Abstract :
The author shows how elastic fuzzy logic (EFL) nets make it possible to combine the capabilities of expert systems with the learning capabilities of neural networks at a high level. ANN (artificial neural network) implementations have advantages in terms of hardware implementation, ease of use, generality, and links to the brain, which is still the only true intelligent controller available. Neurocontrol is useful in cloning experts, tracking trajectories or setpoints, and optimization (e.g., approximate dynamic programming). There has been substantial success in controlling robot arms (including the main arm of the Space Shuttle), chemical process control, continuous production of high-quality parts, and other aerospace applications. A review of the basic designs and concepts, with reference to both the applications and future research opportunities, is given
Keywords :
expert systems; fuzzy control; fuzzy logic; learning (artificial intelligence); neural nets; aerospace applications; approximate dynamic programming; artificial neural network; chemical process control; continuous parts production; elastic fuzzy logic; expert systems; experts cloning; hardware implementation; learning capabilities; robot arms control; trajectory tracking; Artificial intelligence; Artificial neural networks; Biological neural networks; Cloning; Expert systems; Fuzzy logic; Intelligent networks; Intelligent robots; Neural network hardware; Trajectory;
