When I was in middle school I saw this robot documented in discover magazine. I was absolutely fascinated. I had been interested in mechanical design, and the concept of spherical motion transfer and and control totally blew my mind. It was that article that led me to work in robotics. Thanks to those that worked on it.
I'm amazed to only now get this in my recommendations yet this concept is already quite old and I haven't seen an update on it in 5 years. So with the new availability of electronics and microcontrollers, I'm guessing this should be even easier to engineer in our current day and age. The question then arises, why haven't we seen this design being adapted and improved upon?
Dylan Klavins (our former mechanical engineering lead) directed the design and construction of a robot using some of the concepts from this video's design at our high school robotics club a couple years ago. It is not an actively stabilized tall robot like the one in the video, but rather is a lower, faster robot with four balls designed for fast, holonomic navigation over difficult terrain. Here is a video showcasing it: ruclips.net/video/j09PzmVmzJw/видео.html
@@AlnisSmidchens Awesome machine!! Sadly the site with the CAD files doesn't exist anymore, I would love to know more about the construction, I immediately have a lot of creative ideas.
Very cool, I remember building this thing in Uni. I wonder why there hasn't been any notable further development incorporating this Ballbot's concept?? One problem I bet is maybe it's power hungry, actuating the motors constantly in an effort to stabilize the structure due to its inherent unstable idle position. Thus leading to not being economically feasible or user friendly, cuz you gotta charge the thing all the time or need a huge battery :(
that sway in the jaw-axis when moving probably is just a oscillation in the control loop but it adds character. It gives the appearance of a walking (or skating) motion that to some extent anthropomorphizes the machine.
I kept telling boston dynamics that they needed to put a little tail on their robot dog and it would increase sales but they just couldn't wrap their head around it because the tail had no function.....Engineers. But I agree with you completely. And this thing does it as part of it's function which is an happy accident.
Considering the shape (height) of this robot, it seems like it would be advantageous for it to have a solenoid-shiftable mass near its top, to give it the ability to lean just a little. In its current configuration, it seems like it could quickly destabilize and fall if it were to run inti anything (or is pushed into anything). Having a secondary means to rebalance itself in a worst-case scenario seems important, and would also open up quicker accel/decel, higher movement velocity, and potentially better ability to handle uneven terrain. All the balancing stuff is really cool and great tech, but it’s 2020 now and lots of people have been building single-ball balancing robots and they do well up to a point, until they topple over. A secondary balance mechanism could be the missing component that pushes this from cool concept to viable product.
I think 3 ball at bottom will be much better and more stable so balancing would be much easier and then you also can't move it by small force. And also you can move in any direction without any problem.and you can also rotate the robot on any axis without moving
I think I'd much rather such robot over a human looking one they showed at the start but then I'm curious about how safe it is what if there is a power malfunction, what if someone bump into it too hard/fast, will it just fall in an uncontrolled manner? It's probably not light at that too
OK so this obviously could make for very agile ball joints in robotics, a hip joint with good stable sway could make bipedal walking a lot easier. A binding mechanism could be a simple cup with 3 of the same wheels but free-rolling, with an axle extending from and tentioned to the driving assembly. With it bound from each side and basically a universal joint at that point a computer can calculate the needed input to the ball wheels to drive an appendage in the right direction.
Interesting & Innovative home & workplace technology: the Ballbot - A Single-Wheeled Balancing Robot features Ralph Hollis, George Kantor, Umashankar Nagarajan and Michael Shomin explaining how Ballbot might be able to help people and why it stands out as a better alternative to robots with heavy, bulky multi-wheeled bases.
Sigh...I was inventing that earlier today and then found the video. Anyway, nicely done. I'd honestly feel safer having a robot with a low center of mass and a wider base so it can't fall over.
Ty Willis Basically yes, but here is the rub. „Wide bases make robots unsuitable to navigate cluttered human environments.“ But if a balancing robot runs into any clutter on the floor, trying to balance (can be a lego brick), it will fall over. Possibly onto a child or frail elderly, with all its 75kg...
Here is an idea it could be completely useless though. Why not use a gyroscope to control the lean of the robot that should allow a lot more fluidity in the motion and a lot more control also it could also allow the robot to move faster.
me as a technician: "how does it work internally?" me as an engineer: "what happens if: -something is on the ground and block the ball? -someone push the robot in the direction of lowering stairs? -the battery fail? -we combine many of thoose robots to make a wall? -the ball doesn't touch the ground anymore? -the motor control does not respond? -the robot goes crazy (internal software crash / electronical short-cut / power-surge due to battery fail / ...) ? -someone leans too much weight on the robot? -someone is too heavy for the robot to lift? -the robot vision is disturbed due to reflective environement or capacitive environement? ..."
Anybody who has played with some '2D' balancing robots (like Segway) knows that things go a bit crazy in such cases indeed. Maybe better control software can make it better. But I do not really see a big need for balancing on a ball. You could easily make a small stable low wheeled platform and then put an active approximately-balanced high vertical robot on top of it. This makes the balancing much more forgiving because the platform can still provide some extra torques where needed. Compared to a human we put our whole foot to the floor and add balancing because we are very tall compared to our foot size. The ball-robot is more like balancing on the tip of your toe. I do like the high narrow robot concept a lot though.
i love ball bots, but the name is a bit too humorous for soliciting the emotions that are required for appealing to serious minded investor types. what i mean is that the colloquial 'ball' if often used to mean 'testicle' of the male of any species, especially human males. So this name alludes to the collquial for 'testicle bot' . In this case, the robot is then thought to have 'one' ball, as opposed to ' two balls' or a 'pair' or 'pair of nuts' which , is often used to refer to the number of testicles of a human male. And , of course, this doesn't even begin to approach the allusion to the human penis , a topic of much humorous discussion. That would be bad for getting money to build this robot. Essentially I'm saying that , on a practical level, calling this robot a 'ball bot' is going to elicit a lot of snickering from young people and people who laugh at gonad humor. They may even start calling it 'dick bot' or a 'penis bot' or a 'cock bot' . This would not be good for soliciting sales or investment that is always necessary to help accelerate the development of any fine robot. I would like to see this robot become more popular. So, please call it something different. For example, call it a 'sphere bot' or a 'globe bot' or a bowlingbot , anything but a 'ball bot' . This would likely help you actually get (more) money from investors.
Why build a robot to bring u things from the fridge or show u where the fridge is.. That's why guys get married :) Don't even need to ask anything. Automatically knows when the bottle is empty, bring another haha
Anthony Dach But remember, for this robot to be of any use it will eventually carry an unbalanced load, which will take a whole lot more energy to keep steady. Jets are actually very stable, and most have a fly-by-wire(or in this case fly-by-liquid) system if something goes wrong(say, heading/altitude hold). And again - your attempt at refuting my argument fails because you suggest that this system will be stable most of the time - which it won't. Unstable systems do have a much faster response - but at what cost? Nobody's going to need, say, a patient-observing robot to dash to someone when they press the emergency button - a camera with an epi-pen or two won't do you any good. Nor will this transport any large payload - it's simply too difficult to balance. Also, what do you mean by "reducing power consumption when moving"? Yes, leaning will get you some momentum, but that momentum will be that of an inverted pendulum - to crash to the ground, a stable position. You'll have to work the motor continuously, both forward(accelerating) and back(braking) to keep you from either falling to the ground or stopping completely due to an incorrect control loop. I do know what control theory is, so please, don't underestimate my chair-to-chair combat skills.
Anthony Dach If it's unstable, there's absolutely no way it can use the same amount of energy, as I've said, the center of gravity will be offset, and so will be the wheel, and the rest of the weight, and so on, and it keeps getting more unstable the more you add on. Right now I still don't see an advantage to a very unstable inverted pendulum system with absolutely no load capacity and a constant need to use power to be stable(vs. say, a carefully selected wheel platform with less friction and latching brakes).
Because it's a friggin robot and it's balancing on a wheel like a ballerina, and it's NOT unstable. You have no understanding of the beauty of engineering, and lack the creative imagination to see how this could be applied in other areas.
Thanks Ralph! Don't know why they put "nagarajan" in that video. Were they afraid of somebody saying "racist" if didn't put a brown piece of shite in that video?
![Blox Fruits Dragon Rework Update [Full Stream]](http://i.ytimg.com/vi/EqDAp8udhm0/mqdefault.jpg)
When I was in middle school I saw this robot documented in discover magazine. I was absolutely fascinated. I had been interested in mechanical design, and the concept of spherical motion transfer and and control totally blew my mind. It was that article that led me to work in robotics. Thanks to those that worked on it.
I'm amazed to only now get this in my recommendations yet this concept is already quite old and I haven't seen an update on it in 5 years. So with the new availability of electronics and microcontrollers, I'm guessing this should be even easier to engineer in our current day and age. The question then arises, why haven't we seen this design being adapted and improved upon?
Marinus Because it can’t stop suddenly ?
@@edh2246 That's really a non-argument.
Dylan Klavins (our former mechanical engineering lead) directed the design and construction of a robot using some of the concepts from this video's design at our high school robotics club a couple years ago. It is not an actively stabilized tall robot like the one in the video, but rather is a lower, faster robot with four balls designed for fast, holonomic navigation over difficult terrain. Here is a video showcasing it: ruclips.net/video/j09PzmVmzJw/видео.html
@@AlnisSmidchens Awesome machine!! Sadly the site with the CAD files doesn't exist anymore, I would love to know more about the construction, I immediately have a lot of creative ideas.
Very cool, I remember building this thing in Uni. I wonder why there hasn't been any notable further development incorporating this Ballbot's concept?? One problem I bet is maybe it's power hungry, actuating the motors constantly in an effort to stabilize the structure due to its inherent unstable idle position. Thus leading to not being economically feasible or user friendly, cuz you gotta charge the thing all the time or need a huge battery :(
that sway in the jaw-axis when moving probably is just a oscillation in the control loop but it adds character.
It gives the appearance of a walking (or skating) motion that to some extent anthropomorphizes the machine.
I kept telling boston dynamics that they needed to put a little tail on their robot dog and it would increase sales but they just couldn't wrap their head around it because the tail had no function.....Engineers. But I agree with you completely. And this thing does it as part of it's function which is an happy accident.
Considering the shape (height) of this robot, it seems like it would be advantageous for it to have a solenoid-shiftable mass near its top, to give it the ability to lean just a little. In its current configuration, it seems like it could quickly destabilize and fall if it were to run inti anything (or is pushed into anything). Having a secondary means to rebalance itself in a worst-case scenario seems important, and would also open up quicker accel/decel, higher movement velocity, and potentially better ability to handle uneven terrain.
All the balancing stuff is really cool and great tech, but it’s 2020 now and lots of people have been building single-ball balancing robots and they do well up to a point, until they topple over. A secondary balance mechanism could be the missing component that pushes this from cool concept to viable product.
I think 3 ball at bottom will be much better and more stable so balancing would be much easier and then you also can't move it by small force. And also you can move in any direction without any problem.and you can also rotate the robot on any axis without moving
Wonder how it deals with puddles, items on floor like a dropped spoon,pencil, Or a lego block?
Very interesting level of ability. I think the future will be very exciting.
Does it also use gyroscopes or other trickery? Or is it just rotating the ball and moving its arms?
2:15 You mean directly ABOVE.
I wonder if the omni-wheel would work as good or less on this robot. Also couldn't control moment gyroscopes be used to balance the robot?
I think I'd much rather such robot over a human looking one they showed at the start
but then I'm curious about how safe it is
what if there is a power malfunction, what if someone bump into it too hard/fast, will it just fall in an uncontrolled manner? It's probably not light at that too
OK so this obviously could make for very agile ball joints in robotics, a hip joint with good stable sway could make bipedal walking a lot easier. A binding mechanism could be a simple cup with 3 of the same wheels but free-rolling, with an axle extending from and tentioned to the driving assembly. With it bound from each side and basically a universal joint at that point a computer can calculate the needed input to the ball wheels to drive an appendage in the right direction.
I will build one of these that you can ride on think of it as an omnidirectional autonomous Segway
Hey, so it's been 4 years, did you made it ?
Alten Mechatronics has a similar research robot here in the Netherlands. really elegant approach to locomotion. like!
6:18 - it's called a trolley and doesn't cost much
"it can help you pick it up." What if it is too heavy to lift onto a trolley?
Interesting & Innovative home & workplace technology: the Ballbot - A Single-Wheeled Balancing Robot features Ralph Hollis, George Kantor, Umashankar Nagarajan and Michael Shomin explaining how Ballbot might be able to help people and why it stands out as a better alternative to robots with heavy, bulky multi-wheeled bases.
You make Dr. Michael Mauldin (We call him "Fuzzy") proud!
What if we put one more ball next to current one?
Would be great with a screen to display another persons face for live face time etc. Also track lines to help it with the stairs
Its great and it looks as though it is floating.
Amazing control system and engineering but too complex for its applications, what if it falls?
So now we have skynet and i-robot vehicles. Wht next?
Sigh...I was inventing that earlier today and then found the video. Anyway, nicely done. I'd honestly feel safer having a robot with a low center of mass and a wider base so it can't fall over.
This music is straight out of the 80s.
I love this! Funny how it's just now behind recommended to me... 4yrs later😒😒😒
Nice design
Imagine seeing this on a lonely night with thunder and lightning...
Thats cool. I want a Ballbot.
I love this idea!
Danger Will Robinson Danger!
interesting design by itself, when trying to argument about applications, that just doesn't click.
I always wondered what a bolbod looks like
inspiring stuff i love ballbots
Lets assume floors are perfectly flat and free of obstructions...
You mean like in most hospitals and (a decent amount of) schools?
Ty Willis Basically yes, but here is the rub.
„Wide bases make robots unsuitable to navigate cluttered human environments.“
But if a balancing robot runs into any clutter on the floor, trying to balance (can be a lego brick), it will fall over. Possibly onto a child or frail elderly, with all its 75kg...
Why does the face remind me of Jonny 5?
Here is an idea it could be completely useless though. Why not use a gyroscope to control the lean of the robot that should allow a lot more fluidity in the motion and a lot more control also it could also allow the robot to move faster.
*_Я не пойму, зачем так, всё усложнять?_* 🙄🙄🙄🙄🙄🙄🙄🙄🙄
The face of the robot reminds me of Johny 5.
Just realized it's weights (75KG)
That thing looks like a very top heavy reverse pendulum. I get circus artisic vibs from this human sized pencil 🎪😅
Make more vedios ❤️🔥
What happend when it falls down?
It's fu**ed
Oh no we are in the Jetsons
It's real future !
me as a technician: "how does it work internally?"
me as an engineer: "what happens if:
-something is on the ground and block the ball?
-someone push the robot in the direction of lowering stairs?
-the battery fail?
-we combine many of thoose robots to make a wall?
-the ball doesn't touch the ground anymore?
-the motor control does not respond?
-the robot goes crazy (internal software crash / electronical short-cut / power-surge due to battery fail / ...) ?
-someone leans too much weight on the robot?
-someone is too heavy for the robot to lift?
-the robot vision is disturbed due to reflective environement or capacitive environement?
..."
Anybody who has played with some '2D' balancing robots (like Segway) knows that things go a bit crazy in such cases indeed. Maybe better control software can make it better.
But I do not really see a big need for balancing on a ball. You could easily make a small stable low wheeled platform and then put an active approximately-balanced high vertical robot on top of it. This makes the balancing much more forgiving because the platform can still provide some extra torques where needed. Compared to a human we put our whole foot to the floor and add balancing because we are very tall compared to our foot size. The ball-robot is more like balancing on the tip of your toe. I do like the high narrow robot concept a lot though.
@@richardbloemenkamp8532 yeah but my major concern is that it is created to help people. but if its fail ... what happens? where are the security?
I love ballbot.
this dude invented the Dwemer sphere
Nice to have. But in daily life application I don’t really care if it’s a single ball or four-wheeler.
BB8's ancestor ... AA1
WOOW...
bro how about you make that moving part the ball attach with a leg like moving (leg) so that it could even climb.
Its weakness is... stairs. :D
Or steep slopes. Or door frames that are like 1cm higher than the room they are attached too. Even entering elevators might be a issue
The future will be quadropets or wheeled bipedal robots. Bipedal legs are just less efficient than bipedal wheeld systems
Or LEGO
How many old wobbly dudes took one of these down on top of them before you got sued?
Why didn't sci-fi or video games think of this 🤔
BB8 from Star Wars.
it can move to 100 kph from 0 in 2 seconds ................
I don't know, this looks like a stone age robot compared to Boston Dynamics robots.
It's also 15 years old.
What? You havent hit it with a chair!?
it's got balls!
Gizmoduck!!!
This head looks familiar...
Johny 5
why don't we use this technology on cars ?
Walle all grown up
kick it
So cool? I watch it like Star Wars
Dalek in Alpha stage: "Exterminate! Exterminate!"
...just keep your floors clean.
GrabCad:
Omnidirectional spherical induction motor 360 (selfdrive transport) : Planetary gear Omni speed reduction 360 system .(OPEN SOURCE Engine)
drive.google.com/file/d/170Egu-rHYRfjxiaRHs3RZcwwNIysBCMa/view……………
GrabCad:
Omnidirectional electric Motor (Full Concept : OPEN SOURCE Engine)
drive.google.com/file/d/1qS07wafxxulyqFlR6MWwH4DfG4WjpyGf/view…
mo
One word stairs
i love ball bots, but the name is a bit too humorous for soliciting the emotions that are required for appealing to serious minded investor types. what i mean is that the colloquial 'ball' if often used to mean 'testicle' of the male of any species, especially human males. So this name alludes to the collquial for 'testicle bot' . In this case, the robot is then thought to have 'one' ball, as opposed to ' two balls' or a 'pair' or 'pair of nuts' which , is often used to refer to the number of testicles of a human male. And , of course, this doesn't even begin to approach the allusion to the human penis , a topic of much humorous discussion. That would be bad for getting money to build this robot.
Essentially I'm saying that , on a practical level, calling this robot a 'ball bot' is going to elicit a lot of snickering from young people and people who laugh at gonad humor. They may even start calling it 'dick bot' or a 'penis bot' or a 'cock bot' . This would not be good for soliciting sales or investment that is always necessary to help accelerate the development of any fine robot.
I would like to see this robot become more popular. So, please call it something different. For example, call it a 'sphere bot' or a 'globe bot' or a bowlingbot , anything but a 'ball bot' . This would likely help you actually get (more) money from investors.
BB8
wife robot
Bolloxbot.
Why build a robot to bring u things from the fridge or show u where the fridge is.. That's why guys get married :) Don't even need to ask anything. Automatically knows when the bottle is empty, bring another haha
This didn't age well.
Why build something inherently unstable?
You know, there's omniwheels and all that.
THIS is how you waste research time, money, and man-hours.
Anthony Dach But remember, for this robot to be of any use it will eventually carry an unbalanced load, which will take a whole lot more energy to keep steady.
Jets are actually very stable, and most have a fly-by-wire(or in this case fly-by-liquid) system if something goes wrong(say, heading/altitude hold).
And again - your attempt at refuting my argument fails because you suggest that this system will be stable most of the time - which it won't.
Unstable systems do have a much faster response - but at what cost? Nobody's going to need, say, a patient-observing robot to dash to someone when they press the emergency button - a camera with an epi-pen or two won't do you any good. Nor will this transport any large payload - it's simply too difficult to balance.
Also, what do you mean by "reducing power consumption when moving"? Yes, leaning will get you some momentum, but that momentum will be that of an inverted pendulum - to crash to the ground, a stable position. You'll have to work the motor continuously, both forward(accelerating) and back(braking) to keep you from either falling to the ground or stopping completely due to an incorrect control loop.
I do know what control theory is, so please, don't underestimate my chair-to-chair combat skills.
Anthony Dach If it's unstable, there's absolutely no way it can use the same amount of energy, as I've said, the center of gravity will be offset, and so will be the wheel, and the rest of the weight, and so on, and it keeps getting more unstable the more you add on.
Right now I still don't see an advantage to a very unstable inverted pendulum system with absolutely no load capacity and a constant need to use power to be stable(vs. say, a carefully selected wheel platform with less friction and latching brakes).
+Spirit
*You can't beat the inverted pendulum it's always fun.*
+Spirit
*You can't beat the inverted pendulum it's always fun.*
Because it's a friggin robot and it's balancing on a wheel like a ballerina, and it's NOT unstable. You have no understanding of the beauty of engineering, and lack the creative imagination to see how this could be applied in other areas.
Thanks Ralph! Don't know why they put "nagarajan" in that video. Were they afraid of somebody saying "racist" if didn't put a brown piece of shite in that video?