Cookie Notice

We use cookies to make parts of our website work, and to improve your visitor experience.
If you only allow necessary cookies, some features of our website may not work.

Engineering news

How AI is helping engineers invent new ideas

Amit Katwala

Smart, wi-fi enabled lighting has been available for several years now, but there’s still a giddy joy in illuminating a room with a voice command, or a tap of a smartphone app.

Products such as Phillips Hue even allow users to choose exactly which shade of light they want to use.

A new invention takes that one step further, allowing smartphone users to ‘paint’ rooms with the colours they want to see on their walls – choosing different huesand intensities – and then have the lighting system match it as closely as possible.

But perhaps the most fascinating thing about this invention – this lightbulb moment – is that it was inspired by artificial intelligence. Swiss technology company Iprova are taking what they call a data-driven approach to research and development, and using artificial intelligence to come up with new ideas.

“Traditionally, R&D teams tasked with developing inventions consist of individuals with particular areas of specialist knowledge,” explains Iprova CEO and founder Julian Nolan. “This gives individuals comprehensive insight into specific technologies, but less in-depth information about things outside of that area.”

Augmenting human intelligence

His company’s technology draws on a wide range of information and “uses machine learning and other technologies to augment human intelligence,” says Nolan. He argues that bringing in advances from other unrelated domains can help engineers come up with ideas more quickly, and bring together technologies from seemingly distant areas.

For example, Iprova were able to connect advances in autonomous vehicle technology with a problem that needed to be solved in personal healthcare. Their machine learning algorithms spotted that the advanced sensors built into an autonomous vehicle could be used to take measurements from its human passenger.

By controlling small variations in its motion, the vehicle can generate precise forces that can be used to carry out human health checks. This technology could, for example, be used to assess passenger’s balance or core stability, or evaluate the body’s response to certain stimuli. Other AI-aided inventions include a way of incorporating gesture recognition more easily by using a smartphone’s existing display, and a heat-sensitive coating that can allow users to make sure they’re wearing a protective face-mask correctly.

Futurist Maurice Conti says artificial intelligence is moving from the ‘generative’ stage to the ‘intuitive’. In a TED talk, he highlights the work of Airbus, which has used generative AI to help design lighter and better parts for concept planes.

But this is the next step – as deep-learning systems start to develop intuition, and come up with ideas that their inventors couldn’t have. “Very soon, you’ll literally be able to show something you’ve made, you’ve designed to a computer, and it will look at it and say, “Sorry, homie, that’ll never work. You have to try again,”” says Conti. “Or you could ask it if people are going to like your next song, or your next flavour of ice cream. Or, much more important, you could work with a computer to solve a problem that we’ve never faced before.”

The end of the inventor?

He’s worked on a number of projects, including a 3D-printed bridge in Amsterdam, and a car chassis designed based on data collected from a race-car with a nervous system. “We instrumented it with dozens of sensors, put a world-class driver behind the wheel, took it out to the desert and drove the hell out of it for a week,” Conti explains. 

“And the car's nervous system captured everything that was happening to the car. We captured four billion data points; all of the forces that it was subjected to. And then we did something crazy. We took all of that data, and plugged it into a generative-design AI we call "Dreamcatcher.’” This was the result – a car chassis designed using machine learning and artificial intelligence, based on data instead of human input.

(Credit: TED/Maurice Conti)

(Credit: TED/Maurice ContI)

So does this spell the end for would be Edisons? Is the humble inventor just another job that’s set to be replaced by machines? Some certainly think so. “The use of artificial intelligence to create patentable inventions is the next stage in the evolution of innovation,” says Peter Finnie, managing partner at intellectual property firm Gill, Jennings & Every.

Nolan says that the explosion of information and increasing levels of convergence are making life increasingly difficult for companies to envisage future inventions. “As both the amount of information and the level of convergence are increasing at an ever-accelerating pace, it is possible that data-driven invention will become critical for the success of both product and service-based companies in the coming years.”

Conti talks about technology “amplifying our cognitive abilities so we can imagine and design things that were simply out of our reach as plain-old un-augmented humans.”   

Content published by Professional Engineering does not necessarily represent the views of the Institution of Mechanical Engineers.


Read more related articles

Professional Engineering magazine

Professional Engineering app

  • Industry features and content
  • Engineering and Institution news
  • News and features exclusive to app users

Download our Professional Engineering app

Professional Engineering newsletter

A weekly round-up of the most popular and topical stories featured on our website, so you won't miss anything

Subscribe to Professional Engineering newsletter

Opt into your industry sector newsletter

Related articles