A Simple Soft Computing Structure for Modeling and Control

Using the interpolation/extrapolation skills of core function an iterative adaptive controller, a structurally simple single essential layer neural network-based topological structure is suggested with fast and explicit single-step teaching data-retrieving abilities. Its operation does not assume massive parallelism, therefore it easily can be simulated by sequential program codes needing sophisticated data synchronization mechanisms. It seems to advantageous in approximate model-based common, robust, or controllers that compensate for effects minor modeling imprecisions. In this neuron either firing passive (i.e., producing zero output) state. state its activation realizes abstract rotation maps desired kinematic into space necessary control forces. The allows use incremental model modification slowly varying dynamic models. exemplified numerical simulations van der Pol oscillator free motion, within Computed Torque type control. To reveal possibility efficient correction, robust Variable Structure/Sliding Mode Controller applied, too. novel obtained experimental observations as e.g., fuzzy