Machine learning and small robot navigation

Developing successful navigation and mapping strategies is an essential part of autonomous robot research. However, hardware limitations often make for inaccurate systems. This project serves to investigate efficient alternatives to mapping an environment, by first creating a mobile robot, and then applying machine learning to the robot and controlling systems to increase the robustness of the robot system. My mapping system consists of a semi-autonomous robot drone in communication with a stationary Linux computer system. There are learning systems running on both the robot and the more powerful Linux system. The first stage of this project was devoted to designing and building an inexpensive robot. Utilizing my prior experience from independent studies in robotics, I designed a small mobile robot that was well suited for simple navigation and mapping research. When the major components of the robot base were designed, I began to implement my design. This involved physically constructing the base of the robot, as well as researching and acquiring components such as sensors. Implementing the more complex sensors became a time-consuming task, involving much research and assistance from a variety of sources. A concurrent stage of the project involved researching and experimenting with different types of machine learning systems. I finally settled on using neural networks as the machine learning system to incorporate into my project. Neural nets can be thought of as a structure of interconnected nodes, through which information filters. The type of neural net that I chose to use is a type that requires a known set of data that serves to train the net to produce the desired output. Neural nets are particularly well suited for use with robotic systems as they can handle cases that lie at the extreme edges of the training set, such as may be produced by "noisy" sensor data. Through experimenting with available neural net code, I became familiar with the code and its function, and modified it to be more generic and reusable for multiple applications of neural nets.

KittyCat: a cognitive model of structure-form discovery

Cognition is a core subject to understand how humans think and behave. In that sense, it is clear that Cognition is a great ally to Management, as the later deals with people and is very interested in how they behave, think, and make decisions. However, even though Cognition shows great promise as a field, there are still many topics to be explored and learned in this fairly new area. Kemp & Tenembaum (2008) tried to a model graph-structure problem in which, given a dataset, the best underlying structure and form would emerge from said dataset by using bayesian probabilistic inferences. This work is very interesting because it addresses a key cognition problem: learning. According to the authors, analogous insights and discoveries, understanding the relationships of elements and how they are organized, play a very important part in cognitive development. That is, this are very basic phenomena that allow learning. Human beings minds do not function as computer that uses bayesian probabilistic inferences. People seem to think differently. Thus, we present a cognitively inspired method, KittyCat, based on FARG computer models (like Copycat and Numbo), to solve the proposed problem of discovery the underlying structural-form of a dataset.

Hospital humano: etnografia da humanização hospitalar na perspectiva de usuários

This ethnographic work studies the experiences of patients admitted in public (PUH) and private (PRH) hospitals in the Brazilian northeastern region. 28 adult patients of different clinics participated in the study. Data were analyzed by the patient path method, consisting in a combination of complemented and articulated techniques free observation, participating observation, ethnographic interview and patient testimonials collected prospectively during the patients admissions, from their arrival and until their discharge. The analysis was carried out according to the Thematic Categories Analysis Technique and the data were interpreted pursuant to medical anthropology, healthcare humanization and healthcare promotion theoretical references. The ethical principles of Resolution 196/96 were followed. The human hospital, as revealed by the patient, highlights the significance of subjectivity. 225 (54.7%) out of 411 mentioned concepts were collected in a public hospital (PUH) and 186 (45.3%) in a private institution (PRH). The results show that the patient at the PUH and PRH ethnoevaluates different aspects of the healthcare professionals´ human and technical competence, the hospital´s functioning structure, the access to and the ethics in the financial management, and develops overcoming strategies for his stay at the hospital. This ethnoevaluation is mediated by different factors, namely: social and economic status, personality, religiosity, ironic speech, somber diagnosis and satisfied needs, prior hospital experiences and the conditions under which the interview was carried out. A pedagogic proposal for the hospital humanization must include structural, managerial and organizational changes of the offered services and use active methodologies aimed to the political resolution of problematic situations at work and the inclusion of affective and subjective factors, and become as well a tool for the collective learning. This study shows the importance for the user´s ethnoevaluation to be incorporated into the hospital management and care as a guideline in the decision making and clinical action, thus promoting practices that shall lead to a decent and humanized care. The multidisciplinary nature of this study allowed a wide understanding of the user´s perspective as a socially critical ethnoevaluator

LEARNING A COMPLEX MOTOR SKILL FROM VIDEO and POINT-LIGHT DEMONSTRATIONS

The aim of this Study was to compare the learning process of a highly complex ballet skill following demonstrations of point light and video models 16 participants divided into point light and video groups (ns = 8) performed 160 trials of a pirouette equally distributed in blocks of 20 trials alternating periods of demonstration and practice with a retention test a day later Measures of head and trunk oscillation coordination d1 parity from the model and movement time difference showed similarities between video and point light groups ballet experts evaluations indicated superiority of performance in the video over the point light group Results are discussed in terms of the task requirements of dissociation between head and trunk rotations focusing on the hypothesis of sufficiency and higher relevance of information contained in biological motion models applied to learning of complex motor skills

On negative flows of the AKNS hierarchy and a class of deformations of a bihamiltonian structure of hydrodynamic type

A deformation parameter of a bihamiltonian structure of hydrodynamic type is shown to parametrize different extensions of the AKNS hierarchy to include negative flows. This construction establishes a purely algebraic link between, on the one hand, two realizations of the first negative flow of the AKNS model and, on the other, two-component generalizations of Camassa-Holmand Dym-type equations. The two-component generalizations of Camassa-Holm- and Dym-type equations can be obtained from the negative-order Hamiltonians constructed from the Lenard relations recursively applied on the Casimir of the first Poisson bracket of hydrodynamic type. The positive-order Hamiltonians, which follow froth the Lenard scheme applied on the Casimir of the second Poisson bracket of hydrodynamic type, are shown to coincide with the Hamiltonians of the AKNS model. The AKNS Hamiltonians give rise to charges conserved with respect to equations of motion of two-component Camassa-Holm- and two-component Dym-type equations.

Short comments on self-synchronization of two non-ideal sources supported by a flexible portal frame structure

A practical problem of synchronization of a non-ideal (i.e. when the excitation is influenced by the response of the system) and non-linear vibrating system was posed and investigated by means of numerical simulations. Two rotating unbalanced motors compose the mathematical model considered here with limited power supply mounted on the horizontal beam of a simple portal frame. As a starting point, the problem is reduced to a four-degrees-of-freedom model and its equations of motion, derived elsewhere via a Lagrangian approach, are presented. The numerical results show the expected phenomena associated with the passage through resonance with limited power. Further, for a two-to-one relationship between the frequencies associated with the first symmetric mode and the sway mode, by using the variation of torque constants, the control of the self-synchronization and synchronization (in the system) are observed at certain levels of excitations.

On energy pumping, synchronization and beat phenomenon in a nonideal structure coupled to an essentially nonlinear oscillator

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Glass structure and ion dynamics of lead-cadmium fluorgermanate glasses

Glass structure and fluorine motion dynamics are investigated in lead-cadmium fluorgermanate glasses by means of differential scanning calorimetry, Raman scattering, x-ray absorption (EXAFS), electrical conductivity (EC), and F-19 nuclear magnetic resonance (NMR) techniques. Glasses with composition 60PbGeO(3)-xPbF(2)-yCdF(2) (in mol %), with x+y=40 and x=10, 20, 30, 40, are studied. Addition of metal fluorides to the base PbGeO3 glass leads to a decrease of the glass transition temperature (T-g) and to an enhancement of the ionic conductivity properties. Raman and EXAFS data analysis suggest that metagermanate chains form the basic structural feature of these glasses. The NMR study leads to the conclusion that the F-F distances are similar to those found in pure crystalline phases. Experimental results suggest the existence of a heterogeneous glass structure at the molecular scale, which can be described by fluorine rich regions permeating the metagermanate chains. The temperature dependence of the NMR line shapes and relaxation times exhibits the qualitative and quantitative features associated with the high fluorine mobility in these systems. (C) 2004 American Institute of Physics.

Variable structure controller and observer applied to induction machine

This work will propose the control of an induction machine in field coordinates with imposed stator current based on theory of variable structure control and sliding mode. We describe the model of an induction machine in field coordinates with imposed stator current and we show the design of variable structure control and sliding mode to get a desirable dynamic performance of that plant. To estimate the inaccessible states we will use a state observer (estimator) based on field coordinates induction machine. We will present the results of simulations in any operation condition (start, speed reversal and load) and with parameters variation of the machine compared to a PI control scheme.

Human development and distance education: how are interactivity and interaction articulated?

Currently the achievements of technology have not been enough to overcome the misery, in which are numerous human groups. Relegated to its background, the human being sees those scenarios, which brings out discussions about its formation, in which values are regarded as the scientific and technological aspects. The educational research inspired by this framework involving the theme of human development, update and re-frame concepts related to the linkage between interactivity and interaction, two important features presented in the process of Distance Education (DE). The research inquired about how these features have been articulated. It conducted a field study in which two professors were interviewed. The results showed that the integration between interactivity and interaction, involving aspects such as autonomy, critical awareness, relationships among students, the sharing of values and worldviews, is at the base of the educational processes of the DE. They also showed that, on these processes, there are shortcomings regarding the development of values and having to be thought the tutor training strategies from an interdisciplinary view.

A nonlinear and non-ideal wind generator supporting structure

We present a simple mathematical model of a wind turbine supporting tower. Here, the wind excitation is considered to be a non-ideal power source. In such a consideration, there is interaction between the energy supply and the motion of the supporting structure. If power is not enough, the rotation of the generator may get stuck at a resonance frequency of the structure. This is a manifestation of the so-called Sommerfeld Effect. In this model, at first, only two degrees of freedom are considered, the horizontal motion of the upper tip of the tower, in the transverse direction to the wind, and the generator rotation. Next, we add another degree of freedom, the motion of a free rolling mass inside a chamber. Its impact with the walls of the chamber provides control of both the amplitude of the tower vibration and the width of the band of frequencies in which the Sommerfeld effect occur. Some numerical simulations are performed using the equations of motion of the models obtained via a Lagrangian approach.

Damage detection in a benchmark structure using AR-ARX models and statistical pattern recognition

Structural health monitoring (SHM) is related to the ability of monitoring the state and deciding the level of damage or deterioration within aerospace, civil and mechanical systems. In this sense, this paper deals with the application of a two-step auto-regressive and auto-regressive with exogenous inputs (AR-ARX) model for linear prediction of damage diagnosis in structural systems. This damage detection algorithm is based on the. monitoring of residual error as damage-sensitive indexes, obtained through vibration response measurements. In complex structures there are. many positions under observation and a large amount of data to be handed, making difficult the visualization of the signals. This paper also investigates data compression by using principal component analysis. In order to establish a threshold value, a fuzzy c-means clustering is taken to quantify the damage-sensitive index in an unsupervised learning mode. Tests are made in a benchmark problem, as proposed by IASC-ASCE with different damage patterns. The diagnosis that was obtained showed high correlation with the actual integrity state of the structure. Copyright © 2007 by ABCM.

A structure excited by a DC motor of limited power supply and controlled by a pendulum like device

In this paper, a mathematical model is derived via Lagrange's Equation for a shear building structure that acts as a foundation of a non-ideal direct current electric motor, controlled by a mass loose inside a circular carving. Non-ideal sources of vibrations of structures are those whose characteristics are coupled to the motion of the structure, not being a function of time only as in the ideal case. Thus, in this case, an additional equation of motion is written, related to the motor rotation, coupled to the equation describing the horizontal motion of the shear building. This kind of problem can lead to the so-called Sommerfeld effect: steady state frequencies of the motor will usually increase as more power (voltage) is given to it in a step-by-step fashion. When a resonance condition with the structure is reached, the better part of this energy is consumed to generate large amplitude vibrations of the foundation without sensible change of the motor frequency as before. If additional increase steps in voltage are made, one may reach a situation where the rotor will jump to higher rotation regimes, no steady states being stable in between. As a device of passive control of both large amplitude vibrations and the Sommerfeld effect, a scheme is proposed using a point mass free to bounce back and forth inside a circular carving in the suspended mass of the structure. Numerical simulations of the model are also presented Copyright © 2007 by ASME.

Achieving educational quality: what schools teach us: learning from Chile's P900 primary schools

Includes bibliography

Teaching and learning data structures supported by computers: An experiment using CADILAG tool

This paper aims to present the use of a learning object (CADILAG), developed to facilitate understanding data structure operations by using visual presentations and animations. The CADILAG allows visualizing the behavior of algorithms usually discussed during Computer Science and Information System courses. For each data structure it is possible visualizing its content and its operation dynamically. Its use was evaluated an the results are presented. © 2012 AISTI.