Selection of network and learning parameters for an adaptive neural robotic control scheme

This paper presents the results of a study in the design of a neural network based adaptive robotic control scheme. The neural network used here is a two hidden layer feedforward network and the learning scheme is the well-known backpropagation algorithm. The neural network essentially provides the inverse of the plant and acts in conjunction with a standard PD controller in the feedback loop. The objective of the controller is to accurately control the end position of a single link manipulator in the presence of large payload variations, variations in the link length and also variations in the damping constant. Based on results of this study, guidelines are presented in selecting the number of neurons in the hidden layers and also the parameters for the learning scheme used for training the network. Results also indicate that increasing the number of neurons in the hidden layer will improve the convergence speed of learning scheme up to a certain limit beyond which the addition of neurons will cause oscillations and instability. Guidelines for selecting the proper learning rate, momentum and fast backpropagation constant that ensure stability and convergence are presented. Also, a relationship between the r.m.s. error and the number of iterations used in training the neural network is established.     

Bethe-Salpeter equation for non-self conjugate mesons in a power-law potential

We develop an approach to the solution of the spinless Bethe-Salpeter equation for the different-mass case. Although the calculations are developed for spinzero particles in any arbitratry spherically symmetric potential, the non-Coulombic effective power-law potential is used as a kernel to produce the spin-averaged bound states of the non-self-conjugate mesons. The analytical formulas are also applicable to the self-conjugate mesons in the equal-mass case. The flavor-independent case is investigated in this work. The calculations are carried out to the third-order correction of the energy series. Results are consistent with those obtained before.     

Technology combats disabilities

The author focusses on advances in equipment that offset physically functional limitations. He discusses how disabled and elderly people are using aids that enhance mobility (walking a certain distance, lifting a weight, or climbing stairs) as well as sensory faculties (seeing print, hearing conversation, and speaking intelligibly)     

Reliability studies on GaAs MESFETs fabricated using spin-onplatinum source

GaAs MESFETs were fabricated using a spin-on platinum source as the gate material. They were subsequently aged at 200°C for up to 1000 h. DC electrical characterisation of the MESFETs was carried out during various stages of annealing. The aging behaviour of these MESFETs was compared with those fabricated using conventionally evaporated platinum sources. The results show that the performance of the MESFETs fabricated using a spin-on platinum source is comparable to those of MESFETs fabricated using a conventionally evaporated platinum source     

On the time-domain scattering by a passive classical frequency dependent wedge-shaped region in a lossy dispersive medium

The time-scattering of an electromagnetic wave by a wedge-shaped region in a lossy medium, both with frequency dependent electric characteristics, is analyzed for plane wave and line source illuminations. New exact analytical expressions, satisfying explicitely the causality and allowing useful physical decompositions of the field, are obtained in these cases, using the particularities of the harmonic response in Sommerfeld-Maliuzhinets integral, especially original properties of the spectral function attached to it.     

Fractal error for detecting man-made features in aerial images

A technique is proposed for aiding photointerpreters in detecting man-made features in aerial reconnaissance images. The technique, which uses a metric called fractal error, is based on the observed propensity of natural image features to fit a fractional Brownian motion model. Man-made features usually do not fit this model well, and consequently the fractal error metric may be used as a discriminant function for detecting man-made scene features     

10 Gbit/s optical soliton transmission over 7200 km by using amonolithically integrated MQW-DFB-LD/MQW-EA modulator light source

The authors demonstrate 10 Gbit/s optical soliton transmission in a recirculating loop through the use of a monolithically integrated MQW-DFB-LD/MQW-EA modulator light source. The timing jitter due to the Gordon-Haus effect is successfully reduced by the optical bandpass filters conventionally used in the cascaded EDFAs to avoid the accumulation of amplified spontaneous emission. The transmission distance of 7200 km is achieved at a bit error rate of 10^-9     

Electrodialysis of Hydrosols of Ultradispersed Diamonds

The electrical conductivity of hydrosols of ultradispersed diamonds was studied. The electromembrane method for cleaning and concentrating of ultradispersed diamonds in hydrosols was considered. The influence exerted by the concentration of the dispersed phase on the electrical characteristics of the cleaning process was analyzed. A mathematical relation making it possible to evaluate the output characteristics of the process and to determine the geometrical characteristics of the apparatus was proposed.