Return to Video

A friendly, autonomous robot that delivers your food

  • 0:01 - 0:03
    Food delivery.
  • 0:03 - 0:06
    It's the thing that saves
    millennials from starvation.
  • 0:07 - 0:08
    By my calculations,
  • 0:08 - 0:12
    Americans order over 20 million
    restaurant deliveries
  • 0:12 - 0:14
    every single day.
  • 0:14 - 0:17
    Over half of these deliveries
    are actually within walking distance.
  • 0:17 - 0:21
    But nine out of 10 are delivered in cars.
  • 0:22 - 0:24
    So basically,
  • 0:24 - 0:26
    we are moving a two-pound burrito
  • 0:26 - 0:28
    in a two-ton car
  • 0:28 - 0:30
    20 million times a day.
  • 0:30 - 0:32
    The energy to bring you that burrito
  • 0:32 - 0:35
    is actually bringing you
    a two-ton metal cage
  • 0:35 - 0:36
    with heated seats.
  • 0:36 - 0:38
    Let's be honest.
  • 0:38 - 0:40
    We are addicted to our cars.
  • 0:40 - 0:42
    Did you know that in America,
  • 0:42 - 0:45
    for every car,
    there are four parking spots?
  • 0:45 - 0:47
    In some downtowns,
  • 0:47 - 0:50
    over half of the real estate is for cars.
  • 0:50 - 0:53
    We have designed our cities
    around our cars,
  • 0:53 - 0:56
    because we drive
    whether we're going two miles
  • 0:56 - 0:58
    or 200 miles.
  • 0:58 - 1:00
    Solo, or with our whole family.
  • 1:01 - 1:06
    We get into the same SUV
    to go buy coffee or a coffee table.
  • 1:06 - 1:09
    If we could free up
    some of these streets and parking lots,
  • 1:09 - 1:10
    we could build more housing,
  • 1:10 - 1:12
    more social spaces, more parks.
  • 1:13 - 1:14
    But to do that,
  • 1:14 - 1:18
    first, we need to rethink
    how we are using cars today.
  • 1:18 - 1:19
    In the city of the future,
  • 1:19 - 1:23
    if you want to go five blocks,
    you summon a bike or a scooter.
  • 1:23 - 1:26
    If you're in a rush,
    a passenger drone would pick you up.
  • 1:26 - 1:30
    And if you need food,
    no need to have someone drive over --
  • 1:30 - 1:32
    the food will make its way to you.
  • 1:32 - 1:36
    Let's go back to those 20 million a day
    restaurant deliveries.
  • 1:36 - 1:39
    If we could get
    these deliveries off the road,
  • 1:39 - 1:41
    we could reduce the need
  • 1:41 - 1:45
    for as many as one and a half
    million cars just in the US.
  • 1:45 - 1:48
    That's twice the size of San Francisco.
  • 1:48 - 1:52
    Now, think of the impact this could have
    on cities like Delhi,
  • 1:52 - 1:55
    or my birth city of Tehran,
  • 1:55 - 1:58
    where car pollution is killing
    thousands of people every year.
  • 1:59 - 2:03
    So how do we get some
    of these deliveries off the road?
  • 2:05 - 2:06
    Well, that's the question
  • 2:06 - 2:10
    that my team and I have been obsessed with
    over the last three years.
  • 2:10 - 2:13
    And the solution is
    actually one of the building blocks
  • 2:13 - 2:15
    of the city of the future.
  • 2:16 - 2:20
    We've been creating
    small, self-driving robots
  • 2:20 - 2:23
    that navigate quiet alleys and sidewalks
  • 2:23 - 2:25
    on a walking pace
  • 2:25 - 2:28
    and have a secured cargo
    to deliver you food and supplies.
  • 2:28 - 2:31
    Now, before I tell you
    more about the robots,
  • 2:31 - 2:33
    let's do a quick thought experiment.
  • 2:34 - 2:38
    In your mind, picture a city
    with thousands of robots.
  • 2:39 - 2:40
    Is it this one?
  • 2:41 - 2:45
    This Hollywood dystopia
    is what a lot of people expect.
  • 2:45 - 2:50
    But our job is to create a friendly future
    that's designed for people.
  • 2:50 - 2:53
    So instead of making aliens,
  • 2:53 - 2:57
    we set out to create robots
    that are familiar.
  • 2:57 - 3:00
    Robots that would belong
    in our communities.
  • 3:01 - 3:03
    But we also wanted a little surprise.
  • 3:03 - 3:06
    Something unexpectedly delightful.
  • 3:06 - 3:07
    Think about it.
  • 3:07 - 3:08
    You're walking down the street,
  • 3:09 - 3:11
    and you see your very first robot.
  • 3:11 - 3:13
    That's the moment
    when you're going to decide
  • 3:13 - 3:16
    if this is a future you love or fear.
  • 3:17 - 3:20
    And with a lot of people
    having these dystopian ideas,
  • 3:20 - 3:22
    we need to open their minds.
  • 3:22 - 3:25
    We want to surprise and delight them,
  • 3:25 - 3:28
    so that we can win them over
    on first impression.
  • 3:28 - 3:29
    This is what we came up with.
  • 3:29 - 3:32
    It's familiar, but it's also surprising.
  • 3:33 - 3:34
    It's just a shopping cart,
  • 3:35 - 3:38
    but it also looks like
    we crossbred WALL-E with Minions.
  • 3:39 - 3:42
    If you live in San Francisco
    or Los Angeles,
  • 3:42 - 3:45
    chances are one of these
    has already delivered your food.
  • 3:45 - 3:48
    As soon as we put robots
    out on the street,
  • 3:48 - 3:50
    we learned some
    really interesting problems.
  • 3:50 - 3:53
    Like, how should robots cross the road?
  • 3:54 - 3:59
    Or how should robots interact with people
    with visual or mobility impairments?
  • 4:00 - 4:04
    We quickly realized
    that we need to teach our robots
  • 4:04 - 4:06
    how to communicate with people.
  • 4:07 - 4:10
    People on the sidewalk
    come from every walk of life,
  • 4:10 - 4:13
    so we needed to create a new language,
  • 4:13 - 4:15
    kind of a universal language
  • 4:15 - 4:18
    so people and robots
    can understand each other
  • 4:18 - 4:19
    right off the bat.
  • 4:19 - 4:22
    Because no one is going
    to be reading user manuals.
  • 4:24 - 4:27
    We started with eyes,
    because eyes are universal.
  • 4:27 - 4:29
    They can show where the robot is going
  • 4:29 - 4:32
    or if it's confused.
  • 4:32 - 4:34
    Plus, eyes make robots more human.
  • 4:35 - 4:37
    We also used sounds.
  • 4:37 - 4:40
    For example, we created this running sound
  • 4:40 - 4:41
    with frequent gaps
  • 4:41 - 4:45
    so that people with visual impairments
    could locate their robots
  • 4:45 - 4:47
    using the Doppler effect.
  • 4:47 - 4:50
    But it turned out these were not enough.
  • 4:50 - 4:51
    At intersections,
  • 4:51 - 4:53
    cars would cut in front of our robots.
  • 4:53 - 4:56
    Drivers were getting confused sometimes,
  • 4:56 - 5:00
    because robots would take too long
    before they started crossing.
  • 5:01 - 5:03
    Even ordinary pedestrians
    were getting confused.
  • 5:04 - 5:07
    Sometimes, they couldn't figure out
    on which side to pass the robots,
  • 5:07 - 5:11
    because robots make a lot
    of small adjustments to their direction
  • 5:11 - 5:12
    as they move.
  • 5:12 - 5:14
    This actually sparked a new idea.
  • 5:15 - 5:19
    What if we used movement
    to create a universal language?
  • 5:19 - 5:21
    Like, at intersections,
  • 5:21 - 5:24
    robots would gently move forward
    before they start crossing,
  • 5:24 - 5:27
    to signal to drivers that it's their turn.
  • 5:28 - 5:30
    If they see someone in a wheelchair,
  • 5:30 - 5:34
    they yield by pointing themselves
    away from the sidewalk,
  • 5:34 - 5:36
    to signal that they're not going to move.
  • 5:38 - 5:39
    Some of you may remember this.
  • 5:39 - 5:45
    In 2015, Canadian researchers
    sent a robot hitchhiking across the US.
  • 5:46 - 5:47
    It didn't get very far.
  • 5:47 - 5:51
    It turns out that robots
    can also use some social skills.
  • 5:51 - 5:53
    Like, if they're being tampered with,
  • 5:54 - 5:59
    Carnegie Mellon researchers have shown
    that small toy robots should play dead,
  • 5:59 - 6:02
    because people feel bad
    when they think they broke it.
  • 6:02 - 6:04
    But delivery robots aren't toys,
  • 6:05 - 6:07
    they're not small,
    they are out there in public.
  • 6:07 - 6:10
    We found that with delivery robots,
  • 6:10 - 6:12
    to get people to stop tampering,
  • 6:12 - 6:14
    robots need to show awareness.
  • 6:15 - 6:17
    It's kind of the opposite of playing dead.
  • 6:18 - 6:19
    In this case,
  • 6:19 - 6:22
    robots need to acknowledge the situation
  • 6:22 - 6:24
    to get people to step away.
  • 6:25 - 6:26
    Also, a word of advice.
  • 6:26 - 6:29
    If you are a robot and you see small kids,
  • 6:29 - 6:31
    run towards the closest adult.
  • 6:31 - 6:35
    It turns out that some kids
    just love harassing robots.
  • 6:35 - 6:38
    So besides dystopia,
  • 6:38 - 6:41
    Hollywood also promised us
    some really cool robots
  • 6:41 - 6:44
    that would run our errands
    or keep us company.
  • 6:44 - 6:47
    So far, we've really focused
    on food delivery,
  • 6:47 - 6:49
    but in the future,
  • 6:49 - 6:51
    these robots can do more.
  • 6:51 - 6:55
    Like, they could gather excess food
    and bring it to shelters every night.
  • 6:56 - 7:00
    Because in America,
    we waste 30 percent of our food,
  • 7:00 - 7:03
    while 10 percent of our people
    experience food insecurity.
  • 7:03 - 7:05
    These robots could be
    part of the solution.
  • 7:06 - 7:10
    Or when we have hundreds of robots
    running around cities,
  • 7:10 - 7:14
    we could have robots carry
    emergency medications at all times,
  • 7:14 - 7:18
    just in case someone nearby
    has an allergic reaction
  • 7:18 - 7:19
    or an asthma attack.
  • 7:19 - 7:22
    These robots could be on-site
    within a minute or two,
  • 7:22 - 7:24
    faster than anyone else.
  • 7:24 - 7:25
    And during pandemics,
  • 7:26 - 7:28
    robots can be a key part
    of our infrastructure.
  • 7:28 - 7:30
    They can ensure
  • 7:30 - 7:33
    that we can provide our communities
    with the essential needs
  • 7:33 - 7:35
    even during emergencies.
  • 7:35 - 7:38
    Let me leave you with one last thought.
  • 7:38 - 7:44
    Today, objects can't get from A to B
    without human help,
  • 7:44 - 7:47
    because our three-dimensional world
    is quite complex.
  • 7:48 - 7:52
    But new sensors and AI can change that.
  • 7:52 - 7:55
    In a way, technology is like a baby
  • 7:55 - 7:59
    that has just learned to recognize objects
    and understand words,
  • 7:59 - 8:02
    and maybe even hold a basic conversation,
  • 8:02 - 8:04
    but it hasn't learned to walk yet.
  • 8:05 - 8:07
    Now, we are teaching technology
  • 8:08 - 8:10
    how to navigate
    the three-dimensional world
  • 8:10 - 8:11
    without our help.
  • 8:13 - 8:15
    We are entering this new era
  • 8:15 - 8:19
    where insentient objects
    are going to get up and move freely.
  • 8:20 - 8:22
    And when they do,
  • 8:22 - 8:24
    we've got to make sure
    they don't look like aliens.
  • 8:24 - 8:29
    My vision for the future
    is that when things come to life,
  • 8:29 - 8:31
    they do so with joy.
  • 8:31 - 8:34
    You know, less like
    the movie "Terminator"
  • 8:34 - 8:36
    and more like "Toy Story."
  • 8:36 - 8:37
    Thank you.
Title:
A friendly, autonomous robot that delivers your food
Speaker:
Ali Kashani
Description:

Meet the friendly robot that could deliver your next burrito. Ali Kashani introduces us to Postmates' autonomous delivery robot and explains how it could help reduce carbon emissions and free up valuable real estate in cities everywhere. Learn more about how it was specially designed to navigate complex social interactions on busy sidewalks to bring you your food (and more) with joy.

more » « less
Video Language:
English
Team:
closed TED
Project:
TEDTalks
Duration:
08:51

English subtitles

Revisions Compare revisions