Young animals gallop across fields, climb trees and immediately find their feet with enviable grace after they fall.
年轻的动物奔跑穿过田野，爬树，哪怕跌倒也能立刻优雅地重新爬起。
And like our primate cousins, humans can deploy opposable thumbs and fine motor skills to complete tasks such as effortlessly peeling a clementine or feeling for the correct key in a dark hallway.
和我们的灵长类表亲一样，人类也可以运用对生拇指和精细的运动技能来完成一些任务，比如毫不费力地剥开柑橘皮，或者在黑暗的走廊里寻找正确的钥匙。
Although walking and grasping are easy for many living things, robots have been notoriously poor at gaited locomotion and manual dexterity.
虽然行走和抓取对许多生物来说都很容易，但机器人在步态移动和灵巧性方面一直臭名昭著。
Until now.
直到后来。
Writing in Science Robotics, Hwangbo et al. report intriguing evidence that a data-driven approach to designing robotic software could overcome a long-standing challenge in robotics and artificial-intelligence research called the simulation–reality gap.
Hwangbo等人在《科学机器人学》杂志上撰文，报告了一个有趣的证据，表明设计机器人软件的数据驱动方法可以克服机器人和人工智能研究中一个长期存在的挑战，即模拟与现实之间的差距。
For decades, roboticists have guided the limbs of robots using software that is built on a foundation of predictive, mathematical models, known as classical control theory.
几十年来，机器人专家一直在使用一种软件来引导机器人的四肢，这种软件是建立在经典控制理论的预测数学模型基础上的。
However, this method has proved ineffective when applied to the seemingly simple problem of guiding robotic limbs through the tasks of walking, climbing and grasping objects of various shapes.
然而，这种方法在引导机器人肢体完成各种类型的行走、攀爬和抓取任务这一看似简单的问题上被证明是无效的。
A robot typically begins its life in simulation.
机器人通常在模拟中开始它的生命。
When its guiding software performs well in the virtual world, that software is placed in a robotic body and then sent into the physical world.
当它的引导软件在虚拟世界中表现良好时，该软件就被放置在一个机器人体内，然后送入物理世界。
There, the robot will inevitably encounter limitless, and difficult to predict, irregularities in the environment.
在那里，机器人将不可避免地遇到无限的、难以预测的不规则环境。
Examples of such issues include surface friction, structural flexibility, vibration, sensor delays and poorly timed actuators — devices that convert energy into movement.
这类问题的例子包括表面摩擦、结构灵活性、振动、传感器延迟和并非实时的执行器——将能量转化为运动的装置。
Unfortunately, these combined nuisances are impossible to describe fully, in advance, using mathematics.
不幸的是，这些复杂的麻烦是不可能提前用数学来充分描述的。
As a result, even a robot that performs beautifully in simulation will stumble and fall after a few encounters with seemingly minor physical obstacles.
因此，即使是在模拟中表现出色的机器人，在遇到一些看似很小的物理障碍后也会磕磕绊绊。
Hwangbo et al. have demonstrated a way of closing this performance gap by blending classical control theory with machine-learning techniques.
Hwangbo等人通过将经典控制理论与机器学习技术相结合，展示了一种缩小这种性能差距的方法。
The team began by designing a conventional mathematical model of a medium-dog-sized quadrupedal robot called ANYmal (Fig. 1).
研究小组首先设计了一个名为ANYmal的中型四足机器人的传统数学模型(如下图)。
Next, they collected data from the actuators that guide the movements of the robot’s limbs.
接下来，他们从引导机器人肢体运动的执行器中收集数据。
They fed this information into several machine-learning systems known as neural networks to build a second model — one that could automatically predict the idiosyncratic movements of the AMYmal robot’s limbs.
他们将这些信息输入几个被称为神经网络的机器学习系统中，建立第二个模型——这个模型可以自动预测机器人肢体的特殊运动。
Finally, the team inserted the trained neural networks into its first model and ran the hybrid model in simulation on a standard desktop computer.
最后，研究小组将训练好的神经网络插入到第一个模型中，并在一台标准台式电脑上运行混合模型。
The hybrid simulator was faster and more accurate than a simulator that was based on analytical models.
混合仿真器比基于分析模型的仿真器速度快，精度高。
But more importantly, when a locomotion strategy was optimized in the hybrid simulator, and then transferred into the robot’s body and tested in the physical world, it was as successful as it was in simulation.
但更重要的是，当一个运动策略在混合模拟器中进行优化，然后转移到机器人的身体，并在物理世界中进行测试时，会跟模拟时一样成功。
This long-overdue breakthrough signals the demise of the seemingly insurmountable simulation–reality gap.
这一迟来的突破标志着看似不可逾越的模拟与现实鸿沟的终结。
The approach used by Hwangbo et al. hints at another major shift in the field of robotics.
Hwangbo等人使用的方法暗示了机器人领域的另一个重大转变。
Hybrid models are the first step towards this change.
而混合模型是实现这一变化的第一步。
The next step will be to retire analytical models altogether, in favour of machine-learning models that are trained using data collected from a robot’s real-world environment.
下一步将是彻底淘汰分析模型，代之以使用从机器人真实环境中收集的数据进行训练的机器学习模型。
Such data-pure approaches — referred to as end-to-end training — are gaining momentum.
这种纯数据的方法——称为端到端培训——正在成为主流。
Several innovative applications have already been reported, including articulated robotic arms, multi-fingered mechanical hands, drones and even self-driving cars.
一些创新的应用已经被报道，包括铰接式机器人手臂、多指机械手、无人机，甚至自动驾驶汽车。
For now, roboticists are still learning to harness the power of faster computation, an abundance of sensor data and improvements in the quality of machine-learning algorithms.
目前，机器人专家仍在研究如何利用更快的计算速度、丰富的传感器数据以及机器学习算法质量的提高。
It is not yet clear whether it is time for universities to stop teaching classical control theory.
目前还不清楚大学是否应该停止教授经典控制理论。
However, I think that the writing is already on the wall: future roboticists will no longer tell robots how to walk.
然而，我认为这已经是不祥之兆:未来的机器人专家将不再告诉机器人如何行走。
Instead, they will let robots learn on their own, using data that are collected from their own bodies.
相反，他们将让机器人自己学习，使用从自己身体收集的数据。
问题：
文中提到新型机器人在哪方面进行了最重要的创新？
A.机器人外形
B.机器人肢体的数量
C.机器人的材料
D.机器人的学习模型
留言回复正确答案，前十名朋友可以获得红包哦！赶快来试试吧！
感谢关注
跟amber一起看世界
联系客服
