Professor Lotfi Zadeh first introduced the concept of fuzzy logic in 1965, as reported by Pragya Agarwal of the Center for Spatially Integrated Social Science at the University of California. Fuzzy logic takes incomplete or conflicting, or seemly impossible inputs and makes a decision based upon imperfect data. For example, you buy groceries and the total comes to $100 but you only have $75. You start making decisions on which foods to forego based upon your wants and needs. You decide to forego a steak, but you keep the hot dogs. For you, this decision is easy. However (as of 2011), it is impossible for a computer to make this decision without fuzzy logic programming.
Robotics is the chief area where fuzzy logic inputs have to result in a clear output. For example, the Mars rover has to cross a plain. The shortest route is a straight line. However, a boulder is in its way. What route does the internal computer take? A command signal from Earth takes a long time to reach Mars, so the rover has to think on its own. Professor Michael G. Murphy wrote an extensive report to NASA about how a fuzzy logic control system could be used for obstacle avoidance in a Mars exploration robot. In 2011, the problems of fuzzy logic and robotics control for space exploration is in its infancy, but research is ongoing.
The chief difficulty is not with the output, but rather with the inputs. The inputs are not trying to intentionally "lie," but rather to change as the environment changes. Engineers and programmers try to develop control systems based upon ever-changing input. Suppose an airplane has to fly exactly 200 feet above the ground, but the area is full of hills and valleys. The altimeter is constantly changing as the airplane is flying. A fuzzy logic project would take the ever-changing inputs and follow the terrain accordingly.
Engineer Edward Sanzonov wrote a fuzzy logic control game for Clarkson University. In the game, you have to use a crane to manually load a box into a ship. It's a lot harder than some people think. You cannot let the load or the crane cause the platform to sway due to inertia. In real life, crane operators learn how to control sway, a skill developed by years of practice. When the "Fuzzy Logic" control button is pushed, the crane is computer controlled. The computer makes small adjustments to speed and direction, so the load is transferred smoothly. The next step is to install Sanzonov's program on a real crane, to see if fuzzy logic works in a real life situation.