It’s been a dream of humanity to develop a legion of mechanical slaves to do their bidding for generations. And to a large extent, that’s what we’ve done. The reason only about 2 percent of people work on farms is that machines do the work of the other 98 percent, burning vast amounts of energy in the process.
But when we talk about robots, we have something different in mind to combine harvesters. We’re talking about things that look and move like use: general purpose devices that will do our bidding, whatever that happens to be. A combine harvester might be great at scooping up corn, but it’s probably not all that great at making you a salad.
In this article, we’re going to look at some of the things that are stopping robots from being our personal assistants and when we can expect technological breakthroughs to make it possible.
You can think of servos as the equivalent of robot muscles. They’re little motors in robot joints that translate rotational energy into linear movement. The problem is that, compared to biology, they’re poor. Although they have a lot of strength, they use a lot of energy and don’t have the finesse or ability to spin up and down quickly enough to provide robots with real-world balance.
Improvements in software and panel design, however, is changing the status quo. Servos will likely never be as good as muscle tissue, but as companies like Boston Dynamics are making increasingly clear, they are improving. Servos have now advanced to the point where they can provide stabilization for bi-pedal robots, something that looked hopeless just a few years ago.
Google has a room at an undisclosed facility where it is testing robot arms. All of the arms are lined up in a row, attempting to grab objects in boxes in front of them in slightly different ways. Most of the time they fail, but when they succeed, they update smart software which integrates how the arm did it into its software so that it knows what to do next time.
There are, naturally, a lot of failed attempts. But every so often, one of the arms succeeds in picking up one of the irregularly shaped objects and depositing it into a nearby container. When this happens, the instructions that led to that success are shared with all of the other robots, improving their picking ability.
Google will be the first to admit that the technology has a long way to go. But in under a decade, we’re like to see robots like this on production lines and in fulfillment centers picking things up and depositing them elsewhere, so human workers don’t have to.
The trick to life-like robots is to create artificial muscles that behave like their natural counterparts. Researchers have already built synthetic muscle in the lab which contracts when an electrical current is passed through it. Once this technology matures, it will help some robots more naturally mimic the movements of people – movements that we’ve evolved to inhabit planet Earth.
If you like what you’ve read here, please let others know of this post, blog, and site.
And thanks for reading! 🙂