Intellegens joins new Boeing Accelerator

Intellegens, which has an AI toolkit to train deep neural networks from sparse or noisy data, has had a double boost in the run-up to Christmas.

The first cohort of the ATI Boeing Accelerator includes the Cambridge University spinout whose mission is to bolster the growth and competitiveness of the UK aerospace industry. The inaugural accelerator programme will starting in London on January 14 and running until early April.

Created in partnership by the Aerospace Technology Institute (ATI) and Boeing, the programme selected 10 startups after receiving 268 applications in a six-week period. The accelerator is designed and delivered by European accelerator Ignite. GKN Aerospace has joined as an inaugural corporate sponsor.

Selected companies include Cambridge-based Intellegens, a company with a unique deep learning toolset that can train deep neural networks from sparse or noisy data typical of experimental data; Anomalous, a Scottish company speeding up the process of aerospace inspection; and Plyable an Oxford-based company reshaping the moulding and prototyping process using machine learning and applied AI.

“Over the last 18 months we have been working on several exciting projects and really refining our product,” says Intellegens’ CEO Ben Pellegrini. “We are now applying our toolset to industrial formulations or industrial recipes – for example how do you make the perfect material, chemical or drug? In all of these cases you have raw ingredients that are mixed and processed in a certain way to deliver a new substance with very specific target properties. This process can be very expensive and time consuming so deep learning can really help guide some of this process.

“The ATI/Boeing opportunity is really exciting and looking at new materials, composites and processes for aerospace, where advanced materials are going to be crucial to deliver more efficient planes.”

Gary Elliott, CEO at the Aerospace Technology Institute, said: “We created this programme because we are looking for great technology from startups who have a different approach to innovation and will introduce a new way of thinking into aerospace and aviation.

“We are excited about working with new entrepreneurs with a great level of energy and different ways of approaching innovation.”

The selected startups will have the option to receive a £100,000 equity investment from Boeing HorizonX Ventures, first-hand access to ATI, Boeing and GKN Aerospace strategists and technical experts.

In a separate development, a collaboration between Intellegens, ANSYS Granta and the University of Birmingham has secured funding from the Faraday Battery Challenge to optimise the battery manufacturing processes helping ensure the UK leads the world in the transition to a low carbon economy.

Innovate UK, part of the UK Research and Innovation organization, is investing £246million in innovation projects over four years. The Faraday Battery Challenge: Innovation for Feasibility Studies Round 3 is aimed at supporting business-led R&D for the design and development of batteries for electric vehicles. The collaborative grant to the trio of organisations is to apply artificial intelligence solutions to predict optimum process parameters for complex interdependencies in the battery manufacturing process.

It is estimated that around 50 per cent of the value of future vehicles will be associated to the battery and the systems that support it. By 2022, the UK battery demand will be sufficient to support one UK gigafactory (large battery manufacturing plant that produces cells and modules used in electric vehicles). This presents a major global opportunity for the UK, since the complete supply chain for batteries for electrified vehicles at the scale seen in the automotive sector does not yet exist.

“The battery project is looking at processing and chemical formulations to deliver more efficient batteries, again critical for the looming growth in electric cars,” says Ben. “In both cases, we are helping guide the best recipe from limited data, which is our USP – deep learning on sparse data.

“Speeding up the production of these new battery technologies is critical in addressing the urgent need for climate action.”

Intellegens has any number of applications of its Alchemite platform, which dramatically speeds up research times. Dr Tom Whitehead, machine learning scientist at Intellegens, said: “Alchemite demonstrates real-world applicability and has the potential to provide accurate predictions for problems in drug discovery, such as finding active compounds that can counteract tropical diseases like malaria.”