Data Scientist

Data Scientist

he/him

I am a Data Scientist and Researcher. I am Interested in LLMs, Deep Learning, and Computer Vision.

Publications

You can also find my articles on my Google Scholar profile.

Masters Thesis

Over the Horizon of Connected Vehicles: Advancing Security, Reliability, and Cost-Efficiency through Vehicular Edge Computing

Published in IIITH Publication Network for Thesis, 2024

Featured in IIITH Publication Network for Thesis, 2024

Thesis Topic: Connected Vehicles

Recommended citation: Vaddhiparthy, SVSLN Surya Suhas. "Over the Horizon of Connected Vehicles: Advancing Security, Reliability, and Cost-Efficiency through Vehicular Edge Computing." PhD diss., International Institute of Information Technology Hyderabad, 2024.
Download Paper

Journal Articles

Towards energy-aware operational decisions for automated guided vehicles: A review

Published in ITU Journal on Future and Evolving Technologies, Volume 6 (2025), Issue 2, Pages 132-148, 2025

Featured in ITU Journal on Future and Evolving Technologies, Volume 6 (2025), Issue 2, Pages 132-148, 2025

This paper A comprehensive review of energy consumption models provides insight into the physical characteristics of AGVs energy consumption. Finally, we provide variations of task scheduling and path planning problems that explicitly take into account such energy aspects and introduce how literature tackles to solve those problems.

Recommended citation: KyuJin Kyung, GaHyun Lee, SeongJae Chu, Joseph John Cherukara, SVSLN Surya Suhas Vaddhiparthy, Tanniru Abhinav Siddharth, Deepak Gangadharan, BaekGyu Kim. "Towards energy-aware operational decisions for automated guided vehicles: A review" ITU Journal on Future and Evolving Technologies, Volume 6 (2025), Issue 2, Pages 132-148
Download Paper

Conference Papers

KshemaGPT: A Multi-Agent LLM for Agriculture & Enterprise AI

Published in 31st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 2025, 2025

Featured in 31st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 2025, 2025

This paper presents a multi-agent architecture which brings in multiple domain-specific agents such as Policy Agent, Crop Agent, User Agent, Employee Agent, Translation Agent, and Speech Recognition Agent combined through Query Router Agent to handle the diverse input and information requirements

Recommended citation: SVSLN Surya Suhas Vaddhiparthy, Gokulraj R, Rajesh Nani Dasari, et al. Technical Report on KshemaGPT: A Multi-Agent LLM for Agriculture & Enterprise AI. TechRxiv. July 28, 2025
Download Paper

S2P: Two-Stage Superpixel Algorithm for Enhanced Lane Detection on Resource Constraint Edges

Published in 2024 IEEE 14th International Symposium on Industrial Embedded Systems (SIES), 2024

Featured in 2024 IEEE 14th International Symposium on Industrial Embedded Systems (SIES), 2024

This paper proposes a two-stage algorithm compared to traditional HT and PHT methods, with the two-stage algorithm outperforming the standard HT(0.88 vs 0.92) and PHT(0.89 vs 0.95) in terms of Average Accuracy and Recall. Additionally, the two-stage PHT method significantly reduces power consumption compared to the Ultra Fast Lane Detection (UFLD) model, making it ideal for low-power edge devices like the Raspberry Pi 4B and Jetson Nano.

Recommended citation: P Rhuthik, SVSLN Surya Suhas Vaddhiparthy, Pranav Kannan, Deepak Gangadharan. "S2P: Two-Stage Superpixel Algorithm for Enhanced Lane Detection on Resource Constraint Edges" 2024 IEEE 14th International Symposium on Industrial Embedded Systems (SIES)
Download Paper

Attention Meets UAVs: A Comprehensive Evaluation of DDoS Detection in Low-Cost UAVs

Published in 2024 IEEE 20th International Conference on Automation Science and Engineering (CASE), 2024

Featured in 2024 IEEE 20th International Conference on Automation Science and Engineering (CASE), 2024

This paper addresses enhancing cybersecurity for low-cost Wi-Fi-based UAVs against DDoS attacks by developing a detection mechanism that evaluates various machine learning and deep learning algorithms, finding that the Time Series Transformer (TST) is the most effective model with an F1 score of 0.999 for TCP attacks and an overall runtime of 0.1 seconds, while also detailing steps for on-board implementation and hyperparameter optimization.

Recommended citation: Sharma, Ashish, SVSLN Surya Suhas Vaddhiparthy, Sai Usha Goparaju, Deepak Gangadharan, and Harikumar Kandath. "Attention Meets UAVs: A Comprehensive Evaluation of DDoS Detection in Low-Cost UAVs." arXiv preprint arXiv:2406.19881 (2024).
Download Paper

Scalable and Interoperable Distributed Architecture for IoT in Smart Cities

Published in 2023 IEEE 9th World Forum on Internet of Things (WF-IoT), 2023

Featured in 2023 IEEE 9th World Forum on Internet of Things (WF-IoT), 2023

This paper presents a distributed, multi-layered data platform architecture for smart cities, demonstrating significant improvements in throughput and latency through real-world analysis of the Smart City Living Lab at IIIT Hyderabad, enhancing sensor data management and interoperability.

Recommended citation: Pranavasri, V. J. S., Leo Francis, Ushasri Mogadali, Gaurav Pal, SVSLN Surya Suhas Vaddhiparthy, Anuradha Vattem, Karthik Vaidhyanathan, and Deepak Gangadharan. "Scalable and Interoperable Distributed Architecture for IoT in Smart Cities." In 2023 IEEE 9th World Forum on Internet of Things (WF-IoT), pp. 01-06. IEEE, 2023.
Download Paper

Collision-Aware Data Delivery Framework for Connected Vehicles via Edges

Published in 2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall), 2023

Featured in 2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall), 2023

This study presents an optimization framework for collision-aware data delivery in connected vehicles using HCCA and VeMAC MAC protocols, effectively minimizing global slot utilization and enhancing service delivery amid varying vehicle densities in real-world traffic scenarios.

Recommended citation: Vaddhiparthy, Svsln Surya Suhas, Joseph John Cherukara, Deepak Gangadharan, and BaekGyu Kim. "Collision-Aware Data Delivery Framework for Connected Vehicles via Edges." In 2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall), pp. 1-7. IEEE, 2023.
Download Paper

Dynamic Data Delivery Framework for Connected Vehicles via Edge Nodes with Variable Routes

Published in 2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall), 2023

Featured in 2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall), 2023

This work proposes a two-stage optimization framework for efficient data delivery to connected vehicles via edge nodes, addressing dynamic route changes, and demonstrates its superiority over offline optimization and greedy approaches using a real-world dataset.

Recommended citation: Cherukara, Joseph John, SVSLN Surya Suhas Vaddhiparthy, Deepak Gangadharan, and BaekGyu Kim. "Dynamic Data Delivery Framework for Connected Vehicles via Edge Nodes with Variable Routes." In 2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall), pp. 1-7. IEEE, 2023.
Download Paper

A Comprehensive Evaluation on the Impact of Various Spoofing Scenarios on GPS Sensors in a Low-Cost UAV

Published in 2023 IEEE 19th International Conference on Automation Science and Engineering (CASE), 2023

Featured in 2023 IEEE 19th International Conference on Automation Science and Engineering (CASE), 2023

This paper analyzes 16 GPS spoofing scenarios for low-cost UAVs, revealing that effective spoofable distances can reach up to 100 meters under certain conditions, with significant variations in GPS parameters based on environmental factors and satellite visibility.

Recommended citation: Vaddhiparthy, SVSLN Surya Suhas, Garapati Sreya, Prudhvi Raj Turlapati, Deepak Gangadharan, and Harikumar Kandath. "A Comprehensive Evaluation on the Impact of Various Spoofing Scenarios on GPS Sensors in a Low-Cost UAV." In 2023 IEEE 19th International Conference on Automation Science and Engineering (CASE), pp. 1-6. IEEE, 2023.'
Download Paper

A Multi Layer Data Platform Architecture for Smart Cities using oneM2M and IUDX

Published in 2022 IEEE 8th World Forum on Internet of Things (WF-IoT), 2022

Featured in 2022 IEEE 8th World Forum on Internet of Things (WF-IoT), 2022

This paper presents a multi-layer architecture for smart cities, compliant with oneM2M and IUDX, that enhances data management and sharing, demonstrated through a proof-of-concept on a university campus achieving efficient performance metrics.

Recommended citation: Mante, Shubham, SVSLN Surya Suhas Vaddhiparthy, Muppala Ruthwik, Deepak Gangadharan, Aftab M. Hussain, and Anuradha Vattem. "A multi layer data platform architecture for smart cities using onem2m and iudx." In 2022 IEEE 8th World Forum on Internet of Things (WF-IoT), pp. 1-6. IEEE, 2022.
Download Paper

Design of an IoT system for machine learning calibrated TDS measurement in Smart Campus

Published in 2021 IEEE 7th World Forum on Internet of Things (WF-IoT), 2021

Featured in 2021 IEEE 7th World Forum on Internet of Things (WF-IoT), 2021

The paper describes a low-cost IoT-based TDS measurement system for smart campuses that uses non-invasive data collection, wireless connectivity, and machine learning for calibration, achieving a regression model with an R² of 93.96% and RMSE of 27.93.

Recommended citation: Goparaju, Sai Usha Nagasri, SVSLN Surya Suhas Vaddhiparthy, C. Pradeep, Anuradha Vattem, and Deepak Gangadharan. "Design of an IoT system for machine learning calibrated TDS measurement in smart campus." In 2021 IEEE 7th World Forum on Internet of Things (WF-IoT), pp. 877-882. IEEE, 2021.
Download Paper

Preprints

Technical Report: Attention Meets UAVs: A Comprehensive Evaluation of DDoS Detection in Low-Cost UAVs

Published in 2024 IEEE 20th International Conference on Automation Science and Engineering (CASE), 2024

Featured in 2024 IEEE 20th International Conference on Automation Science and Engineering (CASE), 2024

This paper addresses enhancing cybersecurity for low-cost Wi-Fi-based UAVs against DDoS attacks by developing a detection mechanism that evaluates various machine learning and deep learning algorithms, finding that the Time Series Transformer (TST) is the most effective model with an F1 score of 0.999 for TCP attacks and an overall runtime of 0.1 seconds, while also detailing steps for on-board implementation and hyperparameter optimization.

Recommended citation: Sharma, Ashish, SVSLN Surya Suhas Vaddhiparthy, Sai Usha Goparaju, Deepak Gangadharan, and Harikumar Kandath. "Attention Meets UAVs: A Comprehensive Evaluation of DDoS Detection in Low-Cost UAVs." arXiv preprint arXiv:2406.19881 (2024).
Download Paper

Technical Report: Dynamic Data Delivery Framework for Connected Vehicles via Edge Nodes with Variable Routes

Published in 2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall), 2023

Featured in 2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall), 2023

Expanded Version: This work proposes a two-stage optimization framework for efficient data delivery to connected vehicles via edge nodes, addressing dynamic route changes, and demonstrates its superiority over offline optimization and greedy approaches using a real-world dataset.

Recommended citation: Cherukara, Joseph John, SVSLN Surya Suhas Vaddhiparthy, Deepak Gangadharan, and BaekGyu Kim. "Dynamic Data Delivery Framework for Connected Vehicles via Edge Nodes with Variable Routes." In 2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall), pp. 1-7. IEEE, 2023.
Download Paper