For several years, I trekked around North America selling robot arms, among other products, to research labs. It was a dream job and I’m very grateful for having it. It helped shaped my views on the future and how innovation, at least from academic research, occurs.
One area that was always interesting was collaborating robots. It’s a real challenge pass things to a robot. When we pass things from one person to another, we have a kinaesthetic sense of how much weight is being picked up by the collaborator. Sensing and then acting on this, as a robot, requires very fast processing and operating at high frequencies (> 1 KHz).
Knowing the challenges is why today, if I were in a kitchen with a Samsung co-bot, like the one being displayed at CES, I’d be a bit frightened. I definitely wouldn’t give it anything sharp and I’d only stand close to it if its top speed were limited.
The idea of kitchen robots is not new. Heck, 1989’s Back To The Future 2 featured Master-Cook, a robot with a screen and two arms for helping to cook.
But maybe we’re not thinking of the right tasks?
Washing and drying vegetables is a longer tasks than cutting them up. Automatically adjusting heat on the stove top can work through mechanizing the knobs, rather than using a robot arm. Manipulators other than arms might be better at cleaning dishes (maybe inflatable octopus-like tentacles?).
The robot arm seems like an expression of linear thinking. We can stretch further.