Transcript Probabilistic Robotics

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
© sebastian thrun, CMU, 2000
1
Administrative Information








Sebastian Thrun
Josh Bao
Web:
Email list:
Time:
Location:
Appointments:
[email protected]
[email protected]
http://robots.stanford.edu/cs226
tba
Mon/Wed, 9:30-10:45am
380 X
Mon 1:30-3:00 (Sebastian)
tba (Josh)
© sebastian thrun, CMU, 2000
2
© sebastian thrun, CMU, 2000
3
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
© sebastian thrun, CMU, 2000
4
What this course is not
 Intro to robotics
 Little work
 Low on math
© sebastian thrun, CMU, 2000
5
Course Schedule
Localization
March31-April 14
Mapping
April 21-May 5
Decision Making
May 10-May26
Multi-Agent
May 17
© sebastian thrun, CMU, 2000
6
What You Should Do
 Think
 Think differently
 Be critical
 Come up with Original Research
© sebastian thrun, CMU, 2000
7
What Is A Good Project
 tbd
• Haptic Mapping
• Learning Models of Outdoor Traffic Flow
© sebastian thrun, CMU, 2000
8
Requirements
 On your own
• Written assignment(s)
• Warm-up project (mobile robot localization)
• Midterm exam
 In teams of three:
• Research Project
© sebastian thrun, CMU, 2000
9
Your next tasks
 Check out Web site
• Read assigned paper
• Download map+sensor data and program robot
localization algorithm
 Come to class on April 5th (9:30am-10:45am)
© sebastian thrun, CMU, 2000
10
© sebastian thrun, CMU, 2000
11
© sebastian thrun, CMU, 2000
12
© sebastian thrun, CMU, 2000
13
© sebastian thrun, CMU, 2000
14
Five Sources of Uncertainty
Approximate
Computation
Environment
Dynamics
Random
Action Effects
Inaccurate
Models
Sensor
Limitations
© sebastian thrun, CMU, 2000
15
Trends in Robotics
Classical Robotics (mid-70’s)
• exact models
• no sensing necessary
Reactive Paradigm (mid-80’s)
• no models
• relies heavily on good sensing
Hybrids (since 90’s)
• model-based at higher levels
• reactive at lower levels
Probabilistic Robotics (since mid-90’s)
• seamless integration of models and sensing
• inaccurate models, inaccurate sensors
© sebastian thrun, CMU, 2000
16
© sebastian thrun, CMU, 2000
17
Rhino
© sebastian thrun, CMU, 2000
18
Minerva
© sebastian thrun, CMU, 2000
19
The CMU/Pitt Nursebot Initiative
© sebastian thrun, CMU, 2000
20
People Detection
Mike Montemerlo
© sebastian thrun, CMU, 2000
21
Learning Models of People
Maren Bennewitz
© sebastian thrun, CMU, 2000
22
3D Mapping Result
With: Christian Martin
© sebastian thrun, CMU, 2000
23
Multi-Robot Exploration
© sebastian thrun, CMU, 2000
24
Helicopter Control
© sebastian thrun, CMU, 2000
25
Mine Mapping
© sebastian thrun, CMU, 2000
26
Campus Navigation
© sebastian thrun, CMU, 2000
27
NASA DART site
© sebastian thrun, CMU, 2000
28
Campus Map (in Progress)
© sebastian thrun, CMU, 2000
29
What are interesting problems?




Mapping, automatic, manual, guided?
Probabilistic localization, landmarks?, odometer!,
Route planning, collision avoidance
Multi-robot sensor fusion, cooperation
© sebastian thrun, CMU, 2000
30
How can we solve them?
© sebastian thrun, CMU, 2000
31
© sebastian thrun, CMU, 2000
32
Where Am I/?
© sebastian thrun, CMU, 2000
33
Nature of Sensor Data: Uncertainty
Odometry Data
© sebastian thrun, CMU, 2000
Range Data
34
© sebastian thrun, CMU, 2000
35
Warm-Up Assignment: Localization,
Due April 14, 04
© sebastian thrun, CMU, 2000
36
Warm-Up Assignment: Localization
© sebastian thrun, CMU, 2000
37
Warm-Up Assignment: Localization
© sebastian thrun, CMU, 2000
38