Massimiliano Pirani

Advances in Lee--Schetzen Method for Volterra Filter Identification

This paper concerns the identification of nonlinear discrete causal systems that can be approximated with the Wiener–Volterra series. Some advances in the efficient use of Lee–Schetzen (L–S) method are presented, which make practical the estimate of long memory and high order models. Major problems in L–S method occur in the identification of diagonal kernel elements. Two approaches have been considered: approximation of gridded data, with interpolation or smoothing, and improved techniques for diagonal elements estimation. A comparison of diagonal elements estimated, with different methods has been shown with extended tests on fifth order Volterra systems.

Representation of Nonlinear Random Transformations by Non-Gaussian Stochastic Neural Networks

The learning capability of neural networks is equivalent to modeling physical events that occur in the real environment. Several early works have demonstrated that neural networks belonging to some classes are universal approximators of input-output deterministic functions. Recent works extend the ability of neural networks in approximating random functions using a class of networks named stochastic neural networks (SNN). In the language of system theory, the approximation of both deterministic and stochastic functions falls within the identification of nonlinear no-memory systems. However, all the results presented so far are restricted to the case of Gaussian stochastic processes (SPs) only, or to linear transformations that guarantee this property. This paper aims at investigating the ability of stochastic neural networks to approximate nonlinear input-output random transformations, thus widening the range of applicability of these networks to nonlinear systems with memory. In particular, this study shows that networks belonging to a class named non-Gaussian stochastic approximate identity neural networks (SAINNs) are capable of approximating the solutions of large classes of nonlinear random ordinary differential transformations. The effectiveness of this approach is demonstrated and discussed by some application examples.

Diagonal Kernel point estimation of nth-order discrete Volterra-Wiener systems

The estimation of diagonal elements of a Wiener model kernel is a well-known problem. The new operators and notations proposed here aim at the implementation of efficient and accurate nonparametric algorithms for the identification of diagonal points. The formulas presented here allow a direct implementation of Wiener kernel identification up to theth order. Their efficiency is demonstrated by simulations conducted on discrete Volterra systems up to fifth order.

A scalable production efficiency tool for the robotic cloud in the fractal factory

The present paper proposes an effective metrics for production efficiency and a bottleneck detection algorithm of recursive nature for its application on lightweight embedded systems on board of the robotics and automation components of the factory of the future. The proposed methodology is particularly suited if the fractal paradigm is applied to the factory seen as a complex system of systems but with relevant self-similarities across the several layers of components and structures from the shop-floor up to the enterprise level. A performance test has been conducted to demonstrate the viability of the technology for tiny embedded devices with the use of declarative embedded database language. Due to the high scalability of the algorithm and its simplicity, it seems suitable also for the robotic cloud paradigm, where constituent mechatronics, sensors and actuators components are provided as a service. The results provided suggests that, with the use of similar recursive and distributed form of computing, production bottlenecks or fault detection can be scaled to address the complex and pervasive cyber-physical systems problems that characterize the 4th industrial revolution strategies.

Practical notes on two Volterra filter identification direct methods

A survey of direct methods for identification of discrete Volterra systems is given. The Lee-Schetzen method and Korenbergs fast orthogonal algorithm (FOA) are reviewed. These methods are still useful and can be used as reference for other methods. We give practical considerations and suggestions for optimal use of these two methods coming up from the study of statistical simulations up to the third order.

Rapid prototyping of open source ordinary differential equations solver in distributed embedded control application

This paper proposes a rapid prototyping framework based on open source tools for real-time solution of Ordinary Differential Equations (ODE) problems on embedded microcontrollers. A plug-in software for the deployment of ODE solvers, in C language, on embedded industrial environment is proposed. By means of C portability across different microcontrollers architectures, there is the possibility to give a considerable boost in rapid prototyping of mechatronics and complex embedded system applications. Some commercial softwares feature toolsets for the autogeneration of open C code, but they do not provide the translation of the ODE solvers routines for embedded systems. A key for open source software integration has been the adoption of an infrastructure based on a Database Management System (DBMS) on board of the embedded system.

A Stochastic Model of Neural Computing

This paper addresses the problem of neural computing by a fundamentally different approach to the one currently adopted in digital computers. The approach is based on the experience, rather than on the specification of operators as it is done in the conventional mathematical approach and it is well suited for implementation by neural networks.

Information Management and Decision Making Supported by an Intelligence System in Kitchen Fronts Control Process

Measuring systems used for alternative control are the basis of the kitchen fronts acceptance. The assessment of such products is difficult. On the basis of the performed analysis, it can be said that different operators make different decisions that result in different problems in a case study manufacturing company. In the work, the authors propose a cyber-physical distributed computing infrastructure to implement performance metrics methods extended to manage the control system and to support the decisions concerning improvements in the control process. It will result in the minimization or avoidance of operators’ mistakes concerning the wrong acceptance or rejection of the finished product.

A Database-Centric Framework for the Modeling, Simulation, and Control of Cyber-Physical Systems in the Factory of the Future

Abstract The factory of the future scenario asks for new approaches to cope with the incoming challenges and complexity of cyber-physical systems. The role of database management systems is becoming central for control and automation technology in this new industrial scenario. This article proposes database-centric technology and architecture that aims to seamlessly integrate networking, artificial intelligence, and real-time control issues into a unified model of computing. The proposed methodology is also viable for the development of a framework that features simulation and rapid prototyping tools for smart and advanced industrial automation. The full expression of the potentialities in the presented approach is expected in particular for applications where tiny and distributed embedded devices collaborate to a shared computing task of relevant complexity.

Self-similar Computing Structures for CPSs: A Case Study on POTS Service Process

This paper proposes a novel method for the structuring of the knowledge of a service process in order to be processed by lightweight declarative computing infrastructures. Through the identification of self-similarities in the process, the flow of the structured information and the sequence of activities performed in the process are easily implemented by means of cyber-physical systems technologies, in order to timely meet the customer/stakeholder’s requirements. The study was performed in a telecommunication service providing organization. Service teams create a collaborative network. With the use of the CPS proposed in this work they can communicate problems and disseminate solutions. This methodology uses the information of a set of performance indicators of the service organization to achieve a better control of the effectiveness and the bottlenecks in the supply network. The methodology is borrowed from the mechatronics field and it is prone to a natural extension and reuse for the similar information structures in manufacturing processes.