The system showcases connected tech and digital transformation put to proper use
Pursuing its desire to inject artificial intelligence (AI) into digital transformation, Fujitsu has revealed how its latest AI engine can be used to detect potholes in roads before they appear.
For anyone used to driving on British A and B roads, that will be music to their ears, but Fujitsu’s AI tech was put to use by Japanese company Kawasaki Geological Engineering to develop a system that can detect subsurface cavity survey technology.
Pothole seeking AI
Through the use of Fujitsu’s deep learning platform, the complex sounding Cloud Service K5 Zinrai Platform Service Zinrai Deep Learning, Kawasaki Geological Engineering uses the AI tech to process and analyse large volumes of radar imagery collected from underground probes to search for cavities before they form surface potholes.
According to Fujitsu, this system will bypass the need for expert technicians to take up time making visual evaluations of underground cavities.
Furthermore, the system can be connected to Fujitsu’s Traffic & Road Data Service to display problematic spots on a map so that the road authorities can fix the problem before it becomes a fully blown pothole. Essentially, the system uses AI tech to provide predictive maintenance for roads, rather than leave it in the domain of the manufacturing world.
Through the use of deep learning techniques, which use an artificial neural network to disseminate data in a fashion not dissimilar to how a human brain works with information, the AI system will be trained to spot changes in radar reflection, displayed as waveforms, in underground images and determine if what it is seeing are cavities or sewage pipes.
While the system has yet to emerge fully from a trial period, it helps to demonstrate how the combination of connected sensors, big data and AI technology can help transform an organisations processes with efficiency and easing workloads in mind.