The history of robotics is not a long one. It has produced many oddities, some of which capturing the public transient attention with strangely human behavior. It is sometimes amusing to watch a robot that can pick out particular shapes from a random pile of cogs and other paraphernalia, pull out the desired piece and place it carefully upright in a box on the other side of the room. It seems to achieve its task just as we might. The robot appears conscious. To the lay observer, it seems as if the robot is thinking as it hovers over the pile, scrutinizing it’s contents. “Hmm, now which one is it?” might be it’s internal monologue. Of course there is no internal monologue, no idle musing. There is only tedious calculation and geometric analysis of the visual input from the cameras trained on the pile of shapes. Ah, you say, here stands no human! It is simply the mechanical execution of thousands of lines of coding, all put in place by a human programmer. You can expose the dreadful idiocy of the robot if you, for example, ask it to pick up a piece which is too heavy for it’s pincers. The robot swiftly identifies the piece and attempts to pick it out. The piece slips from the robots grasp and falls back into the pile, making quite the clatter. Our robot seems oblivious to this however, and swings around onto the next stage, the rotation of the piece so it may be placed upright in it’s place. Now we hit a dead-end . The robot becomes paralysed, endlessly staring at it’s empty pincers trying to figure out how the nonexistant piece must be rotated to be placed in the box. What at first seemed a surreal robot representation of a human now shows up as just another mindless machine, unable to conceive of the disappearance of its prized piece. There is nothing human here, we say, it was just a carefully constructed illusion, nothing more than a complicated bicycle, and with that, we walk away.
It seems, in this way, that we still cannot in any way replicate a human, or even ever understand the deep seated mystery of consciousness. That remains the realm of the philosophers and armchair theorizing. Neurologists collaborate on huge projects to emulate cognitive ability in robots, and hope in that way to be given some insight into their own heads. The major problem with robots, the unmoveable tree trunk on the road, is that they don’t appear to experience what they are doing. We give them digital brains but they can’t dream. They have cameras, but they can’t “see”. As we would all probably agree, what it means to exist to a human is not simply function. It is to appreciate, to ponder, to experience the signals from our sensory organs and put the contextualize them, to give them meaning. It is as if there is some Theatre, hidden somewhere within the head where there is some tiny man (our soul perhaps; the “homunculus”) observing all the sensory data projected on a screen, making decisions and sending out commands. That obscure image has been dubbed the Cartesian Theatre and it is the remnants of Descartes concept of Dualism. This concept has now been disregarded by most neuroscientists, but it is no easy task putting to bed a concept whose partnering statement; Cogito Ergo Sum is the very foundation of modern Western philosophy.
Now, enter Shakey the robot. Developed at Stanford Research Institute in Menlo Park, California in the late 1960′s, Shakey is perhaps my favorite robot. Wait, scrap that. He is my favorite robot. He didn’t do anything particularly well, and wasn’t a particularly realistic simulation of human pyschology, but he illustrated shortcomings in how we percieve our thoughts in a way that no human could do before him.

Shakey was a box on wheels with a television eye, and instead of carrying his brain around with him, he was linked to it by radio. He lived in a brightly lit room and never ventured outdoors. His flatmates were a ramp, a few boxes, and few platforms and a pyramid, all brightly coloured to make vision easier for Shakey. You could communicate with Shakey by typing messages at a terminal beside his computer brain in a severly restricted vocabulary of semi-english. “Push the box off the platform” would send Shakey out, finding a box on a platform, locating a ramp, pushing the ramp into position, rolling up onto the platform and pushing the box off. Now how did Shakey achieve this? Did he too have, inside him, a tiny man watching a screen and controlling his motorized wheels? Such would be one way of looking at it, though if that was the case, I would imagine Shakeys’ programmers would be more infamous than they are now. In actuality what Shakey was using was a clever system of telling the different objects apart, which you could watch happening on a monitor in another room. A single grainy frame would appear on the monitor, an image of a box, say. The image would then be sharpened and modified in various ways, and then, though a feat of calculation, the boundries of the box would be outlined in white, and the whole image would be transformed into a line drawing. Shakey would then identify each vertex of the line drawing and determine, what was the shape of the object he was facing. If for example, he was facing the box diagonally, and he labeled the top corner facing him a Y, he would know he had a box, as from no angle would a pyramid present a Y vertex. What may have struck the observer as, as they watched on the monitor the image being sketched out, was that Shakey wasn’t looking at it. In fact, Shakey wasn’t looking at any monitors. Whose purpose did the monitor serve then? Only the observer. Whats more, the monitor could be turned off or unplugged without any detriment to Shakey’s abilities. What relation then, did the events they were watching on the monitor bear to the events going on inside Shakey?
Shakey, taking a single frame, would transform the image of a box into a sequence of 1′s and 0′s, 1 representing white and 0 representing dark. Using a set of mathmatical tools, he could then look for the reoccurance of light turning to dark or visa versa over the sequence of numbers and then from that determine a line, or outline of the box. He could then reorganise in his head the data into what we might like to refer to as an image, but in reality was nothing of the sort. This is, of course, an oversimplification, but here is a rough sketch of what Shakey was “thinking”
000000000000000111111111111111
000000000000000111111111111111
000000000000000111111111111111
000000000000000111111111111111
000000000000000111111111111111
^ a vertical line
What is interesting is that each of the processes Shakey went to to come to his conclusion is that each of the processes is, in itself, stupid and mechanical. That is, no part of the computer has to understand what it is doing and why, and there is no mystery about how each of the processes is mechanically done. Nevertheless, the clever organisation of this stupid processes takes the place of an observer telling Shakey what to do. Put the whole system then in a black box, whose whole purpose is then to tell Shakey what he needs to know. Initially, we might be inclined to think that the only way to achieve this would be to put a little man in the black box, watching a screen. But now we can see that this homunculus, with his limited job, can in fact be replaced by a machine. This is Shakey’s gift to us, a simple demonstration that we must put our old spiritualized concepts to bed, and once we do, theres no real reason that we wont’ be able to finally figure all this conciousness business out.
Posted in Uncategorized