Achieving Sensor Fusion for Collaborative Multi-level Monitoring of Pipeline Infrastructures

TL;DR

This paper presents a multi-level sensor fusion system combining wireless sensor networks and UAVs for pipeline monitoring, enhancing data accuracy and reducing communication bottlenecks through Kalman filtering, FUSVAF, and consensus methods.

Contribution

It introduces a novel multi-level architecture integrating sensor fusion techniques for efficient pipeline infrastructure monitoring.

Findings

Kalman filtering improves data accuracy in sensor fusion.

FUSVAF effectively validates sensor data in complex environments.

Consensus-based processing reduces communication load.

Abstract

Large scale monitoring systems enable efficient field level data collection at high temporal and spatial resolutions. One example is the deployment of such systems in pipeline infrastructure applications which have to be monitored for leaks and protected from unauthorized access, with the potential of causing significant environmental and economic damage. The paper discusses a multi-level system architecture for data collection and processing based on the collaborative integration of wireless sensor networks and unmanned aerial vehicles. Three sensor fusion methods: Kalman filtering, Fuzzy Sensor Validation and Fusion (FUSVAF) and Consensus-based processing, are considered for intelligent data reduction and situational awareness while alleviating communication bottlenecks across the multi-level network. Simulation and experimental results are presented.