Presentation Details
  1. Home
  2. Presentation List
  3. Presentation Details
Agentic Sensor Ecosystems for Real-Time Infrastructure Awareness
ID：153 View Protection：ATTENDEE Updated Time：2026-07-27 13:15:17 Hits：5 Online

Start Time：2026-08-01 11:40（Asia/Kolkata）

Duration：15min

Session：S8 Mixed Track Session » S8-1Mixed Track Session

Presentation File

Tips: Only the registered participant can access the file. Please sign in first.

Abstract

Abstract—Advanced civil and urban infrastructures such as bridges, buildings, roads, and public utilities are dependent on dense networks of cameras, heterogeneous sensors, and drones for surveillance and timely maintenance. Most of the implemented supervision systems remained as inactive data pipelines. They can just collect and transmit the data to a unified system, and they are incapable of reasoning, negotiation, and flexible sensing and computation in dynamic network environments. This article introduced a framework of an autonomous software agent that is embedded with sensing, cloud tiers, and edge frameworks for monitoring networks, facilitating a live, self-organizing, situational-aware, rational-aware architecture called Agentic Sensor Ecosystem (ASE). The Bayesian state-fusion mechanism is used for enhancing the dependability of the sensor, edge-cloud offloading and sharing of the digital twin, and the distributed anomalies are aligned into a single framework. Mathematical parameters such as offloading probability, twin-state fusion, weighted multi-agent consensus, and system-level awareness scoring are proposed and integrated in a functioning model with 1600 sensor nodes along bridges, smart roads, and building implementations. The proposed framework accomplished 96.2% fault detection accuracy with an average end-to-end latency of 118 ms with 800 perpendicular nodes, which shows enhanced performance in cloud-only, edge-only, and federated-learning baselines by 5.5-14.1% for accuracy and reduces latency up to 68%. This suggests that agentic coordination is a practical solution for a scalable, reduced-latency framework awareness.
 
Keywords
agentic artificial intelligence; digital twin; edge–cloud computing; Internet of Things; infrastructure monitoring; multi-agent systems; sensor fusion; structural health monitoring; real-time situational awareness
Speaker
Neeththi Aadithiya B
ASSISTANT PROFESSOR M.kumarasamy college of engineering

Submission Author
ANIL SHARMA IES College of Technology, IES University, Bhopal
Siva Skandha S Department of CSE, CMR College of Engineering & Technology, Hyderabad, Telangana, India
Anish Gupta School of Engineering and Technology (SET), CGC University, Mohali
KOLLU DOLAKNATHCHAKRI Shri Venkateshwara Padmavathy Engineering College, Chennai
DEVASENAPATHY K Karpagam Academy of Higher Education Coimbatore
Neeththi Aadithiya B M.kumarasamy college of engineering
Submit Comment
Verify Code Change Another
All Comments
Important Date

  • Conference Date

    Jul 30

    2026

    to

    Aug 01

    2026

  • Jul 26 2026

    Draft paper submission deadline

  • Jul 28 2026

    Registration deadline

Sponsored By
The United Societies of Science
Organized By
Kongunadu College of Engineering and Technology
Supported By
IEEE Section
IEEE Madras Section
Contact Information
Login to view full contact details
Previous Conferences