Nicholas Thompson emitted this, circa 09-06-15 10:25 AM: > Let us imagine that we want to program a robot to do stuff .... many of > you have, I gather. Now, I assume that any robot worth it's salt, will > have a certain amount of self knowledge. It will know, for instance, > where it is. It will know the position of its effectors. It may also > know something about what it has done recently. > > So how do roboteers provide their robots with such knowledge.?
This depends entirely on the robot. Most robots are controlled with a microcontroller. And in that, they are nothing more than general purpose computers with sensors attached. But some robots (e.g. BEAM) are (basically) just coupled oscillators with sensors attached. Perhaps there are other types; but these will serve for your question. For a microcontroller-based robot, the "knowledge" is programmed in by the programmer. All such knowledge is explicitly grounded by the programmer to the I/O data for the programs. For example, if the program is: read magnetic field -> turn motor, then the robot's "knowledge" is about magnetic fields and motors. In contrast, however, for a BEAM robot, what the robot "knows" can be a bit of a mystery to the roboteer because the oscillators conflate to create systemic properties that may not have been obvious to (or amenable to a closed form solution developed by) the roboteer. If it's simple enough (e.g. builds charge from a solar panel into a capacitor that eventually discharges), the robot "knows" how to store energy slowly and expend it rapidly. The knowledge here is implicitly grounded directly to the sensors and actuators. Obviously, any sophisticated robot will be a hybrid of these two extremes. So, to answer your question as directly as possible, roboteers do NOT provide the robots with their "knowledge". The roboteers construct a machine, that's all. If there is any knowledge in a robot, it will remain a mystery to the roboteer until the robots are intelligent enough to report back what they know. [grin] -- glen e. p. ropella, 971-222-9095, http://agent-based-modeling.com ============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org
