June 28, 2011 10:44am ETRobots Learn Language to Navigate Tough Situations Robots, Language & Learning  

Linguist Jeff Heinz and mechanical engineer Bert Tanner are teaching robots about cooperative behavior, by observing how children learn to communicate and how adults learn new languages. Credit: NSF View full size image

Equipping robots with language and learning capabilities could take some of the heat off human handlers, enabling the robots to navigate tough tasks in small groups.

The ability to communicate with others could help the robots in dangerous situations, where they can work in teams to navigate building fires or other disasters. If they can learn and communicate with each other, they could take action without needing a human to give commands.

Now, University of Delaware researchers are designing robot language and learning capabilities based on the complex rules that human language follows. They are teaching robots to communicate using their own kind of sentences, which string together actions the robot performs in a certain order; for example, the robot can't drop a ball until it has first picked on up.

"So a robot's 'sentence' in a sense, is just a sequence of actions that it is conducting," said Jeffrey Heinz, a linguist at the University of Delaware. "And there will be constraints on the kinds of sequences of actions that a robot can do."

Communication between robots gives them an ability to learn about their environments and other robots, if they catalogue this incoming information. "We would like to make the robots adaptive — learn about their environment and reconfigure themselves based on the knowledge they acquire," explains researcher Bret Tanner, also of the University of Delaware.

Each robot would have different abilities and follow a different set of rules, so the robots could work together to accomplish tasks.Each robot would play to its talents and strengths. Ultimately, the robots will need to be aware of their own capabilities, those of the other robots around them and the overall goal of their mission.

"We have eyes that see, ears that hear and we have fingers that touch. We don’t have a 'universal sensory organ,'" Heinz said. "Likewise in the robotics world, we're not going to design a universal robot that's going to be able to do anything and everything."

机器人学语言以应对艰难的处境

2011年6月28日上午10时44分ETRobots学语到导航艰难的情况下机器人，语言和放大器; Learning

语言学家杰夫·海因茨和机械工程师伯特·坦纳是教学有关合作行为的机器人，通过观察孩子如何学会沟通和成年人如何学习新的语言。图片来源：美国国家科学基金会查看原图

装备机器人的语言和学习能力，可能需要一些热关人类的处理程序，使机器人导航艰难的任务三五成群.

与人沟通的能力可以帮助在危险的机器人的情况下，在那里他们可以在团队中工作，以浏览建筑发生火灾或其他灾害。如果他们可以学习和彼此通信，他们可以采取行动，而无需人类，得到命令.

现在，特拉华大学的研究人员正在设计机器人语言和基于人类语言遵循的复杂规则的学习能力。他们正在教机器人利用自己种的句子，这串动作的机器人按照一定的顺序进行沟通;例如，机器人不能降一球，直到它第一次拿起多达.

因此，在某种意义上机器人的“一句话”，是行动，它正在进行，与只是一个序列;杰弗里说亨氏，语言学家在特拉华大学。 ;以及将有约束的各种动作序列，机器人可以做;机器人之间

通讯给了他们一个了解自己的环境和其他机器人，如果目录此输入的信息的能力。;我们希望让机器人适应mdash;了解他们的环境，并重新配置自己的基础上，他们获得的知识，;研究人员解释说布雷特坦纳还特拉华大学，.

每个机器人将有不同的能力，并按照一套不同的规则，所以机器人可以共同完成tasks.Each机器人将发挥其聪明才智和优势。最终，机器人将需要了解自己的能力，他们周围的其他机器人和他们的使命.

的总体目标;我们眼睛看到的，耳朵听到的，我们有手指的那一抹。我们不＆rsquo; t有一个“普遍的感觉器官，海因茨说。;同样，在机器人世界，我们不会设计一个通用的机器人将是能够做任何事情，一切