Is it possible to use intelligent techniques in real-time systems?
AI become mature., there has been growing interest in applying these techniques to controlling complex real-world systems.
In many applications, a computer control system must sense the environment and directly influence it through action.
# intelligent techniques embedded in a mobile application for real-time.
It depends on your fitness function and required
If you use AND colony B colony in every time it searches about the best solution, while other algorithms like Fuzzy and neural networks have moods they are off-line mood (training) this is not in the time of taking decision, but when u use it (off-line mode) it is a very fast comparing to the ordinary method.
The only problem that you may face when using intelligent (Or evolutionary) algorithms such as genetic algorithm is high convergence speed especially when you deal with a problems that consists of a large number of variables. In such a case, you need to do your calculations off-line to find the optimum values then use them in real time system.
Good Luck
Yes. I discussed this in "Variable Depth Resolution for Intelligent Real-Time Systems" delivered at The 8th International Conference on Software Engineering and Knowledge Engineering (SEKE '96). The basic idea is to have an algorithmic technique that yields an acceptable answer in a known amount of time, and an "intelligent" solution that takes a variable amount of time. The OK algorithm is executed first, in parallel, or as late as possible, while the variable-duration "intelligent" algorithm tries to come up with its better solution. If the "intelligent" solution is generated in time, use it, otherwise use the algorithmic solution.
If the "intelligent" solution refines itself, the quick and dirty algorithm could be ignored, aborted, or not run, once the first "intelligent" solution version is created.