<-
The previous definition of robotic systems is a stark reminder of how little progress has been made in the robot part of robotics. Certainly, the automation and machinery part of robotics has made great strides, but there’s more to a robot than automation and machinery. We can look to the recent 2015 Darpa Robotics Challenge finals to see just how far the robot part of robotics has to go before approaching anything even reminiscent of the creatures envisioned by Čapek. As background, the 2015 Darpa Robotics Challenge finals showcased the best “robots” in the world. Each of these robots had been under development for at least three years with many millions of dollars in funding per robot. The tasks the robots faced in the challenge were quite modest. They included walking up a flight of stairs, turning a knob, flipping a switch, and the like. Conceptually these tasks are no more complex than the tasks a fruit fly faces every day. Most of the robots couldn’t complete the tasks, but two of them did. The winning robot weighed 407 pounds, was powerful enough to bench-press 150 pounds, and didn’t look much like a human. That is woefully short of Čapek’s vision.
With the admission that the robots Čapek describes do not exist, our attention turned to why? Is there a fundamental limitation preventing us from developing the robot’s body, from developing the robot’s brain, or is there just not enough demand to drive this development? The limitation does not appear to be in building the robot’s body and frame. That certainly is a technical challenge, but to see that it is not insurmountable, one only needs to look at modern cell phones. The microprocessors, touch screens, GPS, etc. components in these phones require whole industries to support and manufacture. The feature sizes in these devices are measured in nanometers, but we build them because there is demand to support it. Similarly, the world-wide demand for the kinds of robots Čapek described would be expected to be enormous [2]. There is demand for them and we can build them, so then why don’t they exist?
The answer is that there is no path from digital computers based on semiconductor technology to the complexity of the human brain. Those who do say we’ll have robots with brains as powerful as humans generally base their argument on Moore’s law. Moore’s law says that the number of transistors on a computer processor doubles approximately every two years. There are currently billions of transistors on a computer processor [4] and roughly 100 trillion connections in the human brain [6]. If Moore’s law continues to hold, the number of transistors on a computer processor will equal the number of connections in a human brain in about 30 years, but there are two issues with this line of reasoning. First, the interconnections in human brains are much more complex than binary computer connections. The connections in human brains are electro-chemical and involve firing rates. This makes them more like analog signals than digital signals. If we assume the resolution of a connection in a human brain is 8 bits and use the 100 trillion connections estimate, then the calculation for the required number of binary components becomes 300*10^12
transistors. →
|
