(Laughter)
(笑声)
(Laughter)
(笑声)
That's SpotMini. He'll be back in a little while.
这是SpotMini, 它过一会儿还会回来。
I --
我——
(Applause)
(掌声)
I love building robots. And my long-term goal is to build robots that can do what people and animals do. And there's three things in particular that we're interested in. One is balance and dynamic mobility, the second one is mobile manipulation, and the third one is mobile perception.
我喜欢建造机器人。 我远期的目标是发明 可以像人类和动物一样行动的机器人。 我们尤其感兴趣的 有3个方面, 其一是平衡和动态运动, 其二是移动操控, 其三是移动感知。
So, dynamic mobility and balance -- I'm going to do a demo for you. I'm standing here, balancing. I can see you're not very impressed. OK, how about now?
首先,关于动态运动和平衡—— 我来做个演示。 我站在这里。保持着平衡。 我知道你们并没有 觉得很惊讶,那现在呢?
(Laughter)
(笑声)
How about now?
这次呢?
(Applause)
(鼓掌)
Those simple capabilities mean that people can go almost anywhere on earth, on any kind of terrain. We want to capture that for robots.
这些简单的能力使得 我们可以走到世界各个角落, 无论哪种地形。 我们也希望机器人可以这样。
What about manipulation? I'm holding this clicker in my hand; I'm not even looking at it, and I can manipulate it without any problem. But even more important, I can move my body while I hold the manipulator, the clicker, and stabilize and coordinate my body, and I can even walk around. And that means I can move around in the world and expand the range of my arms and my hands and really be able to handle almost anything. So that's mobile manipulation. And all of you can do this.
那么,操控能力呢? 我手里拿着点击器, 我都不用看, 就可以轻而易举地操控它。 但更重要的是, 我可以在拿着操控器的 时候随意变换姿势, 我的身体还是很稳定,保持协调, 我还能随意走动。 也就是说我可以来回走, 即使伸开我的手臂和手, 我也可以拿着几乎任何的东西。 这就是所谓的移动操控。 你们都能做到。
Third is perception. I'm looking at a room with over 1,000 people in it, and my amazing visual system can see every one of you -- you're all stable in space, even when I move my head, even when I move around. That kind of mobile perception is really important for robots that are going to move and act out in the world.
第三是感知。 我看着有1000多个人的房间, 我的视觉系统 让我可以看见你们每一个人, 你们都静静的坐着, 即使我有时转个头, 或者来回走。 这种移动感知对于 要移动和做动作的 机器人来说, 很重要。
I'm going to give you a little status report on where we are in developing robots toward these ends. The first three robots are all dynamically stabilized robots. This one goes back a little over 10 years ago -- "BigDog." It's got a gyroscope that helps stabilize it. It's got sensors and a control computer.
我在这里要分享一个状态报告, 关于我们研究机器人的进展现状。 前三种都是有动态稳固性的机器人, 这个是10年前设计的, BigDog。 它有一个陀螺仪来保持稳定, 装有传感器和控制电脑。
Here's a Cheetah robot that's running with a galloping gait, where it recycles its energy, it bounces on the ground, and it's computing all the time in order to keep itself stabilized and propelled.
这个是猎豹机器人, 以飞快的步伐跑动, 它可以重复循环利用能源, 在地上跳跃, 为了能保持稳定的驱进, 全程都是电脑操控。
And here's a bigger robot that's got such good locomotion using its legs, that it can go in deep snow. This is about 10 inches deep, and it doesn't really have any trouble.
这个更大的机器人, 能够熟练使用自己的腿, 在雪中行走, 这个雪大约 25 厘米厚, 行走完全没有问题。
This is Spot, a new generation of robot -- just slightly older than the one that came out onstage. And we've been asking the question -- you've all heard about drone delivery: Can we deliver packages to your houses with drones? Well, what about plain old legged-robot delivery?
这个新一代机器人Spot, 就比刚才台上的那个 设计得早了一点点, 我们一直在问一个问题—— 你们都知道无人机递送: 我们能用无人机送货上门吗? 那用有腿的普通老式 机器人送货怎么样?
(Laughter)
(笑声)
So we've been taking our robot to our employees' homes to see whether we could get in --
我们把机器人带到员工的家, 看看是否能进去——
(Laughter)
(笑声)
the various access ways. And believe me, in the Boston area, there's every manner of stairway twists and turns. So it's a real challenge. But we're doing very well, about 70 percent of the way.
进门的路径五花八门。 相信我,波士顿这一带, 有着各式各样弯曲的楼梯和转弯。 这真的是一个挑战。 不过大约70%的路, 我们都走得很顺利。
And here's mobile manipulation, where we've put an arm on the robot, and it's finding its way through the door. Now, one of the important things about making autonomous robots is to make them not do just exactly what you say, but make them deal with the uncertainty of what happens in the real world. So we have Steve there, one of the engineers, giving the robot a hard time.
这是移动操控, 我们给机器人放了一个手臂, 它在找方法进门。 不要让它们 完全按照你的指示行动, 是发明自主机器人很重要的一点。 要让它们学会处理 真实世界里发生的不确定事件, 这是我们其中一名工程师,史蒂夫, 他正在给机器人制造麻烦。
(Laughter)
(笑声)
And the fact that the programming still tolerates all that disturbance -- it does what it's supposed to.
事实上,所编写的程序 仍然能让它抵抗干扰, 做它该做的事。
Here's another example, where Eric is tugging on the robot as it goes up the stairs. And believe me, getting it to do what it's supposed to do in those circumstances is a real challenge, but the result is something that's going to generalize and make robots much more autonomous than they would be otherwise.
这是另一个例子, 艾瑞克在机器人上楼梯的时候 在后面拖拽它。 相信我, 在这种情况下, 让它们保持原计划的行动 绝非易事, 不过成果可以进行推广, 让机器人比之前更加自主。
This is Atlas, a humanoid robot. It's a third-generation humanoid that we've been building. I'll tell you a little bit about the hardware design later. And we've been saying: How close to human levels of performance and speed could we get in an ordinary task, like moving boxes around on a conveyor? We're getting up to about two-thirds of the speed that a human operates on average. And this robot is using both hands, it's using its body, it's stepping, so it's really an example of dynamic stability, mobile manipulation and mobile perception. Here --
这个像人的机器人是Atlas, 这是我们发明的第三代的机器人。 待会儿我会介绍一些 硬件设计方面的事情。 我们一直在说: 在一项普通的任务中, 比如在传送带上 移动箱子, 它的性能和速度 跟人类相比还有多少差距? 平均来讲,我们大约可以 达到人力操纵速度的 三分之二。 这个机器人正在使用 它的两只手和身体, 同时也在行走, 这是动态稳定, 移动操控, 移动感觉的很好的例子。 这里——
(Laughter)
(笑声)
We actually have two Atlases.
我们实际上有两个Atlas。
(Laughter)
(笑声)
Now, everything doesn't go exactly the way it's supposed to.
事情不一定会按照 预想的情况发生。
(Laughter)
(笑声)
(Laughter)
(笑声)
(Laughter)
(笑声)
And here's our latest robot, called "Handle." Handle is interesting, because it's sort of half like an animal, and it's half something else with these leg-like things and wheels. It's got its arms on in kind of a funny way, but it really does some remarkable things. It can carry 100 pounds. It's probably going to lift more than that, but so far we've done 100. It's got some pretty good rough-terrain capability, even though it has wheels. And Handle loves to put on a show.
这是我们最新的机器人,Handle, 它很有意思,有着 像腿又像轮子的东西, 因为它一半像动物, 一半像其他的东西。 它的手臂很有趣, 但是确实能做出些 让人惊叹的事情。 它能够承受100磅的重量。 很有可能打算举起更重的东西, 不过目前我们只尝试到100。 它可以很好的适应地形, 即使装有轮子。 Handle还很喜欢表演。
(Laughter)
(笑声)
(Applause)
(鼓掌)
I'm going to give you a little bit of robot religion. A lot of people think that a robot is a machine where there's a computer that's telling it what to do, and the computer is listening through its sensors. But that's really only half of the story. The real story is that the computer is on one side, making suggestions to the robot, and on the other side are the physics of the world. And that physics involves gravity, friction, bouncing into things. In order to have a successful robot, my religion is that you have to do a holistic design, where you're designing the software, the hardware and the behavior all at one time, and all these parts really intermesh and cooperate with each other. And when you get the perfect design, you get a real harmony between all those parts interacting with each other. So it's half software and half hardware, plus the behavior.
我想向你们介绍一点 机器人学的相关知识。 很多人认为机器人就是一个 听从电脑指示的机器, 同时电脑从传感器接受信息。 但这只是其中一半。 事实是,电脑在一端 给机器人指示, 而另一端是物理世界。 包括重力,摩擦力,弹力。 为了研究成功的机器人, 我的观念是,必须要做好全盘设计, 同时要包括软件,硬件 和行为, 这些部分必须相互结合并且协调, 只有你的设计达到完美, 在各个部分相互契合的时候, 才会实现真正的协调。 所以是50%软件和50%硬件, 再加上行为。
We've done some work lately on the hardware, where we tried to go -- the picture on the left is a conventional design, where you have parts that are all bolted together, conductors, tubes, connectors. And on the right is a more integrated thing; it's supposed to look like an anatomy drawing. Using the miracle of 3-D printing, we're starting to build parts of robots that look a lot more like the anatomy of an animal. So that's an upper-leg part that has hydraulic pathways -- actuators, filters -- all embedded, all printed as one piece, and the whole structure is developed with a knowledge of what the loads and behavior are going to be, which is available from data recorded from robots and simulations and things like that.
最近我们在研究硬件部分, 希望能够做到—— 图片左边是传统设计, 可以看到导体、管子、连接原件等等 都被衔接在一起。 而右边的设计则更加集成化, 看起来就像解剖图像。 借助神奇的3D打印, 我们可以把机器人的部件打造得 看起来就像动物被解剖的部位。 大腿部分有液压通路, 包括促进器和过滤器—— 都是嵌入式的, 作为单一部件被打印出来, 这样整体结构就做成了。 根据机器人采集的数据, 和一些模拟计算,我们可以推测出 它能够完成的工作量和行为。
So it's a data-driven hardware design. And using processes like that, not only the upper leg but some other things, we've gotten our robots to go from big, behemoth, bulky, slow, bad robots -- that one on the right, weighing almost 400 pounds -- down to the one in the middle which was just in the video, weighs about 190 pounds, just a little bit more than me, and we have a new one, which is working but I'm not going to show it to you yet, on the left, which weighs just 165 pounds, with all the same strength and capabilities. So these things are really getting better very quickly.
所以这是数据驱动的硬件设计。 变化过程如下, 不只是大腿,还包括其他部位, 我们的机器人从最初 庞大笨拙的造型—— 右边那个大约400傍重—— 进化到中间这个, 刚刚视频中的版本, 大约重190傍, 比我重一点点, 我们还有个新的机器人, 它可以运行,但是我 今天不会给大家演示, 就是左边这个, 重约165傍, 却有着同样的力量和能力。 所以这些东西更新很快。
So it's time for Spot to come back out, and we're going to demonstrate a little bit of mobility, dexterity and perception. This is Seth Davis, who's my robot wrangler today, and he's giving Spot some general direction by steering it around, but all the coordination of the legs and the sensors is done by the robot's computers on board. The robot can walk with a number of different gaits; it's got a gyro, or a solid-state gyro, an IMU on board. Obviously, it's got a battery, and things like that. One of the cool things about a legged robot is, it's omnidirectional. In addition to going forward, it can go sideways, it can turn in place. And this robot is a little bit of a show-off. It loves to use its dynamic gaits, like running --
现在该让Spot出场了, 我们现在来演示一些运动能力, 敏捷性和感知能力。 这是赛斯·戴维斯, 是机器人的操纵者, 他会通过遥控器, 给Spot一些大概的指导, 但是它的腿和传感器的配合, 是由自身配备的电脑控制。 它有很多形式的步伐, 它有一个陀螺仪, 固态的陀螺仪, 搭载有IMU。 很明显,它还有电池等等。 有腿的机器人很酷的一点就是, 它能向全方位行走。 不仅可以直走,还可以横着走, 也可以原地转弯。 这个机器人挺喜欢炫耀。 它喜欢用动态步伐, 比如跑步。
(Laughter)
(笑声)
And it's got one more.
还有一个。
(Laughter)
(笑声)
Now if it were really a show-off, it would be hopping on one foot, but, you know.
如果它确实很喜欢炫耀, 就会展示单脚跳了, 不过还没到那种程度。
Now, Spot has a set of cameras here, stereo cameras, and we have a feed up in the center. It's kind of dark out in the audience, but it's going to use those cameras in order to look at the terrain right in front of it, while it goes over these obstacles back here. For this demo, Seth is steering, but the robot's doing all its own terrain planning. This is a terrain map, where the data from the cameras is being developed in real time, showing the red spots, which are where it doesn't want to step, and the green spots are the good places. And here it's treating them like stepping-stones. So it's trying to stay up on the blocks, and it adjusts its stride, and there's a ton of planning that has to go into an operation like that, and it does all that planning in real time, where it adjusts the steps a little bit longer or a little bit shorter.
Spot装有一系列摄像头, 立体摄像头, 这里中间有一个。 对着观众有点暗, 但是在它跨过它前方障碍的时候, 它会利用镜头看清 前方的地形。 下面这个演示,由赛斯控制, 但是机器人会自己做地形探测, 这是利用镜头得到实时的数据, 制作的地形勘测地图, 红点就是它不想踩的地方, 而绿点就是可行的区域。 它把这些障碍物当做垫脚石。 它会尝试待在板上, 调整步伐, 像这样的行为 包含了许多计划步骤。 它时时刻刻都在计划着, 把步伐调整得大一点, 或者小一点。
Now we're going to change it into a different mode, where it's just going to treat the blocks like terrain and decide whether to step up or down as it goes. So this is using dynamic balance and mobile perception, because it has to coordinate what it sees along with how it's moving.
现在我们来换一个模式, 让它把障碍当成地形来看, 在它走路时候来决定 上来还是下去。 这就用到了动态平衡 以及移动感知, 因为它要根据看到的场景 来协调自己的动作。
The other thing Spot has is a robot arm. Some of you may see that as a head and a neck, but believe me, it's an arm. Seth is driving it around. He's actually driving the hand and the body is following. So the two are coordinated in the way I was talking about before -- in the way people can do that. In fact, one of the cool things Spot can do we call, "chicken-head mode," and it keeps its head in one place in space, and it moves its body all around. There's a variation of this that's called "twerking" --
另外,Spot还有一条机器人手臂。 你们可能觉得长得像头或者是脖子, 但是相信我,这是手臂。 赛斯在驱动这条手臂。 他实际上只需要操控手臂, 下面的身体会跟着运动。 所以这两个相互协调, 就如我之前提到过的—— 就像人类一样。 实际上,Spot还可以做 一件很酷的事,称为“鸡头模式”。 它的头保持不变, 只是来回移动身体。 还有另一种模式叫“电臀”——
(Laughter)
(笑声)
but we're not going to use that today.
不过我们就不演示了。
(Laughter)
(笑声)
So, Spot: I'm feeling a little thirsty. Could you get me a soda? For this demo, Seth is not doing any driving. We have a LIDAR on the back of the robot, and it's using these props we've put on the stage to localize itself. It's gone over to that location. Now it's using a camera that's in its hand to find the cup, picks it up -- and again, Seth's not driving. We've planned out a path for it to go -- it looked like it was going off the path -- and now Seth's going to take over control again, because I'm a little bit chicken about having it do this by itself. Thank you, Spot.
Spot,我有点口渴, 能不能给我一杯饮料? 在这个演示中,Seth不会进行操控。 在机器人后边有LIDAR, 用这些我们放在台上的道具 来定位自己。 它现在走过去了, 在使用手上的摄像头 来找到杯子, 拿起它—— 别忘了,赛斯并没有进行操控。 我们给它安排了路线, 现在看来它好像走偏了—— 现在赛斯要恢复控制, 因为我有点不信任 让它自己完成这事。 谢谢你,Spot。
(Applause)
(掌声)
So, Spot: How do you feel about having just finished your TED performance?
那么,Spot, 完成你的TED表演感觉怎么样?
(Laughter)
(笑声)
Me, too!
我也是!
(Laughter)
(笑声)
Thank you all, and thanks to the team at Boston Dynamics, who did all the hard work behind this.
谢谢大家。 也感谢为这项任务付出努力的 波士顿动力公司的团队。
(Applause)
(掌声)
Helen Walters: Marc, come back in the middle. Thank you so much. Come over here, I have questions.
海伦·沃特斯: 马克,能回到场地中间吗? 非常感谢。 往前站一下,我有几个问题。
So, you mentioned the UPS and the package delivery. What are the other applications that you see for your robots?
你提到了UPS和包裹递送。 你还在探索机器人的其他应用吗?
Marc Raibert: You know, I think that robots that have the capabilities I've been talking about are going to be incredibly useful. About a year ago, I went to Fukushima to see what the situation was there, and there's just a huge need for machines that can go into some of the dirty places and help remediate that.
马克·莱伯特:我认为 机器人将会拥有的能力 会有非常大的用途。 大约一年以前,我去了福岛, 想看看那里的情况, 对能够在污染地进行清理 和重建工作的机器人的 需求会很大。
I think it won't be too long until we have robots like this in our homes, and one of the big needs is to take care of the aging and invalids. I think that it won't be too long till we're using robots to help take care of our parents, or probably more likely, have our children help take care of us. And there's a bunch of other things. I think the sky's the limit. Many of the ideas we haven't thought of yet, and people like you will help us think of new applications.
我认为我们的家里 很快也会有这样的机器人, 还有就是对于照顾老人和伤残者 这方面也有会很大需求。 我同样认为我们依靠 机器人照顾父母 也不会很久了。 更有可能的是,让孩子来照顾我们。 还有很多的事情。 但我知道存在一种界线。 很多点子我们还没有想出来, 你们可以帮助我们 想一些新的应用。
HW: So what about the dark side? What about the military? Are they interested?
沃特斯:缺点有什么呢? 军方呢? 军方感兴趣吗?
MR: Sure, the military has been a big funder of robotics. I don't think the military is the dark side myself, but I think, as with all advanced technology, it can be used for all kinds of things.
莱伯特:当然,军队是 自动机械的发源地。 我自己并不认为军用算是坏处, 但是我认为先进的科技 可以用于所有领域。
HW: Awesome. Thank you so much.
沃特斯:太棒了,谢谢你。
MR: OK, you're welcome.
莱伯特:不客气。
Thank you.
沃特斯:谢谢。
(Applause)
掌声