Let's face it, the real core of robotics is information processing. What is done with the information a robot gathers truly determines its behavior. At the simplest level, that information could be a series of commands and times to perform them. Even at that level, there is still sensing going on; in this case, we are sensing the ticks of a clock. While this is a great start, watching your robot roll off your desk or ram into a wall because it had no sense of its surroundings gets old really fast. Building a robot that responds to its environment is what people are on a perpetual quest for, but, to do that, we need perception. Typical sensors of the day include photo-resistive, photo-reflective, pyroelectric, optical triangulation, quadrature, and many others; however, if you make a list, you will find many optical sensors out there. Optical sensors are great because they are non-contact - they let you sense something without touching it. Of course, the king of optical sensors is the video camera. Wouldn't it be wonderful to attach a video camera to our robots and just see the world? The problem is, often times, we simply do not have the processing power to accommodate a video signal and, honestly, very often we do not need it.
展开▼
