Enhancing the performance of photovoltaic (PV) systems has recently become a key concern because of the market demand for green energy. To obtain the most possible power from the solar module, it is imperative to allow the PV system to operate at its maximum power point (MPP) regardless of the climatic conditions. In this study, a comparison of distinctive Maximum Power-Point Tracking (MPPT) techniques is provided, which are Perturb and Observe (P&O) and Modified Variable Step-Size P&O, as well as Incremental Conductance (INC) and Modified Variable Step-Size INC, using a boost converter for two types of solar panels. Using MATLAB software, simulations have been performed to assess the efficiency of the solar module under several environmental conditions, standard test conditions (STCs), and sudden and ramp variations in both solar irradiance and temperature. The output power efficiency, time response, and steady-state power oscillations have all been taken into account in this study. The simulation results of the improved algorithms demonstrate an enhancement in the PV module performance over conventional algorithms in many factors including steady-state conditions, tracking time, and converter efficiency. Furthermore, a boost in the dynamic response in monitoring the MPP is observed in a variety of climatical circumstances. Moreover, the proposed P&O MPPT algorithm is implemented in a hardware system and the experimental results verified the effectiveness, regarding both fast-tracking speed and lower oscillations, of the proposed Variable Step-Size P&O algorithm and its superiority over the conventional P&O technique.

Driver monitoring systems are becoming an essential part of Advanced Driver Assistance Systems (ADAS) safety features in modern vehicles. The U.S. National Highway Traffic Safety Administration reports that drowsy/fatigued driving results in almost 100,000 road accidents per year. Driver’s fatigue can have different causes, such as lack of sleep, long journeys, restlessness, mental pressure and alcohol consumption. Early monitoring systems relied on data from vehicle sensors, and modern systems commonly use driver’s eye tracking. Recently, there has been growing interest in utilizing machine vision and deep learning for driver monitoring. Using machine vision can create more advanced driver monitoring systems capable of detecting driver attention state as well as other features like smartphone usage while driving and seat belts. Machine vision systems usually require extensive processing power, which raises the cost of such devices. In this paper, we present a low-cost driver monitoring system using a $15 Raspberry Pi Zero 2 W board and deep learning CNN to deliver a system capable of monitoring and identifying different states of the driver like safe driving, distracted, drowsy, and smartphone usage, the system achieves an inference rate for 10 Frames Per Second (FPS) and above 90% accuracy with the testing dataset. In addition to the deep learning CNN which runs on Raspberry Pi CPU, we utilize the Raspberry Pi GPU to run a head pose estimation algorithm to boost the system’s accuracy.

Accurate multi-object tracking (MOT) is essential for autonomous vehicles, enabling them to perceive and interact with dynamic environments effectively. Single-modality 3D MOT algorithms often face limitations due to sensor constraints, resulting in unreliable tracking. Recent multi-modal approaches have improved performance but rely heavily on complex, deep-learning-based fusion techniques. In this work, we present CLF-BEVSORT, a camera-LiDAR fusion model operating in the bird’s eye view (BEV) space using the SORT tracking framework. The proposed method introduces a novel association strategy that incorporates structural similarity into the cost function, enabling effective data fusion between 2D camera detections and 3D LiDAR detections for robust track recovery during short occlusions by leveraging LiDAR depth. Evaluated on the KITTI dataset, CLF-BEVSORT achieves state-of-the-art performance with a HOTA score of 77.26% for the Car class, surpassing StrongFusionMOT and DeepFusionMOT by 2.13%, with high precision (85.13%) and recall (80.45%). For the Pedestrian class, it achieves a HOTA score of 46.03%, outperforming Be-Track and StrongFusionMOT by (6.16%). Additionally, CLF-BEVSORT reduces identity switches (IDSW) by over 45% for cars compared to baselines AB3DMOT and BEVSORT, demonstrating robust, consistent tracking and setting a new benchmark for 3DMOT in autonomous driving.

Autonomous Underwater Vehicles (AUVs) are essential for applications such as seabed mapping, environmental monitoring, offshore infrastructure inspection, and search-and-rescue operations. However, achieving accurate trajectory tracking remains a fundamental challenge due to nonlinear and strongly coupled six-degree-of-freedom dynamics, hydrodynamic parameter uncertainties, environmental disturbances, and limitations introduced by sensor noise, actuator faults, and imperfect modeling of added mass, damping, and Coriolis–centripetal forces. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework, this review systematically classifies trajectory-tracking control strategies for AUVs into three main categories. (i) Classical controllers, including PID and Sliding Mode Control (SMC), are analyzed for their simplicity, robustness, and ease of implementation. (ii) Intelligent controllers, such as Fuzzy Logic Control (FLC), Reinforcement Learning (RL), and Physics-Informed Neural Networks (PINNs), are reviewed for their adaptability, learning capabilities, and effectiveness in handling nonlinearities and time-varying disturbances, while addressing challenges related to data availability and computational cost. (iii) Hybrid approaches, including Adaptive Neuro-Fuzzy Inference Systems (ANFIS), Physics-Informed Reinforcement Learning (PI-RL), Fault-Tolerant Control (FTC), and Sim-to-Real Transfer (SRT) techniques, are examined for their ability to integrate model-based reliability with data-driven adaptability and improve resilience under uncertainties, noise, and faults. Classical controllers provide structural simplicity and robustness but suffer reduced accuracy in highly nonlinear and noisy environments. Intelligent methods like RL and PINNs improve adaptability and reduce tracking errors but demand extensive data and computational resources. Hybrid approaches, particularly ANFIS and PI-RL, achieve high tracking accuracy and maintain robust performance under various uncertainties. Sim-to-Real Transfer (SRT) techniques further enhance real-world deployment. A meta-analysis of 120 peer-reviewed studies quantifies performance trends in terms of root-mean-square error (RMSE), settling time, robustness, and computational cost. The review highlights future research opportunities in domain randomization, adaptive fault-tolerant control, and physics-guided hybrid learning to enable reliable real-world AUV operations.

Distributed deep learning is a promising approach for training and deploying large and complex deep learning models. This paper presents a comprehensive workflow for deploying and optimizing the YOLACT instance segmentation model as on big data clusters. OpenVINO, a toolkit known for its high-speed data processing and ability to optimize deep learning models for deployment on a variety of devices, was used to optimize the YOLACT model. The model is then run on a big data cluster using BigDL, a distributed deep learning library for Apache Spark. BigDL provides a high-level programming interface for defining and training deep neural networks, making it suitable for large-scale deep learning applications. In distributed deep learning, input data is divided and distributed across multiple machines for parallel processing. This approach offers several advantages, including the ability to handle very large data that can be stored in a distributed manner, scalability to decrease processing time by increasing the number of workers, and fault tolerance. The proposed workflow was evaluated on virtual machines and Azure Databricks, a cloud-based platform for big data analytics. The results indicated that the workflow can scale to large datasets and deliver high performance on Azure Databricks. This study explores the benefits and challenges of using distributed deep learning on big data clusters for instance segmentation. Popular distributed deep learning frameworks are discussed, and BigDL is chosen. Overall, this study highlights the practicality of distributed deep learning for deploying and scaling sophisticated deep learning models on big data clusters.

Content Based Image Retrieval, CBIR, is a highly active leading research field with numerous applications that are currently expanding beyond traditional CBIR methodologies. In this paper, a CBIR methodology is proposed to meet such demands. Query inputs of the proposed methodology are an image and a text. For instance, having an image, a user would like to obtain a similar one with some modification described in text format that we refer to as a text-modifier. The proposed methodology uses a set of neural networks that operate in feature space and perform feature composition in a uniform-known domain which is the textual feature domain. In this methodology, ResNet is used to extract image features and LSTM to extract text features to form query inputs. The proposed methodology uses a set of three single-hidden-layer non-linear feedforward networks in a cascading structure labeled NetA, NetC, and NetB. NetA maps image features into corresponding textual features. NetC composes the textual features produced by NetA with text-modifier features to form target image textual features. NetB maps target textual features to target image features that are used to recall the target image from the image-base based on cosine similarity. The proposed architecture was tested using ResNet 18, 50 and 152 for extracting image features. The testing results are promising and can compete with the most recent approaches to our knowledge as listed in section 5.