Not requiring a cloud connection, it uses local processing and can be attached to the outside of a pipe without cutting or plumbing, said the company. It will initially be aimed at commercial and public sector property.
“We have been developing our deep tech sensor technology since 2017,” said company CEO Colin Payne. “Ahead of the formal product launch, we are engaging with customers for pilots, under commercial agreement, from a range of sectors including universities, hospitals, commercial offices, hotels and other property owners and operators.”
“We are looking forward to bringing to market a portfolio of products in due course,” he added.
The sensor uses novel mechanical engineering and deep learning to acquire, cluster and classify data, with the algorithm running on a Syntiant NDP120 ‘neural decision processor’ – which is aimed at audio processing and can run multiple CNN, RNN or fully-connected deep neural networks for <1mW.
“We are delighted to be working with InferSens to deploy cloud-free sensor solutions with our edge AI processor technology,” said Syntiant CEO Kurt Busch. “NDP120 delivers 25x the tensor throughput than our first-generation neural network.”
InferSens’ CTO Jonathan Peace is presenting the technology, and its first application, at the tinyML EMEA Innovation Forum 2022 today in Cyprus.
Earlier this year, InferSens raised $1m from investors including the RO Group, Oxford Innovation Fund and “a number of influential serial entrepreneurs”, it said.
Founded in 2017, Syntiant is based in Irvine California. It is backed Intel Capital, Microsoft M12, Applied Ventures, Robert Bosch Venture Capital, Amazon Alexa Fund and Atlantic Bridge Capital.
