Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
Dr. José Ignacio Núñez Varela [email protected]
MICCS 2015
Probabilistic Robotics and
Models of Gaze Control
Part I:
Probabilistic Robotics
Imagen: http://fullhdwp.com/images/wallpapers/terminator-Wallpaper1.jpg
Imagen: http://www.sideshowcollectors.com/forums/attachment.php?attachmentid=98536&d=1394674038
Why do we need robots?
Imagen: http://upload.wikimedia.org/wikipedia/commons/5/5e/KUKA_Industrial_Robots_IR.jpg
Imagen: http://inapcache.boston.com/universal/site_graphics/blogs/bigpicture/robot_08_12/r15_19549703.jpg
Imagen: http://upload.wikimedia.org/wikipedia/commons/f/fa/Martian_rover_Curiosity_using_ChemCam_Msl20111115_PIA14760_MSL_PIcture-3-br2.jpg
Imagen: http://www.popsci.com/sites/popsci.com/files/styles/large_1x_/public/import/2013/images/2010/03/autosub_6000.jpg?itok=m44HsKug
Imagen: http://www.guidedsolutions.co.uk/news/de/wp-content/uploads/2014/06/Robotic-surgery1.jpg
Imagen: http://upload.wikimedia.org/wikipedia/commons/6/64/US_Navy_090512-N-2013O-013_A_Mark_II_Talon_robot.jpg
Imagen: https://www.nasa.gov/images/content/664141main_iss031e148737_full.jpg
But the kind of robot we
really want is …
Imagen: http://img4.wikia.nocookie.net/__cb20100924163215/bigbangtheory/images/e/ed/The-Big-Bang-Theory-The-Robotic-
Manipulation.jpg
No, not really.
What about this …
Imagen: http://www.jsk.t.u-tokyo.ac.jp/research/irt/images/ar-robot.jpg
Imagen: http://inapcache.boston.com/universal/site_graphics/blogs/bigpicture/robot_08_12/r12_19311581.jpg
Imagen: http://inapcache.boston.com/universal/site_graphics/blogs/bigpicture/robot_08_12/r12_19311581.jpg
Unfortunately, there are
still many things to
solve first …
Basic model of
robot interaction
Picture credits: http://www.3ders.org/images/PrimeSense_apple-3d-sensor-1.jpg
http://cdn.shopify.com/s/files/1/0130/8982/products/midi-cpu-large_1024x1024.jpg
Sensing Planning
Acting
Asimo © Honda
We need intelligent
robots!
Intelligent robots
Learning
Reasoning
Decision-making
Planning
Understanding
Common sense
PR2 © Willow Garage
Robots have to be able to accomodate the
enormous uncertainty that exists in the physical
world.
Imagen: http://www.grumpygratefulmom.com/wp-content/uploads/2011/11/messy-kitchen-1024x769.jpg
But, what is
uncertainty?
Being not certain about
something.
Imagen: http://www.allonrobots.com/images/robot-question.jpg
The estimated percentage by which
an estimated or calculated value
may differ from the true value.
Imagen: http://www.threadbombing.com/data/media/30/motorcycle_jump_failure.jpg
Let’s see an
example
Robot
grasping
The robot might not have a good
estimate of where the object is
What factors contribute to
the robot's uncertainty?
Robot Environments
Well structured environment
<< uncertainty Not structured environment
>>> uncertainty
Imagen: http://upload.wikimedia.org/wikipedia/commons/5/5e/KUKA_Industrial_Robots_IR.jpg
http://www.archimuse.com/mw2001/papers/giannoulis/giannoulis_fig1.jpg
Sensors are limited in what they can perceive (e.g.,
physical limitations affect range and resolution)
Sensors are subject to noise
Sensors can break
Imagen: http://www.longrangecamera.com/fov.gif
Robot Sensors
Motors are, at some extent, unpredictable
Control noise, wear-and-tear, mechanical failure
Imagen: https://content.solarbotics.com/products/photos/8ddc7ea32073c4756a4cbdaedcbda0fa/lrg/IMG_0681.jpg
Robot Actuators
All internal models of the world are approximate
Model errors have often being ignored
Imagen: http://latimesblogs.latimes.com/.a/6a00d8341c630a53ef014e610f9f0e970c-800wi
Robot’s Internal Models
Robots are real time systems, thus limiting the
amount of computation being carried out
Algorithms need to be approximated
Imagen: http://www.pirobot.org/blog/0015/map-1b.png
Algorithmic Approximations
Robots are forced to act even
though they don't have sufficient
information to make decisions with
absolute certainty.
Imagen: http://i.dailymail.co.uk/i/pix/2014/03/11/video-undefined-1C33F90E00000578-54_637x365.jpg
“Managing uncertainty is
possibly the most important
step towards robust real-world
robot systems.”
- Thrun, Burgard and Fox
Probabilistic Robotics
Key idea: Represent uncertainty explicitly
using the calculus of probability theory.
Instead of relying on a single “best guess”,
probabilistic algorithms represent information
by probability distributions over a whole space
of guesses.
Mobile Robot Localization
Imagen: http://clearpath.wpengine.netdna-cdn.com/wp-content/uploads/2013/01/Gallery_TurtleBot_Office.jpg
Mobile Robot Localization
The map is given
The robot wants to know where it is
Mobile Robot Localization
The robot assumes a uniform probability distribution
of where it is
(it is likely to be in any place in the map)
Mobile Robot Localization
The robot’s belief increases after sensing a door
(data is integrated into the old belief)
Mobile Robot Localization
The robot moves some distance
(Its belief moves as well, but the movement
introduces some noise)
Mobile Robot Localization
The robot senses a door once again, and its belief of
where it is increases
Bayes Theorem
Imagen: http://upload.wikimedia.org/wikipedia/commons/d/d4/Thomas_Bayes.gif
Bayes Theorem
Prior probability
Bayes Theorem
Data
Bayes Theorem
Posterior
probability
Part II:
Gaze Control
Imagen: http://www.grumpygratefulmom.com/wp-content/uploads/2011/11/messy-kitchen-1024x769.jpg
Biological
perspective
Gaze Control
Machine
perspective © Jason Babcock © icub.org
Why study gaze control?
© cellfield.ca
Foveal Vision
© Michael Land
Eye Movements
Saccades
• Rapid jump-like movements (900°/sec)
• Ballistic (trajectory cannot change)
• Stereotyped (follow the same pattern)
• Voluntary and involuntary
• Aim: Shift the fovea to obtain high resolution samples
Saccade Sequence
We perform hundreds
or even thousands of
saccades every day!
How does the brain
decide where to
fixate next?
© Ilya Repin
Active Vision
© Yarbus
Task and context
determine where to
fixate next
Vision and Action
© Mary Hayhoe
Uncertainty Reduction
Engineering science goal
What mechanisms a rational decision maker
could employ to select a gaze location
optimally, or near optimally, given limited
information and limited computation time
during the performance of a task?
Human behavioural goal
How humans select the next gaze location?
Gaze Control Processes
iCub Humanoid Robot
© icub.org
Two problems
where to look gaze allocation
Pick & Place Task
Models of Gaze Control
• Based on uncertainty reduction (Uncertainty)
• Based on rewards and uncertainty (Rew+Unc)
• Based on rewards, uncertainty and gain (Rew+Unc+Gain)
“What would happen if I look at entity ei?”
One-step look ahead gaze control
Uncertainty Reduction
“How much uncertainty is
reduced if I look at entity ei?”
X
Reward and Uncertainty
“How much value am I
expected to get after looking
at entity ei?”
X
Reward, Uncertainty & Gain
“Which motor system would
get more benefit if gaze is
allocated to it?”
X
Conclusions
•We need robots!
•There is still much to do before we can
buy our assistant robot
•You can contribute to make this
happen!
Thank You!!
E-mail: [email protected]
Website: http://ciep.ing.uaslp.mx/jnunez
© Botodesigns / Chen Reichert
Reach/Grasp Sensitivity
Observation Noise
Field of View