How artificial intelligence can prevent food wastage ... and clean my floors

A couple days ago I had to throw out all my chicken because it got spoiled over the weekend while I was at my family’s place. Learning to cook hasn’t been going well if you couldn’t tell. As I tossed the chicken I asked myself, “how could this have been prevented?” What if I had never bought the chicken in the first place. Imagine that as soon as I picked up the chicken in the grocery store, my phone was able to scan the expiration date, cross reference with my calendar, identify that I was gone this weekend, and notified me not to buy it. This is a simple way in which artificial intelligence can prevent food wastage, save money, and most importantly, offload some of the decisions you have to make in your busy life.

Whenever I say I work with robots people inevitably ask me if I “do stuff with AI”. I don’t really know how to answer that because “AI” could mean a million different things. When I think of AI, I think of a machine capable of learning from its mistakes and improving its performance and decision making capabilities. In that case, no, I don’t work in AI (yet). But some people view AI as simple pattern recognition and decision trees which in that case, yes, I do “do stuff with AI”. If you haven’t googled it by now, AI stands for Artificial Intelligence which is the broad area of research around a machine’s ability to mimic human intelligence.

Smart homes is probably a buzz word you’ve heard of before. Almost everyone I know has some sort of Google Home, Amazon Alexa, or other nifty voice activated gadget to play Pandora for them. But AI has a lot of potential beyond just playing your favorite playlist. Smart cleaning is the movement towards automating a lot of the manual labor that goes into cleaning, and is the space that companies like SharkNinja are trying to capitalize on. Now you may be thinking to yourself, “I don’t have a problem sweeping the floor every now and then or washing the dishes, and I’m especially not going to pay a boatload of money for it”. Point taken, but that’s not what the rest of the world seems to think. A couple years ago everyone was skeptical of the new Roomba, and for good reason. There were reliability issues, it kept bumping into things, and it was noisy. But if you look up robotic vacuum on Amazon today you’ll find 10 different companies making the same exact model at $250 price point. And these robots are only getting smarter. Most everyday consumers are not aware of the technical details of robots, but that’s why I’m here right? Currently, robotic vacuums are split up into two primary categories. Let’s call them dumb and smart robots for simplicity. Dumb robots don’t use any sort of intelligent navigation system. They kind of just bounce around and clean… and scare your dog. Smart robots on the other hand, use cameras and/or sensor systems to actually map out your home and move around based on where it has identified obstacles and boundaries. You can connect these robots to your phone, enable them through voice activation, and even receive detailed cleaning reports. I sound like an infomercial but these technologies, although still being developed, do actually exist and are driving the smart cleaning industry.

So what is the future for smart cleaning and smart homes in general? For starters, we need to develop smarter products. Funding research and development in machine learning is a good first step for companies who want to be ahead of the curve. Right now, robots still require a lot of human intervention including manually inputting data and providing explicit instructions. In order to offload human decisions, they need to be able to think like humans, which means collecting data and making their own predictions based off patterns. If an “intelligent” robot gets stuck under a chair one day it should never get stuck under that chair again. And I should never have to throw away food again, or maybe I should just remember to check the expiration date on my chicken.