Combat Vehicles Research and Development Establishment

Optimizers play an important role in the accuracy and efficiency of convolutional neural networks by controlling the updates of convolutional neural networks parameters in the training process. The literature does not have a comprehensive discussion of adaptive optimizers such as Adam, AdamW and AdaBelief on the FLIR ADAS V2 thermal imaging dataset in the context of the YOLOv11 architecture, thus this paper is a novel comparison of these adaptive optimizers. The original contributions encompass: (1) the performance evaluation of AdaBelief as an empirical experiment (with mAP@50, mAP@50-95, and training time as key metrics); (2) the discovery of the optimal hyperparameters, in particular, the fact that a momentum of 0.937 is an effective one to use; and (3), the actionable conclusions about the forging of optimal hyperparameters in place of object detecting systems which use infrared cameras. Findings demonstrate that AdaBelief attains better mAP@50 (0.696) at the lowest training cost (6.999 hrs), which is an important point of reference when operating real-time vision in the defense and autonomous devices.

Since the introduction of the information technology business in Thiruvananthapuram, many women indulged in the same field. Thanks to the absence of any barrier for women entering computer-related engineering and other fields, IT professionals have high rate of female professionals. But with the same female IT professionals the scenario is different and the modern era has changed the traditional lifestyle of a woman that was tied to her family. The level of empowering women depends on their level of education and working experience. What should be explored this time is, do the women require training for some work and if all the educated women are working or are contented with their current positions. Here, to identify the impact of vocational training on women's career competencies, machine learning (ML) techniques such as convolution neural network (CNN), bagging decision tree (BRDT), XGBoost, KNN and LR are used to study the importance of IT skills oriented course development on the self success measured in working women. This analysis was taken more precisely with the CNN algorithm.

The development of lightweight, corrosion-resistant metallic lattice structures has gained significant attention in aerospace, defense, and structural applications, where material durability and weight optimization are critical. This study investigates the corrosion behavior of Laser Powder Bed Fusion (LPBF)-fabricated A286 steel honeycomb, Body-Centered Cubic (BCC), and gyroid lattices, comparing their performance against conventional materials such as Rolled Homogeneous Armor (RHA), Maraging High Strength Steel (MHA), and High-Nitrogen Steel (HNS). Corrosion testing was conducted using accelerated salt spray exposure, and the results were validated through computed tomography (CT)-based structural integrity analysis and machine learning-based predictive modeling. The experimental findings revealed that lattice structures exhibited significantly lower corrosion rates than conventional bulk materials, with the honeycomb lattice demonstrating the highest corrosion resistance (1.218 mm/year), followed by BCC (1.311 mm/year) and gyroid (1.671 mm/year). Compared to RHA, the honeycomb lattice exhibited a 57.23% reduction in corrosion rate, confirming its superior electrochemical stability. CT scan evaluations further highlighted differences in density distribution and geometric fidelity, with honeycomb lattices showing the most uniform porosity, while BCC structures displayed localized density variations at nodal intersections. To enhance predictive capabilities, various machine learning (ML) algorithms were employed to model corrosion behavior based on weight-loss measurements and lattice topology. Bayesian Ridge regression outperformed other models, achieving an R² of 0.99849 and RMSE of 0.00049, confirming its robustness in capturing corrosion trends. Linear Regression also performed well, while ensemble models such as Random Forest and XGBoost exhibited higher error margins due to dataset linearity constraints. Residual analysis and graphical interpretations further validated the stability and predictive reliability of ML-based corrosion assessments, demonstrating their feasibility as an alternative to traditional experimental methods. This study presents a comprehensive framework for integrating experimental corrosion testing, computational modeling, and CT-based defect analysis, offering a scalable approach to optimizing micro-lattice designs for corrosion-sensitive applications. The findings highlight the potential of LPBF-fabricated metallic lattices for aerospace, defense, and marine structures, where enhanced corrosion resistance, reduced material degradation, and predictive maintenance strategies are essential for long-term operational performance.

In contemporary warfare, communication breakdowns within armoured fighting vehicles present critical operational hurdles. Traditional audio channels are prone to jamming and interception, potentially jeopardizing the efficacy of military manoeuvres. This paper introduces an innovative gesture-based communication system specifically tailored for armoured fighting vehicles operating in hostile environments. The system seamlessly integrates flex sensors within gloves and harnesses a camera-based recognition framework to accurately interpret hand gestures. Employing Python and the MediaPipe library for gesture recognition and flex sensor data processing, this system establishes an alternative communication channel resilient against audio interception. Furthermore, through a dedicated machine learning algorithm, the system amalgamates inputs from both sensor types, ensuring rapid and reliable transmission of crucial messages. This solution endeavours to bolster communication reliability in armoured vehicles, furnishing a secure conduit for transmitting vital information to crew members or ground control stations while effectively mitigating risks linked to audio channel vulnerabilities.

Introduction: This study presents a comprehensive design and analysis of a 300L steel fuel tank intended for heavy off-road vehicles. The design process integrates numerical simulations and experimental investigations to optimize the tank's performance and durability under various operating conditions.Methods: The design methodology involves CAD model optimization, numerical analysis setup, and experimental validation. CAD model optimization simplifies the tank geometry while retaining structural integrity. Numerical analysis setup includes defining boundary conditions, meshing strategies, and simulation parameters. Experimental validation entails testing the tank under dynamic loading conditions to assess its structural response.Results: Numerical simulations reveal insights into stress distribution, deformation behavior, and fluid dynamics within the tank. Experimental tests confirm the numerical predictions and provide valuable data for model validation. Key results include stress concentrations in critical areas, deformation patterns under different loading conditions, and fluid flow characteristics.Discussion: The integrated approach combining numerical simulations and experimental tests offers a comprehensive understanding of the fuel tank's behavior. Findings highlight areas for design improvement, such as reinforcement of stress-prone regions and optimization of fluid flow dynamics. The study contributes to enhancing the performance, reliability, and safety of fuel tanks for heavy off-road vehicles.

Developing countries like India have large consumer markets driven by huge demands. Commercial vehicles play a critical role in full filing these demands. Commercial vehicles increasingly face stringent emission norms criteria and hence designing an ICE-powertrain with optimum operating efficiency becomes paramount. Intake manifold is the critical part of an internal combustion engine that supplies fuel/ air mixture to all the cylinders combustion chambers. It ensures a uniform mixture at cylinder inlet for better mixing inside the cylinders for better volumetric efficiency. Uneven distribution of fuel/air mixture causes unstable torque and unburnt fuel which fails to meet the emission norms. It also results in uneven temperatures in each cylinder because of cylinder misfiring. In current paper, 3D Computational Fluid Dynamics (CFD) simulations are carried out to investigate the variance and uniformity of CNG/air mixture at the outlet of intake manifold. Commercial CFD tool Ansys Fluent is used to study the flow distribution of mixture inside the manifold and runners. Initial estimation of flow pattern is done by performing a steady state simulation to predict the uniformity index of CNG at cylinder inlet. For detailed investigation, transient simulation is performed by taking fresh air and CNG mass flow rate as a function of crank angle. In this paper, mesh dependence study was done initially to achieve an optimum cell count with good accuracy. A detailed transient analysis using multi-species modelling for air & CNG was done using automated scripts with time steps as small as 1 degree crank angle rotation coupled with injection pressure and injection timing study. This helped to identify critical areas and optimise the design to improve the mass flow rate variance from 15-20% for baseline case to 6-7% for final design, and also improve the uniformity index. It also helped reduce the CNG engine mis-firing issue. The results have been well validated with Laboratory Test Results.