• Notice

    We update New content Daily Just Stay With us

    A Closer Look at the Robots of Boston Dynamics!


    • News often sensationalizes new technologies and they really haven't progressed as far as we may give them credit for.
    • There's no need to fear robots like Atlas and SpotMini. They are a long way away from being able to take over the world according to experts.


    The field of robotics is constantly expanding and improving upon techniques learned from previous design concepts. We have seen a lot of hoopla coming out of this field in the past few years with names like Boston Dynamics, Honda and Toyota often making the news for the latest and greatest in robotic technologies.
    Personally I was amazed at some of the robots that made the news in 2017. I was most impressed with Toyota's T-HR3 and Boston Dynamics' Atlas robot. The thing about Boston Dynamics that really impresses me is the speed at which they are coming out with new models and the features each new robot is equipped with!
    Of course when Boston Dynamics hits the news with a new robot it causes a viral buzz on social media every time it posts a new video of one of its robots moving around the lab.
    Then people start screaming that its the first sign of the inevitable robo-pocalypse! Fortunately, with a little advice from experts, you can see much of that panic is just based on the hype and not the facts. The bottom line is that Boston Dynamics is doing outstanding work in the field of robotics and in this article I'm going to try to seperate some of the hype from the facts.
    So lets get started here shall we? Firstly, if you want to behave like a robotics expert when you watch one of these new release robot videos, you'll need to become more critical. If you look a bit deeper, you'll notice that Boston Dynamics is a private company and not an academic entity. You'd be surprised how rare it is to actually see the company publish its findings.
    This makes it hard to know what's really going on inside the robots. "We have an idea about what approaches they are using" says Ioannis Havoutis, a researcher in robotics focusing on leg locomotion at the Oxford Robotics Institute, "but apart from a few papers, we can only guess what they are doing."

    Ok, so far we have learned that things aren't always what they seem and that robot hype and robot fact may not go hand in hand. One reason we shouldn't panic about what we are seeing is because for robots to actually gain proficiency in most fields they will have to be able to overcome complex challenges with little margin for error. So if we start by understanding the calculations and margins involved in the robots' activities its easy to see that Boston Dynamics' robots have a larger margin of error than most robots are given.
    "Boston Dynamics do not worry about sub-millimetre accuracy, they worry about the functional accuracy," says Thrishantha Nanayakkara, reader in design engineering and robotics at Imperial College London. "[Atlas] is metastable, so it's stable most of the time. There is a probability that it can go wrong, and they take that chance. Most robots we know in the industry don't take that chance." Being metastable means Atlas has to balance itself to stay upright, just like a human.
    But even Atlas' backflip only requires "a very crude calculation to make the jump." "Then when it lands, it makes the corrections. It doesn't have to be perfect, just good enough."
    So we can see that BD is kind of cheating right? Boston Dynamics' has many experiments involving movement. This can include irregular or tricky terrain like grass or snow. As a result robots can be designed with an even greater margin.
    "There is no need for precision there," says Nanayakkara. "The important thing is to finish the calculation within the deadline."
    In the case of BigDog, one of the firm's earlier machines, the calculation is working out how to stay upright, and the deadline is the moment when it can no longer right itself and then falls over.
    Well, BigDog is certainly impressive! In robotics terms having enough time can be a lot different than human terms. Keep in mind calculations can be made in milliseconds!


    Another thing that should be stated is that most of the things we see Boston Dynamic's robots doing in video aren't completely autonomous! Stating this even more clearly, Boston Dynamics' robots aren't usually completing these actions alone, they are usually getting outside help in the videos we see.
    According to experts, it is no secret that human help is still very much needed. "It's mostly done by humans and remote control, but there is onboard computing happening," Nanayakkara explains. "There are some very nice controllers, but it has some assistance from remote interference. There is a human in the loop, but there is an element of autonomy."
    Havoutis seems to mainly agree with the statement above. "These robots are not autonomous in the sense that we are autonomous. They follow a well-defined set of routines, they don't have the ability to decide on their own what to do.
    "You can see in the SpotMini video that the robot can make out where the handle of the door is, but it's been commanded to go through that door or there's another level of planning that sets its course of action."
    The day that these robots no longer need to be told what to do is a day we should probably worry about! But for now most actions are scripted and therefore human intervention is still very much required!

    So why to people get so scared when they watch these Boston Dynamics robo-sensational videos? Well its probably human nature! Humans are always trying to personify robots. "As humans we tend to think of them as 'live' things instead of machines because they look familiar; Atlas looks like a person, and SpotMini looks like a dog," Havoutis says.
    Havoutis expanded on this idea by referring to a video featuring Atlas trying to pick up a box. While the robot tried to pick up the box, a Boston Dyamics employee moved it out of reach and pushed the robot with a hockey stick.
    "A lot of people thought that this was cruel and inhumane. It looks kind of disturbing because we've associated the human form with humans but it's a robot doing what it's programmed and designed for. It doesn't feel sad when the box slips away!"
    Nanayakkara believes there will always be room for robots and humans to coexist. Even if robots take over a large proportion of manufacturing jobs, this will likely elevate the value of humans. "Human value is not in the body, it is in the mind," he says.
    I definitely agree with the statement above but have to admit that downplaying humanity's role in directing technology is unwise. We must be active in both our research and the actions we take as a result of learning where things like AI, gene editing and robotics are going. Are they going in a direction that we want to see them go? If so, then great... But if we see signs of foul play we must unite and speak out against them. We must act.
    Havoutis takes time to discuss the fear that many humans have when it comes to the prospect of robots becoming increasingly under the control of "evil" individuals or groups. "These can be used in a bad manner, but it's the same thing with every other technology. It can be used for good and it can be used for bad. There are a lot of positives that we can expect from developing robotic technology."
    For Nanayakkara, Boston Dynamics is pulling off "a trick" when it comes to showing off the competency of its machines. "They are picking up very difficult problems even for humans to deal with and then demonstrating the capability of robots to deal with this. It shouldn't be generalised to everything. We understand how to solve very complex dynamic problems, but it doesn't mean we understand the mind."
    I agree with Nanayakkara for the most part and knew that the videos only show a small part of a greater picture. However, based on what I have seen so far I do also believe their is cause for concern. Things are progressing far faster then I ever believed possible.

    I hope this article has put some potential worries at ease and has shown another side to this sensational trend of showing off brand new robots for the masses to ooooh and ahhhh at!
    When the next Boston Dynamics video goes viral, hopefully my readers will be able to watch it more critically and ask questions like 'have these results been published in any academic papers yet?'
    Certainly admiring the skilled engineering that goes into these robots is fair play. They are quite impressive after all. However there is always a chance things can go wrong. Most of these robots couldn't do what they do alone. We may however see more employment woes when Boston Dynamics' machines leave the lab and seek their fortune in the outside world but we will just have to face that fact when and if we ever get to that point.
    Thanks for reading.


    OLFA-Shoot4Spins