This study provides a comprehensive evaluation of Convolutional Neural Network (CNN) and Vision Transformer (ViT) models for driver fatigue classification, critical issue in road safety. Using custom driving behavior dataset, state-of-the-art CNN ViT architectures, including VGG16, EfficientNet, MobileNet, Inception, DenseNet, ResNet, ViT, Swin Transformer, were analyzed this to determine the best model practical monitoring systems. Performance metrics such as accuracy, F1-score, training...

Platoon based traffic flow models form the underlying theoretical framework in simulation tools. They are essentially important facilitating efficient performance calculation and evaluation urban networks. For this purpose, a new platoon-based macroscopic model called LWR-IM has been developed [1]. Preliminary analytical validation conducted previously proven feasibility of model. In paper, is further enhanced with algorithms that describe platoon interactions arterials. The proposed...

Finger prostheses are devices developed to emulate the functionality of natural human fingers. On top their aesthetic appearance in terms shape, size and colour, such biomimetic require a high level dexterity. They must be capable gripping an object, even manipulating it hand. This paper presents robotic finger actuated by hybrid mechanism integrated with tactile sensor. The actuation comprises DC micromotor Shape Memory Alloy (SMA) wire. A customized test rig has been measure force stroke...

The issue of driving behavior at night poses significant challenges due to reduced visibility and increased risk accidents. Recent works have leveraged deep learning techniques enhance night-time safety. However, the limited availability high-quality training data lack robustness in existing models present problems. In this work, we propose a novel approach improve recognition using ResNet50 with contrast adapted histogram equalization (CLAHE). We collected new dataset developed more...

Existing antibody language models (

This paper presents the modelling of agarwood oil (AO) significant compounds by different qualities using Scaled Conjugate Gradient (SCG) algorithm. technique involved data collection from Gas Chromatography-Mass Spectrometry (GC-MS) for compound extraction. The development Multilayer perceptron (MLP) is used to discriminate AO chemical high and low quality. input output was transferred MATLAB version R2013a extended analysis. abundances (%) quality either or low. identification, selection...

Air quality monitoring is essential to ensure that the places we work or live in are conducive for our health. Recent technologies have enabled remote of conditions environment such as air and water. This paper presents a based on myRIO-LabVIEW, which could monitor via detection Carbon Monoxide (CO) rate indoor locations within buildings. The sensor use detect CO MQ-7 connected Arduino Mega MyRIO controllers; outputs processed by LabViewsoftware user computer remotely. For purpose testing,...

Lane detection and tracking technique are commonly used for a vehicle to navigate autonomously on the road. Various techniques have been developed by researchers it seems image processing from vision sensors appears be popular approach. Hence, seeing relevance of technique, this research intends develop road lane which comprises OpenCV, Gaussian Blur, Masking, Canny Edge, Hough Transform methods. The was set run using an embedded controller that is connected sensor. They were installed...

Research on developing mathematical and simulative models to evaluate performance of signalized arterials is still ongoing. In this paper, an integrated model (IM) based Rakha vehicle dynamics LWR proposed. The IM which imitates actuated measurement in described using continuous timed Petri net with variable speeds (VCPN). This enables systematic discretized description platoon movement from upstream intersection towards a downstream intersection. integration the notion that speed travel...

This study investigates the performance of Multilayer Perceptron (MLP) classifier in discriminating qualities agarwood oil significant compounds by different based on three training algorithms namely Scaled Conjugate Gradient (SCG), Levernbergh-Marquardt (LM) and Resilient Backpropagation (RP) Neural Network using Matlab version 2013a. The dataset used this were obtained at Forest Research Institute Malaysia (FRIM) University Pahang (UMP). Further, areas (abundances, %) chemical is set as an...

Model‐based optimal control strategies were capable of handling saturated traffic regimes with the ability to maintain optimised network performance. However, upstream queue spillback which was not considered by these renders an inefficient oversaturated regime. Hence, situations needs both flow coordination and regulation. Here, author aims regulate model‐based strategy (MCS) based on Lighthill‐Whitham‐Richards–integrated model (LWR‐IM) model. A simulator developed in a previous work is...

The LWR theory has been notable in modeling traffic progression. However, its reliance on a fundamental diagram and the shock wave to evaluate performance of signalized intersection is rather tedious procedure. Hence, this paper, we propose an integration Rakha vehicle dynamics model "assist" which described using VCPN estimate platoon arrival profile along arterials. It hypothesized that obtained from can be analyzed directly inputoutput analysis less compared shockwave analysis. This paper...

Vehicles' motion on roads causes road surface vibration. Road vibration induced by passing vehicles has been investigated some researchers in attempt to obtain traffic data such as volume, speed, occupancy and vehicle types. This type of non-deterministic study signals employs parametric well non-parametric methods explore important properties the estimate data. In this paper, we present estimation speed from vehicles. A logger which comprises an ADXL355 accelerometer is developed measure...

Malaysia has one of the highest traffic fatality rates in world. The main cause towards increment annual rate on accident is due to distracted driver wheel. Due advancement and integration technologies within society, drivers tend get either by their devices or infotainment that build-in with vehicles. This paper presents application deep learning classify driver's behavior behind implements convolution neural network experiment was conducted drowsiness dataset 10 classes from State Farm 2...

<span lang="EN-US">The autoregressive integrated moving average (ARIMA) method has been used to model global navigation satellite systems (GNSS) measurement errors. Most ARIMA error models describe time series data of static GNSS receivers. Its application for modeling under dynamic tests is not evident. In this paper, we aim real kinematic-GNSS (RTK-GNSS) errors during using linear regression with establish a proof concept via simulation that along trajectory logged by the RTK-GNSS...

A new traffic model called LWR-IM has been developed to simulate dynamics in urban arterial network. The arrival profile produced by the could be analyzed using input-output method which is less tedious compared analyzing LWR Shockwave analysis. This paper presents a comparative study of queues and analysis respectively. reveals that simulated compatible with More importantly, provides more efficient analyze LWR, thus further ascertaining feasibility efficiency model.

The sparse Gaussian process regression (GPR) has been used to model trajectory data from Real time kinematics-global navigation satellite system (RTK-GNSS). However, upon scrutinizing the residuals; GPR poorly fits and exhibits presence of correlated noise. This work attempts address these issues by proposing an integrated modeling approach called GPR-LR-ARIMA where was with linear autoregressive moving average errors (LR-ARIMA) further enhance description data. In this approach, predicted...