This weekend is supposed to be a scorcher. According to the latest meteorologist reports, in Death Valley it might reach 130 degrees. Here is New York, it is so humid that one cannot tell the difference between swimmers drying off and people leaving work. There is one bright point, a cold one tastes better on a hot day.
At Cornell University, Hema S. Koppula, graduate student in computer science, and Ashutosh Saxena, an assistant professor of computer science are have built a robotic bartender for just these warm days. According to their research, PR-2 robot can not only carry out everyday tasks, but even anticipate human behavior and adjust its actions to their needs (like fill my glass when it is empty or find the remote).
PR-2 creation is no small task as grasping is one of the hardest things to teach robots – too hard and its crushed; too soft and it falls down. For example, what would happen when a human unintentionally moves their glass as the robot goes to top it up? This could get very messy, so the robot needs to anticipate possible human actions and adjust accordingly. If it sees someone reaching for the cup, the robot has to know when to stop trying to pour.
The Cornell anticipatory robot avoids embarrassing spills and other accidents by using its Microsoft Kinect scanner (one of the best ones on the market) to build up a 3D map of the objects present and then calculating how they might be used based on the action currently being performed by the person. The robot then manages this by means of a database of 120 3D videos of people performing everyday household tasks, from which it reduces the person’s movements to symbolic skeletons. It then classifies these skeletons into subactivities, such as reaching, pouring and carrying, for example, while associating different objects with different actions.
According to tests carried out by its researchers, the PR2-robot is able to correctly predict actions 82% of the time when looking one0second ahead, 71 percent for three seconds, and 57 percent for 10 seconds. So there’s still some time before this robot shows up at your local watering hole.
“Even though humans are predictable, they are only predictable part of the time,” Saxena said. “The future would be to figure out how the robot plans its action. Right now we are almost hard-coding the responses, but there should be a way for the robot to learn how to respond.”
In my opinion robots will have a lot easier time predicting the behavior of guys (beer=happiness) vs. women (?=happiness), which require such enhanced technology that it could be eons ways.
Contributing Source: Cornell University & The Wall Street Journal