This paper proposes a new method to model partially connected feedforward neural networks (PCFNNs) from the identified input type (IT) which refers to whether each input is coupled with or uncoupled from other inputs in generating output. The identification is done by analyzing input sensitivity changes as amplifying the magnitude of inputs. The sensitivity changes of the uncoupled inputs are not correlated with the variation on any other input, while those of the coupled inputs are correlated with the variation on any one of the coupled inputs. According to the identified ITs, a PCFNN can be structured. Each uncoupled input does not share the neurons in the hidden layer with other inputs in order to contribute to output in an independent manner, while the coupled inputs share the neurons with one another. After deriving the mathematical input sensitivity analysis for each IT, several experiments, as well as a real example (blood pressure (BP) estimation), are described to demonstrate how well our method works.

To classify systolic, mean and diastolic blood pressure using the oscillometric method heavily depends on the computational algorithms. Generally, the algorithms aim at extracting some parameters such as height, ratios of the pulses at certain pressure levels, which are obtained from the cuff pressure. These parameters can be used to form profiles to attribute to blood pressures. Our algorithms are based on fuzzy sets, whose membership functions are determined by using neural networks. We further employ Gram-Schmidt orthogonal transformation to select appropriate features for classification. The effectiveness of neural network solution to systolic blood pressure classification is the focus of this paper.

The authors investigate the local convergence properties of an artificial-neural-network (ANN)-based learning controller, using linearization techniques. The controller utilizes generic multilayer ANNs to adaptively approximate the manipulator dynamics over a specified region of the state space for a given desired trajectory. This generic neural network structure can be viewed as a nonlinear extension of a deterministic autoregressive model which is commonly used in model matching problems for linear systems.>

A method of adapting an entire rule-base to new environmental conditions is introduced. The method, based on curve fitting, is explained using the example of a rule-based mobile robot control. Specifically, it is shown that it is possible to transform a so-called generic rule-base, for operation on a firm surface with an acceleration of 6 ft/sec/sup 2/, to a working rule-base, for operation on a less firm surface with a lower allowable acceleration, using a family of fuzzy mappings.>

The controller for an intelligent mobile autonomous system (IMAS), equipped with vision and low-level sensors to cope with unknown obstacles, is modeled as a hierarchy of decision-making for planning and control. One of the levels (pilot) deals with a distorted 'windshield' view of the world and provides the actuator controller with real-time decisions. This level of IMAS controller is treated as a linguistic controller with fuzzy variables that assume values from possible intervals. The decision-making process at this level of control are presented as a production system with a fuzzy database. The rules in the production system are derived from an analytical system model for minimum-time control. The choice of optimal motion execution commands is performed using fuzzy set operators. Also included is a temporal decision-making mechanism (reporter), which recognizes the persisting conflicts between successive levels of the hierarchy by observing the motion trajectory.>

An Intelligent Mobile Autonomous System (IMAS), which is equipped with vision and low level sensors to cope with unknown obstacles, is modeled as a hierarchy of decision making for motion planning and control. The world description is based on linguistic variables that assume values from possible interval corresponds to the membership grade to a fuzzy set. The decision making is presented as a production system with a fuzzy database. The choice of optimal motion execution commands are performed using fuzzy set operators. Also included in the paper is the procedure of rule derivation and examples based on low level motion control.

This paper represents part of the ongoing research into an Intelligent Mobile Autonomous System (IMAS). An intermediate path planning subsystem (Navigator), for the hierarchy of IMAS, is presented which operates in completely known, partially known, and completely unknown environments. Information is passed from a sensor to the Navigator via the "Cartographer" which performs map updating. The interaction and performance of these subsystems are demonstrated in a simulation of the IMAS hierarchy.