The Kitchen of the Future


In the early days of microprocessors, people selling home computers tried to come up with reasons to have them in the home. The real reason you got one was hobby computing, but the companies wanted to push other purposes. A famous one was use in the kitchen. The computer could story your recipe file, and wonder of wonders, could change the amounts of the ingredients based on how many servings you wanted to make.

This never caught on, but computers have come a long way. But still, I mostly see nonsense applications promoted. For example, boosters of RFID tell us that our fridges will be able to track when things went in the fridge, and when it's time to buy more milk. We should give up huge amounts of privacy to figure out when to order more milk?

With that track record, I should stay away from the area, but let me propose some interesting approaches in the kitchen.

The cooking area should have a screen, of course. Screens are already in the kitchen to watch TV. While you could (and would) put digital recipes up on the screen, I imagine going further, and having TV cooking shows, where you watch a chef prepare a dish. You would be able to pause, rewind and do everything that digital video does, but the show would also come along with encoded instructions tagged to points in the video. When the recipe calls for cooking for 5 minutes, the computer would start appropriate timers.

The computer should have a speech interface, and a good one, allowing you to call out for timers, and to name ingredients and temperatures. More on that later.

The first thing I would like to see is smart, digital wireless scales in a lot of places. A general one on the counter of course, but quite possibly also built into the rack above the burner which holds the pot. You can get scales built into spoons and scoops now, and they could be bluetooth. These scales would be smart and fast. You would not have to zero them. Slap a bowl on the scale and the sudden change in weight would let the computer know you've just put a bowl down. With fast readings it can tell the difference between that and pouring something in, even really fast. Besides, over time it will know the weight of all your pans and bowls. (Of course, it will display the weight of the bowl for you in case it made a mistake, but it will also display the weight of what you add to it.)

Now you can add an ingredient and see the weight. If it knows what the ingredient is, it can also display the volume, calculated from the weight. If there is a target amount to add, it can show you with graphics and also use a rising audio tone to tell you when you are getting closer to, right on or over the target amount. If it doesn't know what the ingredient is, it can show you what volume it would be for a raft of your most common ingredients -- water like liquids, oil, flour, rice, etc.

With a nice speech interface you can say the name of the ingredient, or the target weight, or just watch the one you know it to be. But if you are following a recipe in the computer, it probably just told you, "Add 250g of corn starch" and so it is fully aware of what you are adding, and how much it should weigh.

To make this even better, the computer should have a camera looking over the stove and work surfaces. The camera should include a lower resolution thermal imager able to determine surface temperatures via deep infrared. But through colour and texture mapping, it should also be able to identify a lot of ingredients, or classes of ingredients.

It will also know the weights of ingredients you add without measuring, and be able to calculate other things based on those. Knowing the weight and dimensions of a cut of meat, and knowing the temperature of the pan and the temperature of the meat surface, as well as the colour, it might be able to determine how well cooked things are. In so many cases good cooking is knowing exactly how long to cook things, and not fussing with them while it happens. The same might apply inside ovens and grills.

Of course, you can always figure that out with a probe thermometer, so the system should have heat-tolerant bluetooth probe thermometers with thin probes that can be inserted in lots of ingredients. (I wonder if you could manage to power these off thermocouple power, if the chips inside get low power enough. The thermocouple could charge a capacitor, letting the radio send out a burst of temperature information as often as there is enough charge.)

The next sensor to add are digital chemical sensors, sometimes called digital noses. Above the stove and in the ovens, obviously, but quite possibly they could also be built into pans and pots, or be small devices dropped into dishes. In many cases, done-ness is a matter of chemistry, and certain molecules will be present in certain concentrations at the perfect point. These sensors are going to get better and better -- soon they will be better than a top chef at knowing when the food is perfectly prepared, especially combined with temperature, colour, and deep infrared imaging.

Of course, one of the most popular kitchen appliances today, the rice cooker, is popular because its ability to sense temperature and humidity lets it know exactly when rice is properly done, better than a chef can.

The computer will even be able to do some basic calorimetry, knowing the parameters of the pans, the heat output of the burners and other rules. The computer should be able to adjust and turn off the burners when the food is just right. (At the very least, even today's simple chemical sensors and burner control could stop you from ever having food get burned or boil over.)

We might even see, for the lazy chef, a simple robot arm which can turn over a container of ingredients into a dish when the time is right, allowing you to work on other things if you want. Over time we may see other robotics but that's another level. If you have a water tap at the stove, as many fancy kitchens have, its valve could be under computer control so that it is able to squirt an exact amount of water into your pan.

Ideally the programmers of the cooking show will be programming in all sorts of rules for these controls and sensors.

What about the joy of cooking? Well, I think such tools would provide a great way to learn how to cook new and very tasty dishes, and you might not need the computer's guidance after getting some practice. And they won't stop you from experimenting, in fact they may help. Even a good chef knows a thermometer is a vital tool to making really great food. I also expect such tools to appear in commercial kitchens, where the master chef sets up the program, and junior chefs follow it.

When done, our scales will even help you divide up dishes evenly. No help with the clean-up until the robots come, though.


I think the first kitchen automation devices will be complete automats for making everyday meals, like salad mixes. There was news from Japan about complete automation of making noodles - a good sign of this approach.

Then will be automats for several meals which use one plate, one slicer, one blender, etc.., and one or two robotic arms to move a meal between these.

A "killer-feature" of all kitchen automation is ability to hide actions which are boring to the cook, especially washing dishes and roasting plates (and even moving this stuff into the dish washing machine). No one needs recipe storage automation when they need to do boring things after the meal is done. On the contrary the one wishing to cook custom meal, will be prepared to face the need to do boring things like making sauses and experiment with recipes, so they don't really need automation.

The only who really need kitchen automation are the families and singles that get up, break-fast, then spend the whole day at work and return home tired and emptied. They want just to press button and receive tasty ration for the whole family.

I agree that "Automatic makers" will be a common form of kitchen appliance. Though they face a few problems. People will only buy so many of them, and they also often stop using them. Many automatic bread makers are never used. The device has to produce better product than you can get at the store or easily make yourself, and not take up to much room.

One device that has had great success is the automatic rice cooker. By giving the controller accurate sensors for temperature and humidity, it is able to make better rice than humans, who can't tell as well exactly when it's done. This in part leads me to believe that devices that tell when foods are done -- by detecting chemicals, colours, temperatures or humidity -- will have success. They don't take away the fun of cooking but do produce better food.

But cooking shows are popular, and cooking is popular, so a cooking show you watch while you cook, which pauses, and helps you measure ingredients, also makes sense. This is where I think what I describe starts, and it starts getting better with better sensors and better programming. The goal is that people are still doing the cooking though.

It is also possible that we might start to see a generalized cooking robot. Unlike the custom devices that litter the kitchen, a generalized robot which can do a lot of dishes might well be popular. However, we aren't at that level yet. Particularly since ingredient preparation is such a big part of cooking (and the less fun part) while combining and cooking is the fun part. Cooking shows, as everybody knows, regularly make a dish look very easy because there are pre-prepared bowls of all the ingredients, pre-measured, sitting on the table. The more cooking becomes like that, the more it moves towards being fun, I think.

I don't think anybody resents their rice cooker or probe thermometer being better than a human at determining when something is done.

One appliance where I'd really like to see more processing power is the microwave oven.

The first aspect of this was something I originally read several years ago in the NYTimes. Most microwavable food you buy says something useless like "Microwave on HIGH for 1-4 minutes," which is really no help at all. If a microwave oven had a barcode scanner, and an [Internet-connected] database of the associated foods, it could know how long each food item really needs at the power output of that particular oven. Scan your TV dinner, toss it in the microwave, and press Start. 3 minutes 27 seconds later, it's perfectly cooked.

Second, I want a microwave that can cook perfect popcorn every time by incorporating an audio sensor. A bag of popcorn is done "when popping slows to about 1 pop per second." There's no reason why a small microphone in the oven couldn't detect the transient spikes of popping kernels, and signal completion when they fall to this threshold.

Granted, microwave ovens aren't the realm of the gourmet, but they're already electronically-controlled, and improvements in intelligence like this could make them even more convenient for millions of people.

Microwave ovens with popcorn noise sensors have been around for ages.

Why not just outfit them with bluetooth or wifi and then sell the rest of th rig as a kit? The popcorn sensor could be a suction-cup contact mike. All you really need is a way to interface with existing teeny processor that's already in there, and pass that data back & forth to the household computer. The rest is easy enough.

I just remodeled my kitchen and it now has a 42" TV hanging on a wall. The TV is cabled to the Sat Receiver in the family room and to a PC just on the other side of the wall near a kitchen desk.

As a PC screen, we have a wireless keyboard on the countertop with a built in track pad. That works pretty well. The TV connects to the PC with two cables - a RCA to 3.5mm audio cable and a HD to DVI video cable. The TV connections are component RGB to LR.

It all works pretty well and gets a fair amount of use as both a TV and PC screen. In fact, I wish I had bought a TV that could do PIP so we can do both. The one cable I forgot to run thru the wall is a USB for video cam (for Skype, and Google video chat). That has already come up quite a bit.

Though I do think that no-hands interfaces will become important, because in the kitchen, your hands are busy and often dirty, leaving a keyboard out of the question. That doesn't stop the potential of gestural interfaces and maybe even a washable stick like a wii controller, plus some buttons at the working areas. But I think speech will be a common interface. A touchscreen that is meant to get dirty and easily cleaned could also make sense.

But many of the plans I have would have almost no user interface. If you start pouring stuff in the pot, it can just start showing you the weight, plus the volume of the common things that the camera says it could be. No need to tell the computer anything, just look at it. We're not that far from systems that could even recognize ingredients by image and chemical sensors and figure out what you're doing. You thus only need to use a UI when it gets it wrong. It sees you put down a nice red steak, and knows you make them medium rare, so it tells you when it's done that way. Worst case you just ignore it if that wasn't what you wanted to do, or change it only if it gets it wrong.

But I certainly think it could easily do things like watch you put a pot of water on the stove and set it to boil, and just know to turn the burner down after it reaches boiling, or turns it down if it sees something boiling over etc. You won't have to tell it to do that.

I love seeing retro news reels of stuff like the "Kitchen of Tomorrow"... I remember one that showed your cabinets moving up and down, but the one that seemed most interesting to me was the cabinet that washed your dishes... I remember seeing it and thinking how cool that would be. But then I realized that it would cost a fortune to do. But I kept thinking about it, and finally came up with the poor man's answer to the problem. Install 2 dishwashers.

If you did that, you basically would alternate between one or the other for your clean dishes. As you use plates, bowls, cups, etc. you simply place them into the "dirty" washer. Then, once that was full, or the "clean" washer was sufficiently empty, you'd simply put in the detergent, and push the start button. Now your "dirty" washer is your "clean" washer and vice versa.

For everyday use, this would likely work just fine. And you'd never have to empty the dishwasher or fill it. And if you are entertaining, you could clean twice as many dishes at once.

So, is this genius or madness?

Or rather they sell a dishwasher that is two independent sections which can be run independently, letting you store dishes in it. It's annoying expensive though.

Yes, but that is the point of my idea. For the cost of that 1 fancy dishwasher, I can have two, with the added capacity to boot... It's on my list for the next kitchen remodel.

Add new comment