Abstract:A typical mechatronic system consists of a multitude of components, and the sensors belong to an important and cru-cial class of such components. Optimal matching of the system components is implicit in the current definition of a mechatronic system. The focus of the present paper is the optimal matching of sensors with other hardware in the system. Sensor matching may be based on several concepts such as the operating frequency range (operating band-width), speed of response (and the corresponding rate of data sampling in digital conversion), the device sensitivity (or gain or data amplification), and the effect of component accuracy on the overall accuracy of the system. The pre-sent paper explores all these concepts and presents suitable approaches for sensor matching through those criteria. The relevant procedures are illustrated using case studies.

Abstract:Perhaps hearing impairment individuals cannot identify the environmental sounds due to noise around them. However, very little research has been conducted in this domain. Hence, the aim of this study is to categorize sounds generated in the environment so that the impairment individuals can distinguish the sound categories. To that end first we define nine sound classes--air conditioner, car horn, children playing, dog bark, drilling, engine idling, jackhammer, siren, and street music-- typically exist in the environment. Then we record 100 sound samples from each category and extract features of each sound category using Mel-Frequency Cepstral Coefficients (MFCC). The training dataset is developed using this set of features together with the class variable; sound category. Sound classification is a complex task and hence, we use two Deep Learning techniques; Multi Layer Perceptron (MLP) and Convolution Neural Network (CNN) to train classification models. The models are tested using a separate test set and the performances of the models are evaluated using precision, recall and F1-score. The results show that the CNN model outperforms the MLP. However, the MLP also provided a decent accuracy in classifying unknown environmental sounds.

Abstract:In order to improve the detection accuracy of chaotic small signal prediction models under the background of sea clutter, a distributed sea clutter denoising algorithm is proposed, on the basis of variational modal decomposition (VMD). The sea clutter signal is decomposed into variational modal functions (VMF) with different center bandwidths by means of VMD. By analyzing the autocorrelation characteristics of the decomposed signal, we perform instantaneous half-period (IHP) and wavelet threshold denoising processing on the high-frequency and low-frequency components respectively, and regain the sea clutter signals. Based on LSSVM sea clutter prediction model, this research compares and analyzes the denoising effects of VMD. Experiment results show that, the RMSE after denoising is reduced by two orders of magnitude, approximating 0.00034, with an apparently better denoising effect, compared with the root mean square error (RMSE) of the prediction before denoising.

Abstract:Parkinson's Disease (PD) is a neurodegenerative disease which shows a deficiency in dopaminehormone in the brain. It is a common irreversible impairment among elderly people. Identifying this disease in its preliminary stage is im-portant to improve the efficacy of the treatment process. Disordered gait is one of the key indications of early symptoms of PD. Therefore, the present paper introduces a novel approach to identify parkinsonian gait using raw vertical spatiotemporal ground reaction force. A convolution neural network (CNN) is implemented to identify the features in the parkinsonian gaits and their progressive stages. Moreover, the variations of the gait pressures were visually recreated using ANSYS finite element software package. The CNN model has shown a 97% accuracy of recognizing parkinsonian gait and their different stages, and ANSYS model is implemented to visualize the pressure variation of the foot during a bottom-up approach.

Abstract:The real random number generation is a critical problem in computer science. The current generation methods are ei-ther too dangerous or too expensive, such as using decays of some radioactive elements. They are also hard to con-trol. By the declaration of uncertainty principles in quantum mechanics, real probabilistic events can be substituted by easier and safer processes, such as electron diffraction, photon diffraction and qubits. The key to solve the problem of Schrödinger’s cat is to identify that the atom stays in different states after and before the decay, and the result of the decay is probabilistic according to the wave packet collapse hypothesis. Same matter is able to possess different kinds of properties such as wave-particle duality due to that it can stay in various states, and which state will the matter stay is determined by the chosen set of physical quantities (or mechanical quantities). One eigenstate of a set of physical quantities can be a superposition of other eigenstates of different sets of physical quantities, and the col-lapse from a superposition to an eigenstate it contains is really random. Using this randomness, real random number can be generated more easily.

Abstract:Determination of an age in a particular tree species can be considered as a vital factor in forest management. In this research we have introduced a novel scheme to determine the accurate age of the tree species in Sri Lanka. This is initially developed for the tree species called‘Hora’ (Dipterocarpus zeylanicus ) in wet zone of Sri Lanka. Here the core samples are extracted and further analyzed by means of the different image processing techniques such as Gaussian kernel blurring, use of Sobel filters, double threshold analysis ,Hough line transformation and etc. The operations such as rescaling, slicing and measuring are also used in line with image processing techniques to achieve the desired results. Ultimately a Graphical user interface (GUI) is developed to cater for the user requirements in a user friendly envi-ronment. It has been found that the average growth ring identification accuracy of the proposed system is 93% and the overall average accuracy of detecting the age is 81%. Ultimately the proposed system will provide an insight and contributes to the forestry related activities and researches in Sri Lanka.

Abstract:Machine embroidery is a multi-step process with many variables that govern the quality of the final product. In Sri Lanka a few large-scale factories use machine embroidery process. They import heavy and very expensive machines from the other countries in the world. The main disadvantage of these machines is, they are not suitable for production of small order quantities due to high cost. Also, these machines and the required software for the designing and operating process are comparatively expensive. Small scale apparel factories and domestic industrialists cannot afford the high cost of hardware and software. Besides, it requires persons with high skills for design, operation and maintenance. There are a few domestic type automated embroidery machines, but those are exorbitant. The present study aims to design and fabricate retrofit for ordinary sewing machine that is capable of performing given embroidery pattern. The retrofit comprises of an frame which can be moved in X and Y directions by three controlling motors. The design of the embroidery is converted to a G-code using GBRL software and then the microcontroller governs the movement of the frame to produce the embroidery. The retrofit designed and developed performed embroidery function and the quality of the output was checked and it was found to be satisfactory as far as first conceptual model is concerned. Further de-velopment work is required to improve user requirements and technical requirements including the increase of speed of operation which may be done with synchronization of the three motors.