2. © sebastian thrun, CMU, 20001 CS226 Statistical Techniques In Robotics Sebastian Thrun (Instructor) and Josh Bao (TA) http://robots.stanford.edu/cs226 Office: Gates 154, Office hours: Monday 1:30-3pm

3. © sebastian thrun, CMU, 20002 Administrative Information  Sebastian Thrun thrun@stanford.edu  Josh Baojbao@stanford.edu  Web: http://robots.stanford.edu/cs226  Email list: tba  Time:Mon/Wed, 9:30-10:45am  Location:380 X  Appointments: Mon 1:30-3:00 (Sebastian)  tba (Josh)

4. © sebastian thrun, CMU, 20003

5. 4 Goals  Enable you to program robots and embedded systems in a robust fashion  Enable you to understand the intrinsic assumptions in your robot software  Enable you to pursue original research in probabilistic robotics  Sway you into joining a young and fascinating research field: probabilistic robotics

6. © sebastian thrun, CMU, 20005 What this course is not  Intro to robotics  Little work  Low on math

7. © sebastian thrun, CMU, 20006 Course Schedule Localization March31-April 14 Mapping April 21-May 5 Decision Making May 10-May26 Multi-Agent May 17

8. © sebastian thrun, CMU, 20007 What You Should Do  Think  Think differently  Be critical  Come up with Original Research

9. © sebastian thrun, CMU, 20008 What Is A Good Project  tbd Haptic Mapping Learning Models of Outdoor Traffic Flow

10. © sebastian thrun, CMU, 20009 Requirements  On your own Written assignment(s) Warm-up project (mobile robot localization) Midterm exam  In teams of three: Research Project

11. © sebastian thrun, CMU, 200010 Your next tasks  Check out Web site Read assigned paper Download map+sensor data and program robot localization algorithm  Come to class on April 5 th (9:30am-10:45am)

12. © sebastian thrun, CMU, 200011

13. © sebastian thrun, CMU, 200012

14. © sebastian thrun, CMU, 200013

15. © sebastian thrun, CMU, 200014

16. © sebastian thrun, CMU, 200015 Five Sources of Uncertainty Environment Dynamics Random Action Effects Sensor Limitations Inaccurate Models Approximate Computation

17. © sebastian thrun, CMU, 200016 Trends in Robotics Reactive Paradigm (mid-80’s) no models relies heavily on good sensing Probabilistic Robotics (since mid-90’s) seamless integration of models and sensing inaccurate models, inaccurate sensors Hybrids (since 90’s) model-based at higher levels reactive at lower levels Classical Robotics (mid-70’s) exact models no sensing necessary

18. © sebastian thrun, CMU, 200017

19. © sebastian thrun, CMU, 200018 Rhino

20. © sebastian thrun, CMU, 200019 Minerva

21. © sebastian thrun, CMU, 200020 The CMU/Pitt Nursebot Initiative

22. © sebastian thrun, CMU, 200021 People Detection Mike Montemerlo

23. © sebastian thrun, CMU, 200022 Learning Models of People Maren Bennewitz

24. © sebastian thrun, CMU, 200023 3D Mapping Result With: Christian Martin

25. © sebastian thrun, CMU, 200024 Multi-Robot Exploration

26. © sebastian thrun, CMU, 200025 Helicopter Control

27. © sebastian thrun, CMU, 200026 Mine Mapping

28. © sebastian thrun, CMU, 200027 Campus Navigation

29. © sebastian thrun, CMU, 200028 NASA DART site

30. © sebastian thrun, CMU, 200029 Campus Map (in Progress)

31. © sebastian thrun, CMU, 200030 What are interesting problems?  Mapping, automatic, manual, guided?  Probabilistic localization, landmarks?, odometer!,  Route planning, collision avoidance  Multi-robot sensor fusion, cooperation

32. © sebastian thrun, CMU, 200031 How can we solve them?

33. © sebastian thrun, CMU, 200032

34. © sebastian thrun, CMU, 200033 Where Am I/?

35. © sebastian thrun, CMU, 200034 Nature of Sensor Data: Uncertainty Odometry Data Range Data

36. © sebastian thrun, CMU, 200035

37. © sebastian thrun, CMU, 200036 Warm-Up Assignment: Localization, Due April 14, 04

38. © sebastian thrun, CMU, 200037 Warm-Up Assignment: Localization

39. © sebastian thrun, CMU, 200038 Warm-Up Assignment: Localization