Thinnect has invested together with Tallinn University of Technology over one million euro in developing next generation artificial intelligence sensor technologies, which are applied to make cities smarter and to monitor critical infrastructure. In the scope of the two year project, Thinnect engineers, together with researchers from Tallinn University of Technology developed AI algorithms trained using Machine Learning, which are for example able to detect noise types in an urban environment, classify vehicles and assess air quality. The algorithms and methods crated during the project can determine noise sources in an urban environment and monitor traffic density on individual streets. Additionally the technology can be used to detect a potential threat to critical infrastructure – electric substations by detecting electric corona discharge, which indicates a potential fault in the substation.
During the two years Artificial Intelligence capability was created within Thinnect in collaboration between researchers from Tallinn University of Technology and Thinnect engineers. This increases the value of Thinnect sensors and broadens the application areas for the sensors. Thinnect engineers obtained valuable skills in applying machine learning methods to train neural networks and in applying the neural networks on sensors based on microcontrollers.
One of the important results of the collaboration was that the sensors are low-power, which in combination with Thinnect wireless communication technology, enables to create very energy efficient sensors, which can be battery powered.
Thinnect plans to apply the newly created technology in new products, which can, using Artificial Intelligence, provide more valuable information to local municipalities and help critical infrastructure operators to ensure safe operation of the infrastructure. In addition to these applications the technology can be also applied to building monitoring and control.
The technology development and collaboration project was partially funded by the Smart Specialization applied research grant funded by the European Union European Regional Development Fund via Archimedes.