SLIDE 1
COMP 150: Probabilistic Robotics for Human-Robot Interaction - - PowerPoint PPT Presentation
COMP 150: Probabilistic Robotics for Human-Robot Interaction - - PowerPoint PPT Presentation
COMP 150: Probabilistic Robotics for Human-Robot Interaction Instructor: Jivko Sinapov www.cs.tufts.edu/~jsinapov This week: Planning Announcements Reading Assignment Project Related Article of your choice If working with a team, each
SLIDE 2
SLIDE 3
Announcements
SLIDE 4
Reading Assignment
- Project Related Article of your choice
– If working with a team, each member should read a
different article
– Summary and link on Canvas – Due 3/24
SLIDE 5
Proposal Peer Review
- Each student will review 2 proposals, each
proposal will receive ~3 reviews
- Look for an email from Tyler with instructions
SLIDE 6
Planning
SLIDE 7
Path Planning using A*
[from “Making Shakey 1966-1972”]
SLIDE 8
Planning for Manipulation
[http://arm.eecs.umich.edu/]
SLIDE 9
Whole-Body Motion Planning
SLIDE 10
Whole-Body Motion Planning
SLIDE 11
Planning with Symbols
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
The need for (semi-) autonomy
SLIDE 22
How should autonomy be achieved and organized?
SLIDE 23
Robot Primitives
SLIDE 24
The Early Answer (1967): Sense-Plan-Act
SLIDE 25
The Early Answer (1967): Sense-Plan-Act
SLIDE 26
Early Example of S-P-A
SLIDE 27
Shakey Video
SLIDE 28
Early Work on Planning
SLIDE 29
Early Work on Planning
SLIDE 30
Early Work on Planning
SLIDE 31
A More Realistic Example
SLIDE 32
A More Realistic Example A More Realistic Example
SLIDE 33
A More Realistic Example
Is INROOM(IT,R1) true or false? CONNECTS(D1,R1,R2)? INROOM(IT,R2)? IT
SLIDE 34
Representing Initial State
IT
SLIDE 35
Representing Goal State
IT
SLIDE 36
The “difference” table
SLIDE 37
Logical Difference
- r
SLIDE 38
Finding the Plan
SLIDE 39
Discussion
- How did you solve the problem?
- What are some limitations of planning with
STRIPS?
- Where do the predicates, operators, etc. come
from?
SLIDE 40
Towers of Hanoi with PDDL
[ https://s3.amazonaws.com/ka-cs-algorithms/hanoi-5-init.png ]
SLIDE 41
3-Disk Hanoi
SLIDE 42
Final Plan
SLIDE 43
PDDL
- Editor: http://editor.planning.domains/
- Tutorial:
https://www.cs.toronto.edu/~sheila/2542/s14/A1/ introtopddl2.pdf
- Example PDDL files:
http://www.ida.liu.se/%7ETDDC17/info/labs/plann ing/strips/
SLIDE 44
Actions
- Action name and parameters:
- Preconditions:
- Effects:
SLIDE 45
Planning Exercise
- Consider a service robot operating in a human
environment such as an office or our department
- Specify three high-level actions, with
preconditions and end-effects
– You will need to specify the relevant predicates as
well
– Specify a planning problem within the domain
SLIDE 46
Further Reading
- Planning with STRIPS: A gentle introduction
http://www.primaryobjects.com/2015/11/06/artifici al-intelligence-planning-with-strips-a-gentle-int roduction/
- Cashmore, Michael, et al. "ROSplan: Planning
in the robot operating system." Twenty-Fifth International Conference on Automated Planning and Scheduling. 2015.
SLIDE 47
Next time...planning in stochastic domains
SLIDE 48