Servo - September 2012

Sensors For Mobile Robots -— Part 4

FIGURE 7. An illustration of the rock-paper-scissors robot's operation.

Oku Lab researchers at the University of Tokyo, a high speed overhead camera (shown in Figure 7) controls the mechanical hand that is interconnected through a computer. The camera reads the human’s hand movements in one millisecond and quickly responds with the solution that is fed to the mechanical hand. As the lab personnel say, “In other words, the robot is capable of cheating faster than a human is capable of seeing.”

Smartphone Voice Recognition Apps

With very little new speech recognition hardware being developed during the 2008-2012 time period, smartphone apps took center stage. For example, Siri is an app that seemed to take the technical world by storm in 2011. Shown in Figure 8 as an ad, it is much more advanced than crude programs like ELIZA. Siri was part of Apple’s marketing and is a popular app for all iPhones.

It has been said that “This was more than enough to rejuvenate the speech recognition industry, while putting the pressure back on to create better iterations of speech recognition software.” Google’s Voice Search app became available for the iPhone in 2008. This app relies on Google’s cloud data center to process voice requests, matching them with the huge pool of human-speech samples and search queries collected by Google. Their newer Android has a similar program called ‘the Assistant’ that is supposed to be pushed big time in the fourth quarter of 2012.

The Leap Motion Sensor that Recognizes Human Hand Gestures

The little object that looks like a small candy bar in front of the laptop computer in Figure 9 is actually a motion sensor for computers that is reportedly going to cost only $70. It is called the Leap and is made by Leap

FIGURE 8. The popular Siri smartphone app.

Motion — a small San Francisco company founded by Michael Buckwald and David Holz. As they state, “Leap represents an entirely new way to interact with your computers. It’s more accurate than a mouse, as reliable as a keyboard, and more sensitive than a touch screen. For the first time, you can control a computer in three dimensions with your natural hand and finger movements. This isn’t a game system that roughly maps your hand movements. The Leap technology is 200 times more accurate than anything else on the market — at any price point. Just about the size of a flash drive, the Leap can distinguish your individual fingers and track your movements down to 1/100th of a millimeter.” This unique device caught my eye as the ideal sensor for a small robot since it is a close-up gesture sensor unlike the game-type Kinects by Microsoft with a minimum 40 cm range. The extreme accuracy has been the fodder of many online sites such as extremetech.com, and people have assumed that it uses time-of-flight technology that measures the time it takes light to travel to and from an object. The present models shown here and on different sites are just mockups and not actual working prototypes, so the technology is ripe for all types of robot applications. It’s due out the end of this year or January 2013.