Back to search

BIA-Brukerstyrt innovasjonsarena

ADRIANE: Autonomous Drone system for Inspection of Industrial Assets by Non-Experts

Alternative title: ADRIANE: Autonomt Drone-system for Inspeksjon av Industrielle Eiendeler

Awarded: NOK 10.6 mill.

Project Manager:

Project Number:

313652

Project Period:

2020 - 2023

Funding received from:

Organisation:

The vision for this project is to develop a fully digital process for inspection of tanks and indoor industrial assets with the help of a drone that is able to navigate safely inside of confined spaces. Every day, thousands of industrial assets are inspected manually. For example, a tank in an oil & gas refinery needs to be regularly inspected for corrosion and cracks. Likewise, a ship?s storage and ballast tanks must be inspected at least every 5th year, as required by the International Maritime Organization. Such tanks are often tall and may contain areas that are difficult to reach. It is common to use either scaffolding or a team of climbers, which are both expensive, time and labor intensive, and expose inspectors to health and safety risks, such as e.g. risk of falling, toxic gases, radiation or extreme temperatures. Scout Drone Inspection has developed a drone that is able to navigate safely in confined spaces by taking advantage of an onboard 3D laser scanner that continuously creates a 3D map of the environment and at the same time localizes the drone without the need for GPS. In this project, we will further develop the Scout inspection drone as well as accompanying software with the ambition of creating a fully digital inspection process, all the way from planning to data collection, analysis and reporting. The drone will be equipped with high-resolution cameras and 3D imaging sensors to collect inspection data, replacing the need for manual, visual inspection in dirty and dangerous environments. The system will include software for planning and carrying out inspections, taking advantage of digital twins when they are available, as well as cloud services for report generation and automatic detection of defects, such as cracks, corrosion or deformations. Eliminating the need for human entry to dangerous environments will in turn improve personnel safety, inspection quality, reduce inspection cost and have positive environmental impacts. The project has achieved new and improved methods for localization and mapping in confined spaces and new strategies have been implemented for better illumination and imaging of defects. In addition, a realistic simulator has been made that includes all sensors onboard the drone. ScoutDI's cloud-based portal has been extended with functionality for automatic analysis of inspection data based on machine learning as well as post processing of 3D LiDAR data to produce dense global 3D maps. It has also been demonstrated that the drone system is capable of carrying out inspection flights automatically based on a set of pre-planned inspection points.

-

The vision for this project is to develop a fully digital process for inspection of indoor industrial assets based on an intelligent, tethered drone system that eliminates the need for human entry to confined and often dangerous spaces. The system will encompass software for smart data capture in unknown, GPS-denied environments, automatic position-tagging of indoor inspection data, AI-based detection of anomalies as well as a cloud-based system for live video streaming and reporting. Every day, thousands of industrial assets are inspected manually to ensure proper maintenance and safety. There is a huge potential in digitalization of such inspection processes when it comes to planning, execution, analysis and reporting. Today’s regime is characterized by manual labor and hazardous working conditions, printed inspection programs and reports, as well as a lack of integration with digital representations of the inspection targets. Our ambition is to create a fully automated and end-to-end digital inspection process – from planning to data collection and analysis to reporting. An autonomous drone system equipped with precise 3D and RGB imaging sensors will be developed to collect inspection data, replacing the need for manual, visual inspection in dull, dirty and dangerous (DDD) environments. The system will include software for planning and carrying out inspections, taking advantage of digital twins (i.e. structural models) when they are available, as well as cloud services for report generation and automatic detection of possible defects, such as cracks, corrosion or deformations. Eliminating the need for human operators to enter DDD environments will in turn improve personnel safety, inspection quality, reduce inspection cost, increase asset uptime and potentially have positive environmental impacts – since improved inspection practices can reduce risk of malfunction and accidents.

Publications from Cristin

No publications found

No publications found

No publications found

No publications found

Funding scheme:

BIA-Brukerstyrt innovasjonsarena