Computer scientists concur that there are basically two long-term obstacles to having robots replace people for most repetitive tasks: vision and touch. (Cost is a short-term obstacle. Like every technological advance, mass production and economies of scale will, over time, drive the amortized cost of robots down to a level cheaper than human wages.)

The vision problem is being addressed by companies such as Microvision (MVIS:Nasdaq), as Transformational Technologies Portfolio holding. Now, BBC News reports that a new robotic sensor has the same sensitivity as a human fingertip.

The team says the tactile sensor could, in the future, aid minimally invasive surgical techniques by giving surgeons a “touch-sensation.”

Ravi Saraf, a Professor of Engineering at the University of Nebraska, is a member of the team that accomplished the breakthrough. Traditionally, robotic sensors were limited to distinguishing objects that are millimeters in diameter. That’s small, but not small enough for detail work.

Comments Dr. Saraf, “the resolution of a human fingertip is about 40 microns, about half the diameter of a human hair.”

The University of Nebraska team’s sensor is in the form of a thin film. It’s essentially a sandwich of metal and nanoparticles surrounded by electrodes.

Imagine a ballroom floor with a fine layer of dust scattered across it. Whenever someone might walk across the floor, for an instant some of the dust particles are squeezed between shoe and floor. The pattern of the dust particles will correspond precisely to the shape of the shoe.

In the sensor, this happens on a far smaller scale. Just as a finger is far tinier than a ballroom floor, so a nanoparticle is smaller than a grain of dust.

Any pressure on the film changes a micro-current of electricity in the nanoparticles squeezed. They emit tiny flashes of light that a camera instantly detects.

Not only that, but there’s a direct relationship between the amount of pressure and the intensity of the flash. The device can “touch” a penny and detect the wrinkles in President Lincoln’s clothing and the letters TY in “liberty.”

Not only is the device sensitive, but it is also able to withstand repeated use without degrading in performance. It should be cheap to manufacture, especially in quantity.

Prof. Saraf foresees medical applications. For example, the device could become part of a probe inserted into a patient to determine whether tissue is cancerous (cancerous tissue often feels different from normal tissue). It could probe an arterial wall for weaknesses, or the inside of a colon for polyps.

In principle, there is no reason why the device couldn’t be miniaturized. This would potentially allow further uses, such as monitoring the healing of a wound or surgical procedure.

Dr. Saraf seeks to add temperature sensitivity to the sensor, a capability that would make it fully rival the human finger.

This sensor should enable a superior wiper system for vehicles — one that continuously measures the density of rain spattering a windshield and adjusts the speed to suit.

What of the robotic applications? In principle, there’s no reason why this technology couldn’t enable robots to do household and industrial tasks with far greater efficiency. Currently, the Roomba automatic vacuum cleaners can’t distinguish a fallen ring from a pop-top lid. That’s a problem.

I foresee truly automatic home cleaning systems within five years. As this technology gets adapted to those applications, expect more interesting ones to follow.

For instance, have you tried one of the automated massage chairs or tables? The best ones are good, but not as good as a talented human masseuse. One reason is that they can’t monitor the pressure they’re applying and therefore adjust accordingly. This capability will bring them closer to human performance.

Over time, as the noted computer scientist Dr. Marshall Brain has noted, robots and computers will displace humans from more and more tasks. Giving them a sense of touch is a huge milestone in this advance.

I fully expect this technology to be widely licensed due to its far-ranging applications. I’ll be watching for startups that find ingenious uses for it.

The robotic revolution is coming — and we’ll profit from it.

To your profitable future,

Jonathan Kolber
June 14, 2006

One Small Touch for Man, One Giant Touch for Machine was originally featured in the Penny Sleuth.