Google Cloud and Simple Technology Solutions (STS) have partnered with the US Navy to modernise the process for maintenance and repairs inspection of vessels and facilities.
Leveraging Google Cloud artificial intelligence (AI) and machine learning (ML) technologies on the images captured by the inspection drone, STS will detect, prioritise and predict its maintenance needs.
Due to its potential for technological innovation and commercialisation, this work was awarded to the company as the first phase of the Small Business Innovation Research project.
The US Navy currently spends billions of dollars annually on the maintenance and repair of its fleet and other platforms, such as facilities and aircrafts, which is majorly a labour intensive process.
STS will train Google Cloud AI and ML models on many images to identify corrosion under the first phase with the US Navy.
By using Google AI / ML technology and images, STS will aim to significantly decrease the workload and safety risks related to maintenance inspections.
The company will use the public domain and images from the inspection drone to construct an AI/ML model to permit organisations to set up custom vision models.
It will partner with the corrosion subject matter experts of the US Navy to tag and train the data using the AI Platform Data Labeling Service of Google Cloud.
It will also use custom inspection drone flight data to train and validate the model that will be processed after uploading using Google Cloud Storage.
STS chief technology officer Aaron Kilinski said: “The initial goal for Phase I is to build a model that detects corrosion in drone images with a very high degree of accuracy. The ultimate goal, however, is to move from detection to prediction by expanding the subjects and sensors, and eventually integrating with navy systems.
“We selected Google Cloud AutoML because it allows our engineers to train and test high-quality models quickly. Google Cloud provides an unrivaled degree of specification to meet tough business objectives in compliance with FedRAMP High.”