Radhika Nagpal: What intelligent machines can learn from a school of fish
راديكل نابال: ماذا يمكن لآلات الذكاء الاصطناعي أن تتعلم من مدرسة الأسماك
Taking cues from bottom-up biological networks like those of social insects, Radhika Nagpal helped design an unprecedented “swarm” of ant-like robots. Full bio
Double-click the English transcript below to play the video.
off the coast of the Bahamas.
in the ocean before,
as I put my head in the water
هو عندما وضعت رأسي في الماء
to breathe through the snorkel,
أن أتنفس من خلال أنبوب التنفس
المخططة بالأصفر والأسود
of striped yellow and black fish
suddenly changed its mind,
and then swerved to the right
have had this experience.
and the beauty of it,
just the sheer oneness of it,
with a single collective mind
really ended up determining
أظن أن تلك التجربة انتهت بلعب دورا هاما
is artificial intelligence.
by creating our own computational systems
عبرخلقنا نظمنا الحاسوبية الخاصة
the way we see it in nature.
بالطريقة التي نراها بها في الطبيعة
come from science fiction and the movies,
تأتي غالبا من الخيال العلمي والأفلام
view of intelligence.
وجهة نظر الإنسان للذكاء
kind of intelligence.
of the collective,
and myopic view of what's going on,
isn't really about the individual
of the group itself.
that I still find most remarkable,
يراقبون مدرسة الأسماك هذه
supervising this fish school.
collective mind behavior
سلوك العقل الجماعي المذهل هذا
من تفاعلات الأسماك الواحدة مع الأخرى
of one fish and another.
توجد تلك التفاعلات أو قواعد الاختلاط
or rules of engagement
التي تؤدي لهذا النوع من الذكاء
that lead to this kind of intelligence,
أن نصنع الذكاء الخاص بنا؟
that I work on with my team in my lab.
الذي أعمل عليه مع فريقي بالمختبر
the mathematics behind it.
working closely with experimentalists.
حيث نعمل مع المختبرين
our own collective systems
التي نراها في الطبيعة
that we see in nature,
along this line
في هذا العمل
of a thousand robots.
to exhibit collective intelligence,
في حجم عملة ربع - الدولار -
about the size of a quarter,
مع الآلات الأخرى لاسلكيا
communicate with other robots,
exactly an interaction,
different kinds of rules of engagement
spontaneous synchronization,
and suddenly start all clapping together,
وفجأة يبدأ الكل بالتصفيق معا
for pattern formation,
they're going to take on
from nature's rules.
التي تعلمناها من الطبيعة
that we've learned from nature
entirely new collective behaviors
two different kinds of rules.
يمكنه الحركة حول آخر موضعي
around other stationary robots.
مبني على أساس الأثنين المجاورين له
based on its two nearest neighbors.
in a little pattern seed,
في
are sufficient for the group
في نمط خطي
a simple line pattern.
complicated pattern rules,
مجموعات ترتب نفسها ذاتياً
really complicated self assemblies,
a thousand robots
to self-assemble the letter K.
is that no one is in charge.
من الآليين بجواره
to a small number of robots nearby it,
ليتحرك حول البناء نصف المكتمل
to move around the half-built structure
بناءاً على قوانين النمط
based on its pattern rules.
is doing anything perfectly,
يقوم بمهمته على أكمل وجه
تصل لهدفها
the collective to do its goal
almost so perfect, you know --
تعرفون--
that they're individual robots at all,
in two dimensions,
والقوانين بأبعادٍ ثلاثة
and rules in three dimensions.
that could build together?
آليات يمكنهم البناء سوياً؟
from social insects.
mound-building termites
الباني للتلال
لمستعمراتهم من الطين
complex nest structures out of mud
also have pattern rules
من حشرات أخرى
out of other insects,
to create rules for robots.
لخلق قوانين للآليين
some simulated robots.
لبعض لآليين
through the structure,
عبر مجموعة من قطع البناء
where it looks at groups of blocks
and the right pattern rules,
وقوانين الأنماط الصحيحة
يبنون أي شئ نريده
to build whatever we want.
their own tower.
that go with these rules.
that can climb over blocks,
the very structure that it's on.
that you could imagine.
many different kinds of robot bodies.
that maybe could move sandbags
that built out of soft materials
to shore up a collapsed building --
in different kinds of bodies.
obsessed with army ants,
يمكنهم تسلق أي شئ
that can climb over literally anything
out of their own bodies.
of robot visions become possible.
about the rules that fish schools use.
التي تسخدمها مدارس الأسماك
the bodies to go with that,
التي تتماشى مع ذلك
with a fish school of our own creation.
مع أسماك من صنعنا
أدوات رياضية ومعنوية
the mathematical and the conceptual tools
of collective power,
of future applications,
من التطبيقات المستقبلية
that build flood barriers
لتلقيح المحاصيل
that could pollinate crops
تراقب حركة الشعب المرجانية
that monitor coral reefs,
and we thinking about programming
how to design the rules of engagement
good collective behavior
rules for insects and for fish
والأسماك
المنطبقة على نظامنا الإنساني؟
to our own human collective?
that I'd like to leave you with
of collective intelligence,
fish schools that I study,
التي أدرسها
أطول بكثير لنسلك
evolutionary path to walk.
the science of robot collectives,
علم الآليين الجماعيين
والتفكير في قوانين
and thinking about rules
our own scientific collective.
هو الذي يحدد ما يقوم به العلم
determines what science gets done.
الوقوف هنا
that they could stand here
بل التغيير
not just to understand but to change
حياتهم اليومية
impacts their everyday lives.
our own human collective,
سيكون شيئا رائعاً
it will be beautiful.
ABOUT THE SPEAKER
Radhika Nagpal - Robotics engineerTaking cues from bottom-up biological networks like those of social insects, Radhika Nagpal helped design an unprecedented “swarm” of ant-like robots.
Why you should listen
With a swarm of 1,024 robots inspired by the design of ant colonies, Radhika Nagpal and her colleagues at Harvard’s SSR research group have redefined expectations for self-organizing robotic systems. Guided by algorithms, Nagpal’s shockingly simple robots guide themselves into a variety of shapes -- an ability that, brought to scale, might lead to applications like disaster rescue, space exploration and beyond.
In addition to her work with biologically inspired robots, Nagpal helped create ROOT, a simple robot to teach coding to would-be programmers through a simple user interface suitable for students of all ages.
Radhika Nagpal | Speaker | TED.com