In most of the surveillance applications of autonomous underwater vehicle (AUV), very often it is intended to follow the desired horizontal way-points, where some oceanography data need to be collected. In view of this, the motion planning algorithm using way-points is investigated in this study. The proposed work involves identification of dynamics of AUV and design of adaptive model predictive controllers which includes linear adaptive model predictive controller (LAMPC) and non-linear adaptive model predictive controller (NAMPC). Owing to the fast convergence rate and robustness property, on-line sequential extreme learning machine (OS-ELM) is employed for estimating the dynamics of AUV. To improve the OS-ELM modelling performance, Jaya optimisation algorithm is applied to optimise the hidden layer parameters. The desired surveillance region is formulated in terms of way-points using heading angle obtained from desired line-of-sight path. Simulations are performed using MATLAB by applying proposed NAMPC, LAMPC and a previously reported optimal controller, namely inverse optimal self-tuning PID (IOSPID) controller. Subsequently, real-time experimentation is performed using a prototype AUV in a swimming pool. From the simulation and experimental results, it is observed that the proposed controller exhibit efficient tracking performance in face of actuator constraints as compared to LAMPC and IOSPID controller.

This study addresses the task of supervised cross-domain image generation, which aims to translate an image from the source domain to the target domain, guided by a reference image from the latter. The key difference between the authors setting and the recently proposed domain guided photogeneration is that their image generation is bidirectional, i.e. images are generated in both domains. Thus, they call it bidirectional cross-domain image generation. This novel task poses new challenges as it requires the model learning to decouple personalised and shared semantics of the two domains. For this purpose, they propose a framework to learn a feature space, which breaks into three parts, namely, two personalised semantic subspaces, which encode patterns that are unique for each domain, and a shared semantic subspace that captures the common patterns. The three subspaces are automatically decoupled through end-to-end training. Extensive experiments on the modified shoe and handbag datasets show that their framework can generate high-quality images in both domains.

Cognitive radio networks (CRNs) have the capacity to be aware of the conditions of their operating environment, and dynamically reconfigure their own characteristics in order to reach the best available performances. These performances may be seriously impacted when the number of users in CRNs grows significantly. This study deals with efficient energy consumption and interference avoidance in large CRNs. To enhance the network lifetime, a new framework combining cognitive hierarchical clustering and the coalitional game is introduced. In this study, a new CRLEACH protocol is proposed and the well-known LEACH protocol is used in CRNs. The authors prove theoretically that their coalition model with a new strategic learning algorithm leads to Nash equilibrium. Finally, the network performances of their framework are illustrated by numerical results.

In this research work, a deep learning algorithm is applied to the medical domain to deliver a better healthcare system. For this, a deep learning framework for classification the region of interest pattern of complex hyperspectral medical images is proposed. The performance of computer-aided diagnosis by verifying the region in hyperspectral image by pre and post-cancerous region classification is enhanced. For this a deep Boltzmann machine (DBM) architecture of the bipartite structure as an unsupervised generative model was developed. The performance of DBM is compared with deep convolutional neural network architecture. For implementation, a three-layer unsupervised network with a backpropagation structure is used. From the presented dataset, image patches are collected and classified into two classes, namely non-informative and discriminative classes as labelled classes. The spatial information is used for classification and spectral-spatial representation of class labels is formed. In the labelled classes, the accuracy, false-positive predictions, sensitivity are obtained for the proposed fully-connected network. By the proposed cognitive computation technique an accuracy of 95.5% with 93.5% sensitivity was obtained. From the obtained classification, accuracy and success rate DBM provide a better classification of complex images compared to traditional convolution network.