[신간의 별자리x] 우리/미술, 그리고 ‘슬픔의 박물관’

Artificial neural networks (ANNs) constitute a class of flexible nonlinear models designed to mimic biological neural systems. In this entry, we introduce ANN using familiar econometric terminology and provide an overview of ANN modeling approach and its implementation methods. † Correspondence: Chung-Ming Kuan, Institute of Economics, Academia Sinica, 128 Academia Road, Sec. 2, Taipei 115, Taiwan; ckuan@econ.sinica.edu.tw. †† I would like to express my sincere gratitude to the editor, Professor Steven Durlauf, for his patience and constructive comments on early drafts of this entry. I also thank Shih-Hsun Hsu and Yu-Lieh Huang for very helpful suggestions. The remaining errors are all mine.

Artificial neural networks

By 2050, the world is expected to generate 340 billion tons of waste annually, increasing drastically from today’s 201 billion tons What a Waste 20: A Global Snapshot of Solid Waste Management to 2050 aggregates extensive solid waste data at the national and urban levels It estimates and projects waste generation to 2030 and 2050 Beyond the core data metrics from waste generation to disposal, the report provides information on waste management costs, revenues, and tariffs; special wastes; regulations; public communication; administrative and operational models; and the informal sector

What a Waste 2.0: A Global Snapshot of Solid Waste Management to 2050

Most of the signal processing that we will study in this course involves local operations on a signal, namely transforming the signal by applying linear combinations of values in the neighborhood of each sample point. You are familiar with such operations from Calculus, namely, taking derivatives and you are also familiar with this from optics namely blurring a signal. We will be looking at sampled signals only. Let's start with a few basic examples. Local difference Suppose we have a 1D image and we take the local difference of intensities, DI(x) = 1 2 (I(x + 1) − I(x − 1)) which give a discrete approximation to a partial derivative. (We compute this for each x in the image.) What is the effect of such a transformation? One key idea is that such a derivative would be useful for marking positions where the intensity changes. Such a change is called an edge. It is important to detect edges in images because they often mark locations at which object properties change. These can include changes in illumination along a surface due to a shadow boundary, or a material (pigment) change, or a change in depth as when one object ends and another begins. The computational problem of finding intensity edges in images is called edge detection. We could look for positions at which DI(x) has a large negative or positive value. Large positive values indicate an edge that goes from low to high intensity, and large negative values indicate an edge that goes from high to low intensity. Example Suppose the image consists of a single (slightly sloped) edge:

http://ieeexplore.ieee.org/iel5/5/31307/01456462.pdf

Linear systems

System Identification I

When the problem is considered of obtaining a periodic description in state-space form of a linear process which can be modelled by linear difference equations with periodic coefficients, it is natural to ask whether it is possible to preliminarily derive a polynomial equivalent form of such equations, which in the periodic case plays a role similar to the Rosenbrock's polynomial matrix description of a linear time-invariant process. In this paper a polynomial time-invariant description of a linear periodic process is introduced. It is shown that such a polynomial description gives a simple characterization of the dimension of the space of the solutions corresponding to the null input function, i.e., of the order of the periodic model under consideration. In addition, it allows us to introduce a transfer matrix for the computation of the output responses corresponding to null initial conditions, and to deduce conditions for the periodic model to be causal. These results, as well as the possibility of defining strict system equivalence between two periodic models through their time-invariant polynomial descriptions, in a similar sense as in the time-invariant case, show the relevance of such a polynomial time-invariant description for the problem under consideration.

A polynomial approach to deriving a state-space model of a periodic process described by difference equations

RMAS architecture for industrial agents in IEC 61499

Unmanned Aerial Vehicles (UAVs) and in particular quad-rotors are gaining an increasing interest owing to their flexibility and versatility. Today the challenge is to integrate these versatile platforms in a wide fleet in order to perform cooperative tasks as surveillance, search & rescue, inspection. The coalition formation problem is a pre-condition for cooperative missions. This problem can be solved with different methodologies from biologically-inspired algorithms to parasocial consensus sampling ones. In this paper a new framework for simulation of unmanned vehicles in cooperative scenarios is first presented. Then a novel method that exploits the benefits of Model Predictive Control (MPC) for a coalition formation problem (leader-follower) is introduced. The obtained results evidence the good performance of MPC to solve the problem of coalition formation for unmanned aerial vehicles finding the optimal solution taking into account different kind of constraints. The developed framework allows also to easily change from simulated agent to real one.

A simulation framework for coalition formation of Unmanned Aerial Vehicles

A characteristic of structured environments is that objects tend to lie in straight lines representing the vertical surfaces of walls, doors, objects, etc. This paper proposes feature-based modeling of the environment in which line segments or surfaces are used for modeling an indoor environment and for improving the pose estimation of a mobile robot. Line features are detected making use of occupancy grids, based on ultrasonic range readings, and of a digital camera. A multisensor fusion approach for fusing ultrasonic and video data is proposed. The Hough transform (HT) is used for line feature extraction from both an occupancy grid map and digital images. Using a Bayesian approach to the HT, a rigorous probability value of the main straight line features extracted from a digital image is also proposed.

http://ieeexplore.ieee.org/iel5/8325/26388/01176066.pdf

A Bayesian approach to the Hough transform for video and ultrasonic data fusion in mobile robot navigation

In this paper a wireless sensors network (WSN) has been designed according to the IEEE 1451 standard and installed in a manufacturing cell. On-line monitoring data have been collected and stored in a database through a Web2Py framework. A case study on the usage of WSN to perform an industrial performance analysis is made. A set of measures, comprehensive of all the production parameters of 4 different line productions, have been used to analyze the performances of the company. Effectiveness and reliability indexes, such as the mean time to repair (MTTR) and mean time between failures (MTBF) of the machines, overall equipment effectiveness (OEE) and the total effective equipment productivity (TEEP) have been computed and compared to the world-class manufacturing (WCM) standards.

Sauro Longhi

Papers

A polynomial approach to deriving a state-space model of a periodic process described by difference equations

RMAS architecture for industrial agents in IEC 61499

A simulation framework for coalition formation of Unmanned Aerial Vehicles

A Bayesian approach to the Hough transform for video and ultrasonic data fusion in mobile robot navigation

Application of a wireless sensor networks and Web2Py architecture for factory line production monitoring