SLIDE 1 CS 378: Autonomous Intelligent Robotics
Instructor: Jivko Sinapov
http://www.cs.utexas.edu/~jsinapov/teaching/cs378/
SLIDE 2
Announcements
FRI Summer Research Fellowships: https://cns.utexas.edu/fri/beyond-the-freshman-lab/fellowships Applications are due March 1st but apply now! Funding is available for 4-5 students per FRI stream
SLIDE 3
Announcements
Homework 3 is due Friday “night”
SLIDE 4
SLIDE 5 Readings for this week: Behavior-based robotics
- R. Brooks (1986). ``A Robust Layered Control System for a
Mobile Robot'', MIT AI Memo 864, Vol RA-2, No. 1. p. 14-23
- R. Brooks (1991). "Intelligence Without Representation",
Artificial Intelligence, Volume 47 , Issue 1-3
SLIDE 6
Progression
2D simulation 3D simulation 2D simulation Real World
SLIDE 7 The Gazebo 3D simulator
- Install gazebo_ros package:
sudo apt-get install ros-indigo-gazebo-ros
roslaunch gazebo_ros rubble_world.launch
SLIDE 8
Today
1) Behavior-Based Robotics 2) ROS Services (part 2)
SLIDE 9
Behavior-Based Robotics
SLIDE 10
A Bit of History
First Industrial Robot (~60s)
SLIDE 11
Modern Industrial Robots
SLIDE 12
Teleoperation
SLIDE 13
Teleoperation
SLIDE 14
Teleoperation
SLIDE 15
Teleoperation
SLIDE 16
Teleoperation
SLIDE 17
Teleoperation
SLIDE 18
Robotics Timeline
SLIDE 19 Teleoperation vs Telepresence
- An early attempt to improve teleoperation
was to add more cameras / displays
- Telepresence aims for placing the operator
in a virtual reality that mimics the robot's surroundings
SLIDE 20 Telepresence Robots
http://www.pilotpresence.com/wp-content/uploads/2011/01/remote-presence-systemsv2.jpg
SLIDE 21
Telepresence Robots
Example: https://www.youtube.com/watch?v=bVe66UW2XUU
SLIDE 22
The need for (semi-) autonomy
SLIDE 23
How should autonomy be achieved and organized?
SLIDE 24
Robot Primitives
SLIDE 25
The Early Answer (1967): Sense-Plan-Act
SLIDE 26
The Early Answer (1967): Sense-Plan-Act
SLIDE 27
The Early Answer (1967): Sense-Plan-Act
SLIDE 28
Early Example of S-P-A
SLIDE 29
Early Work on Planning
SLIDE 30
Early Work on Planning
SLIDE 31
Early Work on Planning
SLIDE 32
Early Work on Planning
SLIDE 33
A More Realistic Example
SLIDE 34
A More Realistic Example
SLIDE 35
A More Realistic Example
SLIDE 36
A More Realistic Example
Is INROOM(IT,R1) true or false? CONNECTS(D1,R1,R2)? INROOM(IT,R2)? IT
SLIDE 37
Representing Initial State
IT
SLIDE 38
Representing Goal State
IT
SLIDE 39
The “difference” table
SLIDE 40 Logical Difference
SLIDE 41
Eliminating the Difference
SLIDE 42
Eliminating the Difference
SLIDE 43
Eliminating the Difference
SLIDE 44 Discussion
- What are some limitations of planning with
STRIPS?
- Where do the predicates, operators, etc.
come from?
SLIDE 45
Then comes Rodney Brooks...
SLIDE 46 GOFAI
- GOFAI: good old-fashioned artificial
intelligence
- Typically implemented as a central planner
- perating on a set of symbols (predicates)
- Tools: logic, predicate logic, PROLOG,
Search algorithms, etc.
- Solution: sense → model → plan → act
SLIDE 47
Brooks' opinion: GOFAI failed
SLIDE 48
Abstraction is a dangerous weapon
SLIDE 49
Toy worlds vs. Real worlds
SLIDE 50
Toy worlds vs. Real worlds
SLIDE 51
Brook's opinion: GOFAI failed
SLIDE 52
Alternatives to Sense-Plan-Act
Sense-Plan-Act Reactive Hybrid
SLIDE 53
Reactive Paradigm
SLIDE 54
Reactive Paradigm Example
SLIDE 55
Reactive Paradigm Example
SLIDE 56
Reactive Paradigm Example
. . . .
SLIDE 57
Reactive Paradigm Example
SLIDE 58
Reactive Paradigm Example
. . . .
SLIDE 59
The Hybrid Paradigm
SLIDE 60
Functional vs. Behavioral Decomposition
SLIDE 61
Words of Wisdom
SLIDE 62
Where is Brooks now?
SLIDE 63 Credits
- “Introduction to AI Robotics” by Robin
Murphy
- Slides by Lorenz Hillen from Universität
Bielefeld
SLIDE 64
ROS Services
SLIDE 65
Messages vs. Services
SLIDE 66
Messages vs. Services
SLIDE 67 Messages vs. Services
Publisher Subscriber Client Server t
i c r e q u e s t r e s p
s e
SLIDE 68 Messages vs. Services
Publisher Subscriber Client Server t
i c r e q u e s t r e s p
s e
t1 t2 t3
SLIDE 69
Calling Services in ROS
1) From the command line:
rosservice call <service_name> <request>
2) From code
SLIDE 70
THE END