wwwuasvisioncom 20141210mistletoe quadcopter drawsblood ROS Example Tiered mobile robot architecture based on a temporal decomposition Strategiclevel decision making Controls the activation of behaviors based on commands from planner ID: 617499
Download Presentation The PPT/PDF document "TGI Friday’s Mistletoe Drone" 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
TGI Friday’s Mistletoe DroneSlide2
www.uasvision.com
/2014/12/10/mistletoe-
quadcopter-draws-blood/Slide3
ROSSlide4
Example
Tiered mobile robot architecture based on a temporal decomposition
Strategic-level decision making
Controls the activation of behaviors based on commands from planner
Low level behaviorsSlide5
Real World Environment
Perception
Localization
Cognition
Motion Control
Environment Model, Local Map
Position
Global Map
PathSlide6
Color Tracking Sensors
Motion estimation of ball and robot for soccer playing using color tracking
4.1.8Slide7Slide8
Robot Position:
ξ
I
=
[
x
I
,y
I
,θ
I]T
Mapping between frames ξ
R=R(θ)
ξI,
ξI=R(θ
)-1
ξ
R
R
(
θ
)=
Each wheel
contributes to speed:
rφ
/
2
For rotation, right wheel contributes:
ω
r
=
rφ
/2lSlide9
Model of the world
ExecuteSlide10
Occupancy Grid, accounting for C-Space
start
goalSlide11
Visibility GraphSlide12
Exact Cell Decomposition
How can we improve the path?
Plan over this graphSlide13
Matching
®
®
®
Global Map
Local Map
…
…
…
obstacle
Where am I on the global map?
Examine different possible robot positions.Slide14
General approach:
A: action
S: pose
O: observation
Position at time t depends on position previous position and action, and current observationSlide15
Localization
Sense
Move
Initial Belief
Gain Information
Lose
InformationSlide16
Uniform Prior
Observation: see pillar
Action: move right
Observation: see pillarSlide17
Example 2
0.2
0.2
0.2
0.2
0.2
Robot senses yellow.
Probability should go up.
Probability should go down.Slide18
Kalman
Filter
Sense
Move
Initial Belief
Gaussian:
μ, σ
2
μ’=
μ+u
σ
2
’=σ
2
+r
2Slide19
Particle FilterSlide20
EKF-SLAM
Grey: Robot Pose Estimate
White: Landmark Location EstimateSlide21
“Using robots to get more food from raw
materials” = Chicken Chopping Robot!
http://www.sciencedaily.com/releases/2014/12/141209081648.
htmSlide22
Gradient Descent
Minimum is
found by following the slope of the function
“Like climbing Everest in thick fog with amnesia”Slide23
Genetic algorithmsSlide24
Brainstorm
How could a human teach a robot?
What are the goals (why would this be useful)?
What would the human do?
How would the robot use this data?
Example applications
Flip a pancake
Play ping pong
Help build Ikea furnitureSlide25
HRI
Acceptance
Dehumanize
Institutionalize
Enabling
Chair
Horse
Autonomy
Passive
Consent
PrivacyCamerasSecuritySlide26
Robots: an ME’s perspectiveSlide27
Robots for
GerontechnologySlide28
DARPA robotics ChallengeSlide29
Learning & VisionSlide30Slide31
Multi-Robot (Multi-Agent) Systems
Homogenous / Heterogeneous
Communicating / Non-CommunicatingCooperative / CompetitiveSlide32
Going Forward
Robots are becoming:
More commonMore powerful
Less expensive