Gerardo Beni

A validity measure for fuzzy clustering

The authors present a fuzzy validity criterion based on a validity function which identifies compact and separate fuzzy c-partitions without assumptions as to the number of substructures inherent in the data. This function depends on the data set, geometric distance measure, distance between cluster centroids and more importantly on the fuzzy partition generated by any fuzzy algorithm used. The function is mathematically justified via its relationship to a well-defined hard clustering validity function, the separation index for which the condition of uniqueness has already been established. The performance of this validity function compares favorably to that of several others. The application of this validity function to color image segmentation in a computer color vision system for recognition of IC wafer defects which are otherwise impossible to detect using gray-scale image processing is discussed. >

The concept of cellular robotic system

The author discusses the conceptual basis for the theory and engineering of a new type of robotic system. The system is composed of autonomous robotic units which accomplish tasks in cooperation. After describing the relevance of this system and contrasting it with cellular automata and neural networks, the author establishes the fundamental properties of the system and their consequences for the structure of the robotic units, the space on which they operate, and the algorithms by which they accomplish the global tasks. The significance of the concept of cellular robotic systems for distributed computing, molecular computing, self-organization, and reliability is explained.<<ETX>>

Theoretical problems for the realization of distributed robotic systems

The authors consider the specific problems most likely to be critical to progress in distributed robotic systems. This perspective suggests that the most promising approach to the design of practical algorithms for complex tasks might be to develop hybrid solutions comprising three stages: (1) use of interacting many-body techniques to form one dimensional substructures in continuous space; (2) application of cellular robot systems techniques to reconfigure the substructure with the desired qualities; and (3) application of standard control theory techniques for maintaining the configuration during subsequent operation.<<ETX>>

Distributed computing problems in cellular robotic systems

Cellular robotic systems (CRS) employ a large number of robots operating in cellular spaces under distributed control. In this paper, the relationship between CRS and distributed computing is discussed. Two problems encountered in designing pattern generation protocols for CRS, the n-way intersection problem and the knot detection problem, are related to distributed mutual exclusion problem and distributed deadlock detection problem, respectively. Solutions to these two problems, derived from their counterparts in distributed computing, are presented in the CRS context.<<ETX>>

Pattern generation in cellular robotic systems

A computational model of cellular robotic systems (CRSs) is introduced. It employs a large (but finite) number of autonomous robots operating on a cellular space under distributed control. No synchronous clock or shared memory is assumed. Robots in such systems have to cooperate to accomplish prespecified global tasks under the government of a protocol. The problem of robots in a CRS spatially rearranging themselves to form desired patterns within the field of operation is discussed. Example protocols that generate alternating and arbitrarily predefined patterns in one-dimensional linear arrangements and protocols that make robots seal one side or all sides of two-dimensional grids are presented. Mathematical proofs of the correctness of these protocols have been obtained. A Sun-Unix based simulation platform has been implemented upon which the pattern forming protocols are exercised.<<ETX>>

On the asymptotic stability of the hybrid position/force control scheme for robot manipulators

Conditions are presented for the global asymptotic stability of the hybrid position/force control of robot manipulators, using Lyapunovs direct method. Although hybrid control schemes have been studied in depth, the stability problem has not been clarified completely. The authors deal with the stability of the hybrid control scheme in Cartesian coordinates of the work space, especially from the viewpoint of coupling between position and force control. Results show not only the global asymptotic stability conditions but also the physical interpretation or meaning of the stability conditions by citing an example of a two-degree-of-freedom manipulator.<<ETX>>

A new fuzzy clustering validity criterion and its application to color image segmentation

A new strategy for the determination of the degree of validity of a fuzzy partition is introduced. The strategy is mathematically justified via its relation to hard partition validity measures. The relationship between this fuzzy validity function and the most general, and well defined, hard clustering validity function (the separation index of J.C. Dunn (1974)) for which the condition of uniqueness has already been established is derived. Using this relationship, the existence of a unique fuzzy c-partition produced by the fuzzy validity function is also proved. Examples of applications to segmentation of color image for IC defects give encouraging results.<<ETX>>

Color Vision for Microelectronics Inspection

The rapid and accurate inspection of microelectronic structures is necessary to increase the yield in semiconductor manufacturing. Machine inspection using black and white vision is increasingly being used to automate this function. Recently, color vision has become available as a tool for machine vision. In this paper we present the use of color vision for the measurement of oxide thickness on a silicon wafer. We have achieved a resolution of less than 30 Angstroms in thickness discrimination. The system can also be used to monitor uniformity of oxide thickness across the surface of a wafer. The sensitivity and stability of the method, along with the factors that affect the two, are examined in order to ensure long term stability and repeatability.

On the logical and physical combination of evidence in intelligent machines

We give a rigorous interpretation of the ways of combining evidence in Mycin‐type expert systems. We demonstrate the probabilistic inexactness of the Mycin model and its incompleteness for combining evidence from quantitative data extracted in well‐defined physical systems. Finally, we present a new method of combining evidence including both logical and physical combinations of evidence. the method is applicable to intelligent machines capable of gathering evidence via sensors.

Inference in intelligent machines: Application to a thermal evaporator

We present a new method of inference applicable to robots and other intelligent machines. Inferences drawn by intelligent machines are based on measurements gathered through sensory perception. We demonstrate that the methods for managing uncertainty of meaning, which recently have been extended to a wide variety of non-human systems, generally yield qualitatively incorrect results when applied to the uncertainty of evidence available to an intelligent machine. We show that even in very simple machines, no amount of sophistication in the mathematical algorithms can compensate for incorrect assumptions about the physical model. Conversely, we also demonstrate that once the essential structure of the physical model is correctly described, classical probability theory yields simple algorithms for the evaluation of the degree of evidence as it propagates through complex inference networks, including diagnostic trees and multicausal nets. As a first application, we have derived the probability algorithms relevant to diagnosing the malfunctioning of a thermal evaporator. For this system, an inference network has been constructed and compared to an implementation based on a MYCIN-type expert system. The laboratory implementation of the system is also described.