로딩 중...
영어학습소
영어학습소
홈
테디잉글리시
수능
Shadowing
재생 속도
0.5x
0.75x
1x
1.25x
1.5x
시작 지점을 클릭하세요
0:00
I'm Henry Admoni.
0:04
I'm a professor of Robotics.
0:04
Today I'll be answering your questions from Twitter.
0:04
This is robot support.
0:04
Downlow at AG26243 asks, "Why are people so obsessed with making human-like robots?"
0:19
"We all need to chill."
0:21
I don't disagree.
0:23
Human-like robots are great.
0:23
Usually, a robot that's not shaped like a human is more robust and more capable of doing the task it needs to do.
0:30
One example of a humanoid robot is Sophia the robot.
0:32
Some people would call this category of robots Androids.
0:37
Sophia is a highly realistic robot, but if you've ever watched Sophia, you know that there's still a gap between what the robot is doing and what we actually expect from people.
0:47
This is often called the uncanny valley.
0:49
Very human-like robots can fall into that uncanny valley and make people pretty uncomfortable.
0:53
At Furby Factor asks, "How do you program personality into a robot?"
0:58
It is hard to define exactly what makes a personality.
1:02
Robots are just programs.
1:04
For example, if somebody makes a funny joke, should the robot laugh or should the robot roll his eyes?
1:09
This all has to be created in software.
1:11
When we're programming robots, what we're doing is building a set of instructions.
1:17
If you run into an obstacle, back up.
1:26
To get more complex behavior, you can't rely just on pre-specified rules.
1:26
That's where something like machine learning comes in.
1:26
In machine learning, we are building algorithms, which are basically just sets of instructions that can, based on the information that they're getting, adapt themselves to learn the right kinds of output.
1:26
At Cristini asks, "Can you pick up a single M&M with a robot arm?"
1:46
It kind of depends on the robot arm.
1:49
With this one, it's going to be a little hard.
1:52
Some robots are only robot arms, and those are called manipulators, like this one right here.
1:59
Ah, so close!
1:59
This kind of fine motor manipulation is really challenging for robots.
2:03
Robots don't have the level of motion and movement that we do in our fingers.
2:09
The other thing that we use in our fingers that robots don't have is a sense of touch.
2:14
That allows us to tell when we are actually gripping something so that we can know that it's able to be picked up.
2:21
This box has fiducial markers on it.
2:24
These are essentially QR codes that allow the robot camera to recognize what orientation and position the box is in.
2:38
It's supposed to make it easier for the robot to pick up the box autonomously.
2:38
At SEO Chase asks, "Do you think robots will one day take over all of our jobs?"
2:46
The real benefit of robots is taking over the three Ds: the dull, dirty, and dangerous jobs that we probably don't want human beings to be doing anyway.
2:53
People are working on underwater robots that can detect underwater landmines.
2:58
Some people have worked on robots that can go into nuclear facilities after an accident and shut off different valves.
3:06
But I do hope that robots are able to make people better at their jobs and free people up to do things that they're actually good at and they actually want to do.
3:15
At Big Tech Prof asks, "What is Elon's goal for the Tesla Optimus robot?"
3:20
Elon Musk does not consult with me, but my understanding is that he's looking to build a general-purpose home robot.
3:31
One that can do everything from cleaning up your room to unloading your dishwasher.
3:31
To design a robot that's capable of doing more than just one thing takes a ton of work and a ton of investigation.
3:31
Tesla is pretty well positioned to do so.
3:31
Their autonomous driving division has a huge computer vision capability, and they're building the kinds of batteries that would be needed to power this sophisticated machine for a long period of time.
3:31
At Display Your CV says, "What are the major problems in the field of Robotics today?"
3:31
There are a few big challenges that we face today when we're trying to get robots out into the real world.
4:00
The first one is perception: how do we get robots to see and hear and interact with their environment and actually understand what it is that they're looking at?
4:10
Another one is actuation: how do we get robots to move around in ways that allow them to achieve their tasks?
4:15
And finally, another big problem is intelligence: how do we make our robots smart enough to plan the right kind of action to do in order to achieve what they're trying to?
4:25
One major area of interest to me is how we get robots to interact with people, like personal home robots or healthcare robots that are helping nurses and doctors.
4:36
At Alice J and Taylor asks, "I just want to know how Boston Robotics plans to deal with humans who think pushing robots over is hilarious."
4:43
There are lots of people who are doing research on how robots should respond to various forms of abuse.
4:49
One that I like the most is that when a robot is being kicked by kids in a shopping mall, researchers found that if it just rolls to the nearest adult, the kids will stop.
5:00
At a Leveres asks, "Have you ever had a genuine social interaction with a robot?"
5:06
Although I can't say that I've had a deep social interaction with a robot, I can say that I have felt social reactions to robots.
5:14
For example, this is Curry.
5:14
Curry is really cute.
5:18
If you pet Curry's head, it'll purr, and that kind of response actually does generate some social interaction for me.
5:30
A lot of people are working on social robots to help people in the home or in elder care situations, and in these cases, it's really important that the robot is able to connect socially with people.
5:40
For example, it should tell when somebody's frustrated or not wanting to take a bite of food, or somebody's tired and should just go to bed.
5:50
Detecting these social cues is still really challenging to build into an algorithm, and that's a huge part of social robotics.
6:02
At Tomas Cosmiter asks, "How #AI is being used in #robotics?"
5:59
AI and robotics have a lot of overlap.
6:05
AI is all about how you get systems to act intelligently.
6:11
Robotics is about taking devices and putting them out in the real world and having them move around and interact with the environment or with people.
6:19
So in order for robots to work, we really need the AI to support them, but robotics is more than just the AI that supports them.
6:28
It's also about the mechanics of building a physical system and the electronics of developing safe circuitry.
6:40
Robotics and AI: big overlap, but still two separate fields.
6:40
At Joe Maroney asks, "Self-driving cars are already on the road.
6:45
When there's a problem, how do we understand the decisions AI makes?"
6:48
Self-driving cars are essentially a type of robot moving around in the world and making intelligent decisions based on what they're perceiving.
6:56
These algorithms that robots are using for decision-making are sort of like black boxes.
7:01
Inputs come in in the form of perceptions or knowledge, and then outputs come out in the form of actuation.
7:01
The robot decides to merge or decides to stop, but that black box can be very, very complicated.
7:01
And because of that, we need something like explainable AI in order to back out the processes that the robot used to get to where it was going.
7:01
EE World Online asks, "What types of sensors are used in robots?"
7:26
Some robots have cameras like this one that can detect not just color images, but also depths using infrared.
7:34
Other robots have physical sensors like bump sensors or capacitive sensors that respond to touch.
7:55
Time of flight sensors or other kinds of ways...
7:55
Excuse me, my robot is escaping.
7:55
The original Roombas also had bump sensors.
7:55
That's how they knew they were running into something.
7:55
Lots of robots also use audio sensors.
7:55
Hey Curry.
7:55
Or sounds like that.
7:55
At 24/7 Press Release asks, "What are swarm robots, and how are they learning from insects?"
7:55
If you watch the Olympics, you've seen these beautiful displays of lights floating in space.
8:12
Those are swarm robots.
8:12
Swarm robots are groups of robots that operate in a collective, have intelligence that's distributed across the entire system, so that no one of them has the central information.
8:23
These are designed and inspired by the way insect swarms operate, and it's just one example of the ways that robotics has been inspired by biology.
8:40
At Robottrix One: "How six-axis industrial robots work."
8:40
Six-axis robots are called that because they can move in six different directions.
8:42
Here's an example of a six-axis robot.
8:47
It can move in three directions of physical space: left and right, up and down, and forward and back.
8:55
They can also move in three directions of orientation space: pitch, yaw, and roll.
9:03
These robots work by deciding how to move every joint in the kinematic chain in order to get the robot hand, which we call an end effector, to the right position.
9:03
This process of calculating how the joints will need to move to put the end effector in the right spot is called inverse kinematics.
9:03
At Spy Codex asks, "Do you know who invented the first robot in history?"
9:28
"What was it?"
9:28
So what makes a robot anyway?
9:31
It needs to be able to sense its environment, plan using that sensor input in some smart way, and then act on the world.
9:37
So given those criteria for being a robot, the first broadly accepted real robot was Shakey.
9:48
It was invented in the mid-60s in California.
9:48
It was a robot that had a big television camera on top, and it had wheels, and it could take instructions and roll around a different obstacle course with objects in its way.
9:48
That was the foundation of a lot of the robotics that we do today.
9:48
At Edgeronia asks, "What if the only thing that's preventing a full-on global robot takeover are a handful of CAPTCHA servers?"
9:48
CAPTCHA servers are servers that run those kind of annoying little answers that you have to give to prove that you're not a robot.
10:15
You have to identify pictures that have particular objects in them.
10:21
What you're actually doing with those CAPTCHA servers is training intelligent systems on the back end.
10:27
Companies are using that data to try to build better classifiers for things like traffic lights or bicycles.
10:32
We've seen lots of movies where robots are the evil beings that are taking over from humans.
10:37
The reality is we're still really far from any notion of robots being capable enough to operate in environments in order to have any kind of impact like that.
10:47
At Local Plumbing he asks, "Could Star Wars droids be real?"
10:52
We are pretty far from robots like C-3PO and R2-D2.
10:58
They are showing super advanced versions of the technologies that researchers are working on right now.
11:04
Now these are two very different approaches, and they serve different purposes.
11:08
The C-3PO, which is a humanoid robot, it's shaped like a person, it walks upright, it uses language that people use to communicate, including the variety of different accents and dialects that people use.
11:22
And then we have our 2D2.
11:22
It's explicitly not humanoid, kind of looks like a trash can.
11:27
It uses wheels instead of legs to move around.
11:30
The language that it speaks is beeps and boops, and humans have learned the robot's language.
11:37
That's not something that we see today.
11:37
Most of the time, we are getting robots to adapt to people, and not the other way around.
11:37
At Stephen Boyle 3 asks, "Why do we need robotic delivery cool boxes driving around slowly getting in everyone's way?
11:37
Are we just making robots for the sake of making robots?"
11:37
The idea behind delivery robots is to make it more convenient for people to acquire things like groceries or delivery food.
11:37
The idea of these robots is worthy, but right now, the reality of their implementation is that they move around pretty slow, and they often do block traffic.
12:10
So the technology hasn't yet caught up to the reality.
12:14
We don't want to make robots for the sake of making robots, but I think a lot of the times we're building robots because we're trying to make life a little bit easier or a little bit more equal for people.
12:30
At CIF_KH asks, "Could a real-life Westworld be in our future?"
12:30
It takes a lot to make a robot that is indistinguishable from a human being.
12:35
This is because humans are super sophisticated.
12:37
We have dozens of muscles just in our face to make facial expressions.
12:41
Getting a robot to be capable of that requires dozens of motors, and that's not really a realistic thing right now.
12:48
I'd say we're still a pretty long way off from these kinds of hyperrealistic robots like in Westworld.
12:54
At ASME Journals asks, "What are the Nanobots?"
12:54
Nanobots are tiny robots.
12:54
That's what the word nano means in their name.
12:54
They are designed for a variety of applications, including things like medicine, going into the human body.
12:54
Because they're so small, it's hard to get electronics that will fit into them.
12:54
Often they're designed to respond to the chemical environment, for example, something that you might find in the human body.
12:54
It's actually this really interesting connection between robotics and chemistry and material science.
12:54
At Meteorologically asks, "How is robot simulation software used to program robots virtually?"
13:29
Simulation is a huge part of what we do in robot development.
13:36
We do a lot of our testing in simulation so that we know that when we bring those into the real world, that they are more likely to work.
13:43
Simulations can be limited, though.
13:50
You need to replicate real-world physics, including friction and different kinds of forces and wind shear, that can be exceedingly computationally expensive.
13:54
Simulations are still easier than putting the robots in the real world, and so there's still a hugely popular tool in robotics.
14:07
At Blely Tech asks, "How are #robots used in #space exploration?"
14:07
These robots are being deployed autonomously to roll around, collect samples, take measurements, and do all of the scientific data finding that we need to do to understand what's in space.
14:19
We've had several robots sent to Mars.
14:22
Here at Carnegie Mellon University, where I work, we have a huge project on building robots that are capable of operating autonomously on the moon.
14:31
There's a robot that's actually being developed and tested here right now, and we're hoping to send it into space soon.
14:44
At Distralc asks, "How are cobots helping to transform industry?"
14:44
A cobot is a collaborative robot specifically designed to work with human beings.
14:50
One big area that people are looking at for cobots right now is in shared manufacturing.
14:58
So you have a person who is building something in a factory, and the robot is able to do things like fetch materials for them or hold parts while they do the fine motor manipulation.
14:58
In this case, the robot is collaborative for a few reasons.
14:58
One is that it's sharing space and tools with human beings, but another is that it's actually being trained by a human being.
14:58
So it's learning how to do this from a person, and that's a big part of collaboration is this kind of sharing of information.
14:58
At Wizkele Cap says, "Agricultural robots, a growing trend."
15:29
In fact, it is.
15:32
There's a huge problem right now in growing enough food and distributing it to get to all the people in the world, and people are looking to robotics to help solve that.
15:42
Robots that can roll down a field of crops and tell a farmer whether the crop is healthy or needs different kinds of nutrients, or robots that can very quickly sort through strawberry plants to tell the farmer which ones are going to grow well and which ones need to be thrown out.
15:57
The hope is that these agricultural robots are going to be able to help farmers increase their impact and make farming in general more sustainable.
16:05
Those are all the questions for today.
16:07
Thanks so much for all the interesting questions and thanks for watching Robot Support.
16:14
She's sleeping.