arXiv subjects

Indranil Bhattacharya

Publications and source records attributed to Indranil Bhattacharya.

7 recordsLinked to original sources

arXiv2025

Cybernaut: Towards Reliable Web Automation

The emergence of AI-driven web automation through Large Language Models (LLMs) offers unprecedented opportunities for optimizing digital workflows. However, deploying such systems within industry's real-world environments presents four core challenges: (1) ensuring consistent execution, (2) accurately identifying critical HTML elements, (3) meeting human-like accuracy in order to automate operations at scale and (4) the lack of comprehensive benchmarking data on internal web applications. Existing solutions are primarily tailored for well-designed, consumer-facing websites (e.g., Amazon.com, Apple.com) and fall short in addressing the complexity of poorly-designed internal web interfaces. To address these limitations, we present Cybernaut, a novel framework to ensure high execution consistency in web automation agents designed for robust enterprise use. Our contributions are threefold: (1) a Standard Operating Procedure (SOP) generator that converts user demonstrations into reliable automation instructions for linear browsing tasks, (2) a high-precision HTML DOM element recognition system tailored for the challenge of complex web interfaces, and (3) a quantitative metric to assess execution consistency. The empirical evaluation on our internal benchmark demonstrates that using our framework enables a 23.2% improvement (from 72% to 88.68%) in task execution success rate over the browser_use. Cybernaut identifies consistent execution patterns with 84.7% accuracy, enabling reliable confidence assessment and adaptive guidance during task execution in real-world systems. These results highlight Cybernaut's effectiveness in enterprise-scale web automation and lay a foundation for future advancements in web automation.

cs.SE↗

arXiv2025

SOP-Bench: Complex Industrial SOPs for Evaluating LLM Agents

LLM-based agents struggle to execute complex, multi-step Standard Operating Procedures (SOPs) that are fundamental to industrial automation. Existing benchmarks fail to capture the procedural complexity and tool orchestration demands of real-world workflows. We introduce SOP-Bench, a benchmark of 2,000+ tasks from human expert-authored SOPs across 12 business domains (healthcare, logistics, finance, content moderation, etc.). Using a human-AI collaborative framework, experts crafted authentic SOPs while AI generated artifacts (tools, APIs, datasets), all human-validated, yielding realistic tasks with executable interfaces and ground-truth outputs. SOP-Bench serves as a research enabler for systematically investigating agent architectures, model capabilities, and deployment considerations across diverse procedural tasks. We demonstrate its utility through illustrative experiments with a subset of frontier models across Function-Calling (FC) and ReAct agents, revealing critical insights. For example, (1) newer models do not guarantee better performance - Claude 4 family outperforms Claude 4.5 family on ReAct tasks (Claude 4 Opus: 72.4% vs. Claude 4.5 Sonnet: 63.3% task success rate), demonstrating that production upgrades require validation; (2) no single model-agent combination dominates: best performances range from 57% to 100% depending on domain. These examples illustrate how SOP-Bench enables isolating and studying specific dimensions of agent performance without costly production experiments. Our goal is not to rank model capabilities or build optimal agents, but to provide a rigorous evaluation framework that enables the researchers and practitioners to systematically investigate agent design choices, model selection, and deployment strategies. We release the benchmark at https://github.com/amazon-science/sop-bench.

cs.AI↗

arXiv2022

Equivalent Circuit Modeling and Analysis of Metamaterial Based Wireless Power Transfer

In this study, an equivalent circuit model is presented to emulate the behavior of a metamaterial-based wireless power transfer system. For this purpose, the electromagnetic field simulation of the proposed system is conducted in ANSYS high frequency structure simulator. In addition, a numerical analysis of the proposed structure is explored to evaluate its transfer characteristics. The power transfer efficiency of the proposed structure is represented by the transmission scattering parameter. While some methods, including interference theory and effective medium theory have been exploited to explain the physics mechanism of MM-based WPT systems, some of the reactive parameters and the basic physical interpretation have not been clearly expounded. In contrast to existing theoretical model, the proposed approach focuses on the effect of the system parameters and transfer coils on the system transfer characteristics and its effectiveness in analyzing complex circuit. Numerical solution of the system transfer characteristics, including the scattering parameter and power transfer efficiency is conducted in Matlab. The calculation results based on numerical estimation validates the full wave electromagnetic simulation results, effectively verifying the accuracy of the analytical model.

eess.SY↗

arXiv2022

Modeling and Analysis of Grid Tied Combined Ultracapacitor Fuel Cell for Renewable Application

In this manuscript, the performance of an ultracapacitor fuel cell in grid connected mode is investigated. Voltage regulation to the ultracapacitor was achieved with a three level bidirectional DC-DC converter while also achieving power flow from the grid to the ultra-capacitor via the bidirectional converter. The choice of a bidirectional three level converter for voltage regulation is based on its inherently high efficiency, low harmonic profile and compact size. Using the model equations of the converter and grid connected inverter derived using the switching function approach, the grid's direct and quadrature axes modulation indices, Md and Mq, respectively were simulated in Matlab for both lagging and leading power factors. Moreover, the values of Md and Mq were exploited in a PLECS based simulation of the proposed model to determine the effect of power factor correction on the current and power injection to grid

eess.SY↗

arXiv2020

Adaptive Step Size Incremental Conductance Based Maximum Power Point Tracking (MPPT)

Extracting maximum power available from photovoltaic arrays requires the system operating at the maximum power point (MPP). Therefore, finding the MPP is necessary for efficient operation of PV arrays. The MPP changes with multiple environmental factors, mainly temperature and irradiance. Traditionally, the incremental conductance technique with fixed step size was used to find the MPP, which suffers from a trade-off between speed of convergence and accuracy. In this work, we propose an incremental conductance maximum power point tracking (MPPT) algorithm with a variable step size, which adaptively changes the step size after each iteration based on how far away the current operating point is from a new MPP. This mitigates the aforementioned trade-off drastically by achieving faster convergence speed without the loss of accuracy. A series of simulations involving variations in temperature and irradiance were performed using MATLAB, and the speed of convergence and accuracy were compared with the traditional IC technique.

eess.SY↗

arXiv2020

Material selection method for a perovskite solar cell design based on the genetic algorithm

In this work, we propose a method of selecting the most desirable combinations of material for a perovskite solar cell design utilizing the genetic algorithm. Solar cells based on the methylammonium lead halide, CH3NH3PbX3, attract researchers due to the benefits of their high absorption coefficient and sharp Urbach tail, long diffusion length and carrier lifetime, and high carrier mobility. However, their poor stability under exposure to moisture still poses a challenge. In our work, we assigned stability index, power conversion efficiency index, and cost-effectiveness index for each material based on the available experimental data in the literature, and our algorithm determined the TiO2/CH3NH3PbI2.1Br0.9/Spiro-OMeTAD as the most well balanced solution in terms of cost, efficiency, and stability. The proposed method can be extended further to aid the material selection in all-perovskite multijunction solar cell design as more data on perovskite materials become available in the future.

cond-mat.mtrl-sci↗

arXiv2019

Novel high efficiency quadruple junction solar cell with current matching and quantum efficiency simulations

A high theoretical efficiency of 47.2% was achieved by a novel combination of In0.51Ga0.49P, GaAs, In0.24Ga0.76As and In0.19Ga0.81Sb subcell layers in a simulated quadruple junction solar cell under 1 sun concentration. The electronic bandgap of these materials are 1.9 eV, 1.42 eV, 1.08 eV and 0.55 eV respectively. This unique arrangement enables the cell absorb photons from ultraviolet to deep infrared wavelengths of the sunlight. Emitter and base thicknesses of the subcells and doping levels of the materials were optimized to maintain the same current in all the four junctions and to obtain the highest conversion efficiency. The short-circuit current density, open circuit voltage and fill factor of the solar cell are 14.7 mA/cm2, 3.38 V and 0.96 respectively. In our design, we considered 1 sun, AM 1.5 global solar spectrum.

physics.app-ph↗