Vijay Kumar: The future of flying robots
Vijay Kumar - Roboticist
As the dean of the University of Pennsylvania's School of Engineering and Applied Science, Vijay Kumar studies the control and coordination of multi-robot formations. Full bio
autonomous aerial robots
that you can buy today,
to determine their position.
cameras and laser scanners,
relative to those features,
all these features into a map,
to understand where the obstacles are
we did inside our laboratory,
to go for longer distances.
what the robot sees with the camera.
by a factor of four --
the map that it's building.
of the corridor around our laboratory.
you'll see it enter our lab,
by the clutter that you see.
of building high-resolution maps
or outside the building
without actually going inside,
what happens inside the building.
with robots like this.
about 100 watts per pound.
a very short mission life.
that end up being very expensive --
and the processors.
can you buy in an electronics store
that has sensing onboard and computation?
smartphone that you can buy off the shelf,
can download from our app store.
reading the letters, "TED" in this case,
of the "T" and the "E"
to make sure if the robot goes crazy,
these small robots,
behaviors, like you see here.
at two to three meters per second,
as it changes direction.
smaller robots that can go faster
very unstructured environments.
gracefully coordinating its wings,
to grab prey out of the water,
hoagie that it's grabbing out of thin air.
going at about three meters per second,
coordinating its arms, its claws
to achieve this maneuver.
how the robot adapts its flight
than the width of the window.
and adjust the altitude
to make these even smaller,
in particular by honeybees.
and this is a slowed down video,
the inertia is so lightweight --
they bounce off my hand, for example.
that mimics the honeybee behavior.
you get lower inertia.
you're resistant to collisions.
we build small robots.
is only 25 grams in weight.
up to six meters per second.
ten times the speed of sound.
collision, at one-twentieth normal speed.
of two meters per second,
prevents the propellers from entangling,
and the robot responds to the collisions.
to these small robots.
of the number of Band-Aids we've ordered
to compensate for these disadvantages.
to form large groups, or swarms.
we try to create artificial robot swarms.
about networks of robots.
of sensing, communication, computation --
quite difficult to control and manage.
three organizing principles
to develop our algorithms.
need to be aware of their neighbors.
and communicate with their neighbors.
hijacked by a human operator, literally.
interact with each other,
able to lead this network of followers.
know where they're supposed to go.
to the positions of their neighbors.
the second organizing principle.
with the principle of anonymity.
to the identities of their neighbors.
you introduce into the formation,
reacting to its neighbor.
to form the circular shape,
without central coordination.
essentially give these robots
of the shape they need to execute.
as a function of time,
start from a circular formation,
stretch into a straight line,
kind of split-second coordination
that we are very interested in.
that we're facing worldwide.
in this earth is malnourished.
has already been cultivated.
in the world is improving,
efficiency is actually declining.
shortage, crop diseases, climate change
called Precision Farming in the community.
aerial robots through orchards,
precision models of individual plants.
to treat every patient individually,
models of individual plants
what kind of inputs every plant needs --
fertilizer and pesticide.
traveling through an apple orchard,
two of its companions
building a map of the orchard.
of every plant in this orchard.
that are being used on this robot.
a standard color camera.
is a thermal camera.
a three-dimensional reconstruction
as the sensors fly right past the trees.
we can do several things.
thing we can do is very simple:
how many fruits she has in every tree
the yield in the orchard,
to the amount of leaf area on every plant.
photosynthesis is possible in every plant,
how healthy each plant is.
and infrared information,
you can essentially see
not doing as well as other crops.
detecting the early onset of chlorosis --
by yellowing of leaves.
can easily spot this autonomously
that he or she has a problem
that can improve by about ten percent
the amount of inputs such as water
aerial robot swarms.
the people who actually create the future,
and Giuseppe Loianno,
demonstrations that you saw.
About the speaker:Vijay Kumar - Roboticist
As the dean of the University of Pennsylvania's School of Engineering and Applied Science, Vijay Kumar studies the control and coordination of multi-robot formations.
Why you should listen
At the General Robotics, Automation, Sensing and Perception (GRASP) Lab at the University of Pennsylvania, flying quadrotor robots move together in eerie formation, tightening themselves into perfect battalions, even filling in the gap when one of their own drops out. You might have seen viral videos of the quads zipping around the netting-draped GRASP Lab (they juggle! they fly through a hula hoop!). Vijay Kumar headed this lab from 1998-2004. He's now the dean of the School of Engineering and Applied Science at the University of Pennsylvania in Philadelphia, where he continues his work in robotics, blending computer science and mechanical engineering to create the next generation of robotic wonders.
Vijay Kumar | Speaker | TED.com