Home / Blog / Slicing an onion: tougher than it looks at first glance

Slicing an onion: tougher than it looks at first glance

There is a lot of math that goes into cutting a potato. This robot had to learn it and now it is proficient in slicing onions, carrots and more… Image credit: Iowa State University, Ames

I have to confess that I never thought about the complexity of slicing a salami or a potato. I guess I have a knack for it (and my wife is usually charging me of any slicing chore…). But so are most people (if not all of them!). Slicing, chopping, cutting is so natural that you are not giving it a second thought. Yet, if you are asking a robot to do the slicing of an onion you realise how complex this seemingly simple act actually is.

I stumbled on an article published by some researchers at the Iowa State University, Ames, where they report on the work done to teach a robotic arm to slice an onion, a potato… It turns out (watch the clip) that there is plenty of math involved and a lot of signal processing as well since the robot needs to become aware of the characteristics of what it is supposed to slice, its resistance to the blade, the stickiness and of course its position and dimension.

This makes me realise what a wonderful analogue processing machine we got in our skull and how seamlessly it is processing data coming from sensors distributed over our body (from the eyes of course, but also from our fingers, hands, joints…). The computation that is going on as you cut a single slice of salami is mind-boggling. And yet we don’t perceive it at all.

This also shows that there is quite a way to go before we can have a robot that can be as flexible as we are, able to face a variety of trivial situations and act appropriately. It is probably the gap we are seeing today between artificial intelligence and general artificial intelligence. Taking a focussed view the distance is not big, if you have a machine that is not sufficiently smart to do soemthing specific you just need to work a bit on its intelligence to upgrade it to the point where it gets smart enough. Yet extending the intelligence to cover “all” trivial situations is much more complex and not yet within reach.

It would be nice to see the researchers at the Iowa University shifting from teaching the robotic arm to slice an onion to teaching it to learn, autonomously,  how to deal any kitchen chore. Of course, that might decrease my value in the my wife’s eye…

About Roberto Saracco

Roberto Saracco fell in love with technology and its implications long time ago. His background is in math and computer science. Until April 2017 he led the EIT Digital Italian Node and then was head of the Industrial Doctoral School of EIT Digital up to September 2018. Previously, up to December 2011 he was the Director of the Telecom Italia Future Centre in Venice, looking at the interplay of technology evolution, economics and society. At the turn of the century he led a World Bank-Infodev project to stimulate entrepreneurship in Latin America. He is a senior member of IEEE where he leads the Industry Advisory Board within the Future Directions Committee and co-chairs the Digital Reality Initiative. He teaches a Master course on Technology Forecasting and Market impact at the University of Trento. He has published over 100 papers in journals and magazines and 14 books.