Detection,localisation and tracking of pallets using machine learning techniques and 2D range data

The problem of autonomous transportation in industrial scenarios is receiving a renewed interest due to the way it can revolutionise internal logistics, especially in unstructured environments. This paper presents a novel architecture allowing a robot to detect, localise, and track (possibly multiple) pallets using machine learning techniques based on an on-board 2D laser rangefinder only. The architecture is composed of two main components: the first stage is a pallet detector employing a Faster Region-Based Convolutional Neural Network (Faster R-CNN) detector cascaded with a CNN-based classifier; the second stage is a Kalman filter for localising and tracking detected pallets, which we also use to defer commitment to a pallet detected in the first stage until sufficient confidence has been acquired via a sequential data acquisition process. For fine-tuning the CNNs, the architecture has been systematically evaluated using a real-world dataset containing 340 labelled 2D scans, which have been made freely available in an online repository. Detection performance has been assessed on the basis of the average accuracy over k-fold cross-validation, and it scored 99.58% in our tests. Concerning pallet localisation and tracking, experiments have been performed in a scenario where the robot is approaching the pallet to fork. Although data have been originally acquired by considering only one pallet as per specification of the use case we consider, artificial data have been generated as well to mimic the presence of multiple pallets in the robot workspace. Our experimental results confirm that the system is capable of identifying, localising and tracking pallets with a high success rate while being robust to false positives.

     

Self-consistent kinetic theory with nonlinear wave-particle resonances

We have developed, based on the oscillating-center transformation, a general theoretical approach for self-consistent plasma dynamics including, explicitly, effects of nonlinear(higherorder) wave-particle resonances. A specific example is then given for low-frequency responses of trapped particles in axisymmetric tokamaks. Possible applications to transport as well as nonlinear wave growth/damping are also briefly discussed.     

Triennial 1951: Post-War Reconstruction and “Divine Proportion”

This research centres on the conference dedicated to the theme “Divina Proporzione” organised on the occasion of the 1951 Milano Triennale. Participating in the meeting were some of the most important art historians, architects, engineers, mathematicians and artists of the day: from a very young James Ackerman to Pier Luigi Nervi, from Le Corbusier to Carlo Mollino, from Lucio Fontana to Ernesto Rogers. That historic moment, even given the plurality of voices and approaches, represented the synthesis of a reflection that was as urgent as it was rich in implications, since it brought the concept of “proportion” into the very heart of post-war reconstruction.     

A model for drop and bubble breakup frequency based on turbulence spectra

In this article, a new Eulerian model for breakup frequency of drops induced by inertial stress in homogeneous isotropic turbulence is developed for moderately viscous fluids, accounting for the finite response time of drops to deform. The dynamics of drop shape in a turbulent flow is described by a linear damped oscillator forced by the instantaneous turbulent fluctuations at the drop scale. The criterion for breakup is based on a maximum value of drop deformation, in contrast with the usual critical Weber criterion. The breakup frequency is then modeled as a function of the power spectrum of Weber number (or velocity square), based on the theory of oscillators forced by a random signal, which can be related to classical statistical quantities, such as dissipation rate and velocity variance. Moreover, the effect of viscosities of both phases is included in the breakup frequency model without resorting to any additional parameter. © 2018 American Institute of Chemical Engineers AIChE J, 65: 347–359, 2019     

Theoretical Hypothesis of a Double Barrier Regarding the D-D Interaction in a Pd Lattice： A Possible Explanation of Cold Fusion Experiment Failures

During the past 15 years, disputable experimental evidence has built up for LERN （low energy nuclear reaction） phenomena in specialized heavy hydrogen systems. Actually, it can not be said that a new branch of science is beginning. In spite of experimental contributions, the real problem is that the theoretical statements of LERN are unknown. In this work, the authors analyze the deuteron-deuteron reactions within palladium lattice by means of the coherence theory of nuclear and condensed matter and, using this general theoretical framework accepted from ＂cold fusion scientists＂, it will be shown the low occurrence probability of fusion phenomena. In fact, in the coherence approach, the D-D potential exhibits double barrier features and, in this way, the D-D fusion is hampered.     

Glycosylation,denaturation, and aggregation of soy proteins in defatted soy flakes flour: Influence of thermal and homogenization treatments

This article provides a systematic study of the impact of different thermal treatments (62 ± 2°C, without and with relative humidity control, 79%) on soy protein in defatted soy flour and their aqueous dispersions. The effect of dispersing treatments (magnetic stirring, high-speed, and high-pressure homogenization) on dispersions also was assessed. Changes in protein solubility (water and 0.2 g/100 g potassium hydroxide solution), apparent-reactive lysine content, urease and trypsin inhibitor activities, protein denaturation, and Fourier transform infrared spectra were studied. Glycosylation, aggregation, and denaturation of storage and biologically active soy proteins were observed in different degrees, being mainly promoted by the control of relative humidity and the dispersibility of the sample.     

Numerical and experimental characterization of antenna positioning in a dual-radio mesh router

This paper investigates the influence of the position of the antennas in a dual-radio mesh router by characterizing the router components from an electromagnetic point of view. Since the behavior of a wireless router equipped with multiple devices and antennas can be affected by mutual coupling among several close conducting bodies, numerical and experimental tests are carried out to provide some suggestions for an accurate positioning of the antennas. A reference configuration is implemented in an electromagnetic Computer-Aided Design (CAD) simulator. The most significant numerical results are validated by performing a campaign of measurements in anechoic chamber and deploying a testbed for the derivation of the experimental values of throughput, return loss, power density, and directivity.