What I find so interesting about this little guy is this guy can BALANCE! Good heavens! Who would have thought? This is so fascinating. And he does this by just the handlebars. Just like us humans. Wow! And pedals the bike with his own feet. Oh my goodness.
There have been robots riding bicycles for some time now, created by Murata, but right now, they need giant gyroscopes and thick wheels to keep themselves upright. This little guys just needs his handlebars. But he is small, hardly a foot tall, so maybe that has something to do with it.
Japanese robot creator Masahiko Yamaguchi is the gentleman who created this little guy. He did this to see if it was possible to build a robot that would incorporate skills into the artificial intelligence. His thinking is that this is the way to make robots independent so they can go about completing their assigned tasks without needing a human to hang over them all the time. This will free up people to do the jobs only humans can do. Now it’s not to say there aren’t any robots today that can do this because there are. Many of them. But this little guy is geared toward improving all that.
But so far, this robot needs a remote control to help him along, tell him where to go. This little robot is only the beginning of Mr. Yamaguchi’s experiments. He made his own control board, which is in the robot’s backpack, using the SH7125 CPU core. PID (proportional–integral–derivative) is the usual control method. It’s used to calculate how far to turn the handlebars when the bike wants to tilt over. But he intends to find a way to build the robot so it will be intelligent enough to ride by itself. He’s calling this one the KHR3HV bipedal robot. Or ‘primer v2’. ( so far, you can buy him for about $2200). Also, this tiny bike doesn’t have brakes. The back wheel stops when the pedaling stops, which helps the robot stop the bike, but the little bot has to also put his toes to the floor to brake. This also helps him keep upright while he is stopped. Check out the video link on our website home page to see what I am talking about.
Mr. Yamaguchi used a mass produced, two-legged robot which he modified with special joints. A small gyroscope in the robot’s body, takes note of how the robot’s body tilts, which is used then to calculate how far to turn the handlebars so it can remain upright while moving. There is a control unit in the little bot’s backpack that Mr. Yamaguchi assembled to ensure high processing speed. He designed this little bot to explore possibilities in technology to mimic human skills. And as it happens, his project has drawn attention to the complexities of the mental processes that are involved in doing things we humans find quite simple. Such as stopping a bike manually by simply dragging our feet. In less than a nanosecond, our brains calculate the angle between the ground and our legs and hips, and the velocity we were going, before we can properly stop the bike. Mr. Yamaguchi wants to build a robot that will do that, too. If he can, there would be an infinite number of applications where this could be used.
Now you may be thinking the computers we already have surely do calculate things far faster than we do. But they don’t. We have to program our computers first before they can calculate anything, and have to include every single bit of information that it takes to calculate anything correctly. That takes a whole lot of work. Far more than we are capable of right now, because we don’t know all there is to know about our brains. And it is our brains we need to know completely before we can build robots that will mimic us humans correctly.
Mr. Yamaguchi believes intelligence and skills have equal value. He wants to build robots that link them together. And in so doing, because robots can be more precise than we humans can, there can be far better improvements for the ones already being used in hospitals, in labs, in law enforcement (defusing bombs, for instance), in our military, etc., etc., etc. Who knows where it will go? Personally, I am looking forward to seeing that. But let me ask you something. Would you enter a robot cycler into a competition? Tell us what applications you can foresee.