Also cookies Also also want to work on an appbusiness idea This week Normal lab hours Weekendnext week Will email 125 No class but lab will be open 128 Takehome due 1213 Projects due ID: 482913
Download Presentation The PPT/PDF document "` http://www.uasvision.com/2013/12/03/wa..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
`
http://www.uasvision.com/2013/12/03/waterstones-trumps-amazon-with-owl-book-delivery-service/
Also, cookies
Also, also, want to work on an app/business idea?Slide2
This week: Normal lab hours.
Weekend/next week: Will email
12/5: No class, but lab will be open
12/8: Take-home due
12/13: Projects due
In Class:
Presentation (20%)
Due at 8am on Angel
Write-up (60%)
Video (20%)
Code (Ungraded, but project not considered submitted without)
https://skylight.wsu.edu/student/Slide3
Multi-Robot (Multi-Agent) Systems
Homogenous / Heterogeneous
Communicating / Non-Communicating
Cooperative / CompetitiveSlide4
Multi-Robot Systems
Example 1: Target Capture
Drone finds a target
Turtlebot needs to go to the target
Why is this much harder than either of the individual tasks?
What are the risks / challenges?Slide5
Risks / Challenges?
Drone’s hovering? Landing? FIRE?!?
Trust human vs. other robot
Obstacles the turtlebot can’t get around? Path planning is very different for two robots
Shared map?
Different requirements for different vehicles
Lots of data (point cloud)Slide6
Multi-Robot Systems
Example 1: Target Capture
Drone finds a target
Turtlebot needs to go to the target
Why is this much harder than either of the individual tasks?
What are the risks / challenges?
What if it’s a drone and a
human
? How do things change?
What if it’s a
human
and a turtlebot?Slide7
Example 2: Ad-hoc networking
Robots can communicate
5 mobile ad-hoc network stations
Want to maximize network strength, but can only take
small scale
movements
How could we optimize this
?
Centralized / Decentralized
Model based / free
Backtracking? Slide8
Example 3: Robot Patrol
1 robot
2 robots
n
robotsSlide9
Example 3: Robot Patrol
1 robot
2 robots
n
robotsSlide10
Example 3: Robot Patrol
1 robot
2 robots
n
robots
Different value targets?
Different robot capabilities
?
What if could be observed?Slide11
Example 3: Robot Patrol
1 robot
2 robots
n
robots
Different value targets?
Different robot capabilities?Slide12
Example 4: Hide and seek
“Hider” is stationary
Solution techniques?Slide13
Example 4: Hide and seek
“Hider” is stationary
Solution techniques?
What if “hider” is
learning
?Slide14
Example 4: Hide and seek
“Hider” is stationary
Solution techniques?
What if “hider” is
learning
?Slide15
Example 5: Predator / Prey,
without
communication
1 fixed policy prey
4 learning predators
Need to surround the preySlide16
Example 5: Predator / Prey,
with
communication
1 fixed policy prey
4 learning predators
Need to surround the preySlide17
Example 5: Predator / Prey,
with
communication
1 fixed policy prey
4 learning predators
Need to surround the prey