Challenge", –In Proceedings of the IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION, IEEE, 1998, pp.2605-2612 n Manuela Veloso, William Uther, Masahiro Fujita, Minoru Asada, and Hiroaki Kitano, "Playing Soccer with Legged Robot", –In Proceedings of the INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, 1998, pp.437-442 Let us go deeper

53. How to find more about RoboCup? Web Pages: Web Pages: http://medialab.di.unipi.it/Project/Robocup n http://www.robocup.org n http://www.dsv.su.se/~robocup