This paper presents the concept of `Airborne Intelligent Monitoring of Infrastructures´ (AIMI), which is a work in progress at Thyssengas since about 2 years. Operators of long-distance gas pipelines are obliged to regularly monitor their pipeline grid for potential third-party interferences (TPI), e.g. construction sites, depositions of material.
Today this inspection is performed by helicopter. The reported events are often unspecific and located inaccurately, which make the inspection at the ground more time consuming. Helicopters are increasingly perceived as a nuisance to people and the environment because of their noise and CO2 emissions. Thyssengas is working on a system that combines different technologies to solve those issues and building highly automated, integrated workflow – the AIMI process.
The use of unmanned aircraft (drones) equipped with cameras will significantly reduce emissions. An Artificial Intelligence (AI) is trained to detect predefined objects on aerial images that indicates potential risks for the pipeline. Combining the cameras with advanced positioning technology allows a precise location of the AI detected events and the derivation of a corresponding georeferenced orthophoto for every potential TPI. This data quality improvement will reduce efforts of the ground staff and thus fetch internal savings. The system is also adaptable to manned aircrafts.
AIMI’s core part is the ability to transform helicopter co-pilots' expertise into "artificial intelligence". As part of the R&D project ANNeBEL – ‘Aufbau eines neuronalen Netzwerkes für die Luftbildauswertung zur autonomen Überwachung von Erdgastransportleitungen‘, the fundament was laid for the development of AI-based pipeline monitoring.
Thyssengas is using the knowledge gained here and combines it with experience in the sector of remote sensing data acquisition to provide the integrated service.
The aim is to provide an end-to-end service that enables a scalable pipeline monitoring process that can be adapted to the individual requirements of different infrastructure network operators.