Neural network-based flight control systems: Present and future

As the first review in this field, this paper presents an in-depth mathematical view of Intelligent Flight Control Systems (IFCSs), particularly those based on artificial neural networks. The rapid evolution of IFCSs in the last two decades in both the methodological and technical aspects necessitates a comprehensive view of them to better demonstrate the current stage and the crucial remaining steps towards developing a truly intelligent flight management unit. To this end, in this paper, we will provide a detailed mathematical view of Neural Network (NN)-based flight control systems and the challenging problems that still remain. The paper will cover both the model-based and model-free IFCSs. The model-based methods consist of the basic feedback error learning scheme, the pseudocontrol strategy, and the neural backstepping method. Besides, different approaches to analyze the closed-loop stability in IFCSs, their requirements, and their limitations will be discussed in detail. Various supplementary features, which can be integrated with a basic IFCS such as the fault-tolerance capability, the consideration of system constraints, and the combination of NNs with other robust and adaptive elements like disturbance observers, would be covered, as well. On the other hand, concerning model-free flight controllers, both the indirect and direct adaptive control systems including indirect adaptive control using NN-based system identification, the approximate dynamic programming using NN, and the reinforcement learning-based adaptive optimal control will be carefully addressed. Finally, by demonstrating a well-organized view of the current stage in the development of IFCSs, the challenging issues, which are critical to be addressed in the future, are thoroughly identified. As a result, this paper can be considered as a comprehensive road map for all researchers interested in the design and development of intelligent control systems, particularly in the field of aerospace applications.     

Structural modifications and fibre processing of hydroxy-functionalised mesogenic polyazomethines

A number of semiflexible homopolyazomethines and copolyazomethines based on hydroxy-functionalised mesogenic cores have been synthesised and characterised. The reported polymers include structural modifications such as an alteration of the coaxiality, shortening of the flexible spacers or copolymerisation to attain materials with reduced melting temperatures and suitable processability. These structural alterations have been introduced by condensing different ,ω-bis-[(4-formyl-3-hydroxyphenyl)oxy]alkanes with 2-methyl-1,4-phenylenediamine and/or 4-methyl-1,3-phenylenediamine. Fibres have been melt extruded from those nematic polyazomethines with the most favourable thermal properties. As-spun and tension-annealed fibres have been investigated by thermal analysis (TGA and DSC), X-ray diffraction and SEM microscopy in an attempt to infer a relationship between microstructure and tensile properties. From the results presented here, copolymerisation incorporating flexible spacers of different lengths seems to be the best strategy to balance ease of processing and tensile properties. Some of the fibres have improved mechanical properties compared with those previously reported for this class of semiflexible polyazomethine. A nematic polyazomethine with a decamethylenic spacer has also been modified with low percentages of several metal ions [Fe(III), Zn(II), V(IV) and Ni(II)] in order to establish a comparison with previously reported Cu(II)-modified fibres.     

Effects of solvent and temperature on the coupling agent/epoxy resin interface

Increased wetting of the coupling agent/epoxy resin interface was observed when γ-glycidoxypropyltrimethoxysilane, polyfunctional aminosilane and γ-aminopropyltriethoxysilane were applied respectively from methyllethylketone, dimethylformamide and water on woven glass cloths which had been cleaned at 300°C. However, when factory-applied coupling agents were burnt off the woven cloths and fresh coupling agents re-applied, it was found that the nature of the factory-applied coupling agent influenced subsequent wetting. Thinner glass fibres showed a greater improvement in wetting rate than thicker fibres in those solvents identified to be good for improved wettability, irrespective of the heat-cleaning temperature.     

Learning to Perceive and Act by Trial and Error

This article considers adaptive control architectures that integrate active sensory-motor systems with decision systems based on reinforcement learning. One unavoidable consequence of active perception is that the agent's internal representation often confounds external world states. We call this phoenomenon perceptual aliasingand show that it destabilizes existing reinforcement learning algorithms with respect to the optimal decision policy. We then describe a new decision system that overcomes these difficulties for a restricted class of decision problems. The system incorporates a perceptual subcycle within the overall decision cycle and uses a modified learning algorithm to suppress the effects of perceptual aliasing. The result is a control architecture that learns not only how to solve a task but also where to focus its visual attention in order to collect necessary sensory information.     

Biodegradable hollow microfibres to produce bioactive scaffolds

Biodegradable hollow microfibres containing particles loaded with specific active agents can be potentially employed to produce a special kind of substrate for tissue engineering, able to function as a scaffold and at the same time to act as a drug‐releasing system. Biodegradable hollow microfibres based on poly(lactic acid) were produced by a dry–wet spinning procedure. Drug‐loaded microparticles were prepared by a simple oil‐in‐water emulsion and entrapped inside the fibres. The morphology of both fibres and particles was investigated by scanning electron microscopy. The mechanical and thermal properties of the fibres were investigated by tensile tests and differential scanning calorimetry. In vitro tests were performed to evaluate the release of the drug from the fibres loaded with the particles Copyright © 2004 Society of Chemical Industry     

Effect of Bonder at Skin/Core Interface on the Mechanical Performances of Sandwich Structures Used in Marine Industry

The present work is aimed to extend the knowledge of mechanical properties of sandwich structures used for marine applications focusing on the possibility to increase the performances of such structures by adding a bonder at the skin/core interface. Therefore, three sandwich structures that are utilised in different structural components of a yacht were realised by manual lay-up. The mechanical characterisation was performed by flatwise compressive, edgewise compressive and three point flexural tests. The tests execution has allowed both to determine the mechanical performances and to understand the fracture mechanisms that take place when the bonder is added in the stacking sequence of the samples.     

Behaviour of ±45° glass/epoxy filament wound composite tubes under quasi-static equal biaxial tension–compression loading: experimental results

The primary aim of this paper is to present results describing in detail the behaviour of ±45° E-glass/MY750 (GRP) tubes, of various wall thicknesses, subjected to equal biaxial tension–compression loading, generated under combined internal pressure and axial compression. The role played by the non-linear lamina shear has also been assessed by comparing various shear stress–strain curves for embedded laminae (extracted from tests on ±45° tubes subjected to circumferential: axial stress ratios SR=1:0, 1:−1 and 2.3:−1) with that of an ‘isolated’ lamina (measured from torsion of 90° tubes). Extracted shear failure strains, for embedded laminae, were more than four fold larger than those measured at ultimate failure for an ‘isolated’ lamina. Soft characteristics were observed in the embedded lamina and these were believed to be due to interaction between early matrix damage initiation (and propagation) and shear. Factors affecting the behaviour of the tubes, such as bulging, scissoring, thermal stresses and stress variation through the thickness are discussed.     

Multiagent learning using a variable learning rate

Learning to act in a multiagent environment is a difficult problem since the normal definition of an optimal policy no longer applies. The optimal policy at any moment depends on the policies of the other agents. This creates a situation of learning a moving target. Previous learning algorithms have one of two shortcomings depending on their approach. They either converge to a policy that may not be optimal against the specific opponents' policies, or they may not converge at all. In this article we examine this learning problem in the framework of stochastic games. We look at a number of previous learning algorithms showing how they fail at one of the above criteria. We then contribute a new reinforcement learning technique using a variable learning rate to overcome these shortcomings. Specifically, we introduce the WoLF principle, “Win or Learn Fast”, for varying the learning rate. We examine this technique theoretically, proving convergence in self-play on a restricted class of iterated matrix games. We also present empirical results on a variety of more general stochastic games, in situations of self-play and otherwise, demonstrating the wide applicability of this method.     

Neural Network Feedback Control: Work at UTA’s Automation and Robotics Research Institute

This is an outline of research in neural networks for feedback control done since the mid 1990s at the Automation and Robotics Research Institute (ARRI) of The University of Texas at Arlington (UTA). It shows how the developments of Intelligent Control Systems based on neural networks have followed three main generations. This statement provides a short, broad-brush perspective on the development of intelligent neural feedback controllers.     

Creep of aromatic polyamide fibres

Creep of Kevlar 29, Kevlar 49 and PRD 49-III fibres was investigated. The fibres exhibited transient creep and the strain-time relationship was represented by a logarithmic time law. The creep strain recovered with time when the load was removed. Upon reloading to the same creep stress the strain-time relationship was again logarithmic but the creep rate was reduced. Modulus measurements were made during the creep test and these showed that the modulus increased with time. This result indicated a crystallite rotation mechanism which could account for the experimentally observed creep strain. Creep in PRD 49-III fibres exhibited a small temperature dependence over the temperature range 20°C to 150°C. The apparent creep activation energy was consistent with the range of values reported for hydrogen bonding. This suggests one possible creep mechanism in which the combined action of stress and thermal activation causes rearrangement of intercrystalline bonds in the crystallite boundaries resulting in boundary creep. Boundary creep allows crystallite rotation which produces the macroscopic creep strain. Boundary creep is discussed in terms of the fibre morphology and a model of delayed elasticity.