Advanced Materials Research Vol. 1117

Abstract: The present work was focused on development and studies of mechanical properties that natural fibres have in the woven reinforcements made from hemp and flax as well as hybrid yarns of hemp and glass fibres. Natural fibres such as hemp and flax are biodegradable, have low weight and show good flexibility. Glass fibre is widely used in the industry when low cost and good performance is required. The hemp yarns (100 Tex and 1186 Tex), the flax yarns (678 Tex) and the hybrid yarn of hemp and glass fibres (1644 Tex) were used to develop woven reinforcement structures. Average surface density for reinforcements of hemp yarns is 83- 529 g/m² and for reinforcements of hybrid yarns 738- 741 g/m².

Abstract: The intelligent systems of smart textile should contain flexible electronics, for example, sewn or stitched textile elements that function as conductive traces, sensor elements, electrodes or switches. The experiments for development of sewn touch and push switches are performed. For this reason silver coated multifilament polyamide and multifilament stainless steel yarns were used and their properties and suitability tested. Tests include the changes of electrical resistance of yarns during tension, measurements of yarn voltage loss at 50 mA current; experiments to assess the yarn suitability for sewing; reliability tests of switches during use.

Abstract: Nowadays, in the field of control engineering, neural networks (NNs) are very useful, because of their learning ability. In case of control tasks, they are often used to adapt to the unknown or changing behavior of the system to be controlled. When the system is unknown, or partially known, at the beginning, it is very difficult to set the controller properly, and imprecision may occur in the control process. In this case, some extra calculations are needed to get more accurate results. One of the possible solutions is the application of Robust Fixed Point Transformations (RFPT). In this paper, it is shown that RFPT can improve the accuracy of the control achieved by a traditional NN controller.

Abstract: In many outdoor robotic applications several factors have to be taken into account during a path planning process. In different situations the importance of these factors vary. This paper presents a path planning method for mobile robots that incorporates decision theory to guide the search. A neural structure is proposed to determine the relative importance of the objectives that makes the robot capable of planning in unfamiliar situations. The method is able to handle an arbitrary number of objectives simultaneously and also enables the incorporation of human logic into the planning process. All parts of the algorithm suit real-time implementation.

Abstract: Fuel economy improvement has become the most important issue in automobile engine developments nowadays. The measurements of various parameters in the engine parts, such as temperature, strain and gap have become more important due to the development of fuel-saving technologies. The conventional measuring method is difficult to be adapted to the moving parts at high engine speeds due to the durability of the wire in the measurement system. On the other hand, the telemetric system enables the measuring information to be transferred wirelessly. Therefore, we have applied this system to piston temperature measurements that need to be conducted under various operating conditions. The experimental results of the piston temperature measurements at high engine speeds up to 6500[rpm] are shown in this report.

Abstract: In present work we develop the stable methods for technical and technological diagnostics of unsteady objects and/or processes by means of the L.A. Urban GPA (Gas Path Analysis) method: the dynamic system classes, which are suitable to be applied by technical and technological diagnostics by means of the L.A. Urban GPA method, are allocated; regularizing algorithms for construction of a stable diagnostic matrix are developed; methods for finding the optimal regularization parameter are proposed; development of methods for optimal regularizing parameter choice.

Abstract: In this paper, a new filter network is presented that is based on Radial Base Function Networks (RBFNs). The output layer of the network is modified, in order to make it more effective in certain fuzzy control systems. The training of the network is solved by a clustering step, for which two different clustering methods are proposed. The suggested structure can efficiently be used for data classification.

Abstract: Reliability of voice-controlled Ambient Assisted Living (AAL) systems depends mostly on real-time recognition of speech commands coming from a noisy environment. This paper discusses the aspects of constructing anytime algorithms for robust speech recognition in AAL systems. In this case, command recognition is a speaker-dependent classification on a limited set of isolated phrases. An interruptible search can rule out every command that is not a possible solution according to various parameters of speech, and can identify a spoken command with a sufficiently high probability, while keeping time constraints in mind.

Abstract: This paper proposes a novel anytime fuzzy supervisory expert system for online signal processing. We demonstrate via simulations that this system is able to follow slowly varying signals and heal the signal in case of missing input data. In the presence of contaminating noise, the supervisory system performs the automatic wavelet shrinkage procedure selection, which ensures to pick the proper algorithm that is the most efficient in the given scenario. The necessary level of wavelet decomposition is determined online by the fuzzy supervisory expert. The system applies orthogonal wavelet functions in order to reduce significantly the processing time of reconstruction. The paper also shows how the online threshold estimator selection module ensures the highest denoising efficiency by selecting the most suitable algorithm.

Abstract: In this paper, the usage possibilities of personal statistics are introduced, which can be applied to improve the patient-specific evaluation in health monitoring systems. The aim of these techniques is to obtain reliable results based on previous measurements. This goal can be achieved by membership function tuning or modification, as well as by a pre-processing method, which is used to judge whether a situation is normal or not. In the latter case, a further requirement, that the appropriate result should be available in time, can also be fulfilled. If the situation is judged to be critical then a reduced model is evaluated instead of the full one.