B go beyond – Quadcopter/RC car combo | Hacker News Reader

B go beyond – Quadcopter/RC car combo

kickstarter.com

191 pointsneonkiwi51 comments

I wonder if there's an efficient mechanism to allow the hubless wheels to tuck up (rotate 90 degrees about the center) when the machine takes flight.

The value could be twofold: the wheels could act as blade guards and the wheel would be outside of the prop's feed and wash airstream, which would probably lead to more desirable flight characteristics.

Maybe something for Revision 2 - as is, this is a cute combination of existing ideas into a cool new machine.

Dog bones and/or CVDs don't allow for the hub to be turned 90 degrees. You would need some sort of drives haft like Trancas uses to allow this. Though that complicates things a bit more.

It doesn't have any suspension from what I can see, so there probably aren't any dog bones.

I was thinking about it from a regular rc car. This can simply use a micro servo motor to turn it sideways

That was my first thought as well. It's too bad it doesn't happen.

It kinda makes me think of the Hovercraft pickup in Carmageddon 3 (and Back to the Future).

That'd be a great feature... Spin the wheels up and turn the props into ducted fans.

Here are a couple of ideas to make the video better

1. Use a better microphone. The quality of your voice is not very good in the video.

2. Don't keep re-using the computer generated shots, it gets somewhat annoying.

Here are some ideas that could make for a better demo of B

- Test the flying mode somewhere more interesting like in the city where there are multiple levels to reach with flying.

- make it go from the first level of a house to the second level of a house by going outside of the window. It would be even more impressive on a higher building - go from the ground level to the roof level (and back).

There are projects that you want to donate to just to show support for the creative mind for doing the seemingly impossible. This is one of them

What really worries me looking at this design is sand getting into the propeller motors. These need to be open because of cooling requirements (multicopter motors will burn out very quickly if you run them without airflow from the propeller). In this design they are very close to the ground and likely getting sand/earth thrown inside them as it gets kicked up from the ground by tires or as it gets transported upwards by tires and then falls off into the motors.

At the beginning of this video I thought this was going to be a Kickstarter project where a non-functioning model is used to sell an idea... But this guy actually built this contraption! Very impressive.

Are there spokeless wheels like that available as an off-the-shelf R/C component?

I'm curious about the patent he mentions filing, for placing the quadrocopter blades inside of the wheel. I'm not familiar with this field - is that a novel contribution? If so, kudos to him for coming up with a simple invention I would have thought would be patented or in use already.

It doesn't really seems as simple or obvious as you make it seem. The engineering that went into building those wheels with the blades in the middle seems quite impressive and not something that your average joe would be able to design and build.

The build process looks quite straightforward to me; it's a hubless wheel (common) with a brushless outrunner (common) screwed into the middle.

The idea is certainly novel as far as I know, though.

You might think of this as a "second wave" patent where people have begun patenting applications (or even implications) of previously invented technology.

This project (and thousands of others like it) have just recently been made possible by three things.

1) Small, high powered brushless motors reaching hobbyist level prices because of neodymium magnets and manufacturing techniques.

2) Control circuits becoming possible at this level as well thanks to the microcontrollers and sensors reaching the market that brought us things like arduino. (10 years ago a micro with 256k of onboard flash and a 3 axis gyro was the size of ham sandwich and cost as much as your car)

3) Cheap but very high power density lithium polymer batteries becoming available.

We're now at the point where we can make just about any little toy fly by festooning it with little propellers. So you can head down to the patent office every time you think of a new place to stick one. It will seem novel because even a few years ago, the "real" technologies that provided you with those little propellers didn't exist.

This toy is still as cool as all get-out. I just think he's wasting time if he goes any further than "pending" before he gets millions of $ in sales.

10 years ago a micro with 256k of onboard flash and a 3 axis gyro was the size of ham sandwich and cost as much as your car

genuine newbie question: is it true? 10 years ago they wire still so big and expensive?

If so, that expense and size was coming from the 3 axis gyro. 10 years ago you could get a (massively overpriced) z80 graphic calculator with several times that much flash for around $100. (The size of a small ham sandwich, yes, but that they were calculators.)

Its kinda true. In 2003 gyros were a bit larger, they were just starting to come out for R/C helicopters. And while an Arduino is a great chip its equivalent in something like a 68HC11 or other more "general purpose" microprocessor was larger in both surface area and design footprint (usually a separate eprom + CPU vs built in) the the performance was equivalent.

The real killer has been rapid update rate inertial measurement units (IMUs) which refresh rapidly enough to make interesting feedback loops practical. Small and fast IMUs have arisen out of the the development of MEMS (or Micro Electro Mechanical Systems). Integrating the accelerometer and signal conditioning is the thing that really kicked things off. Putting them into phones got the volumes to the point where they are very inexpensive.

MEMS accelerometers are a fairly recent invention

I wonder if the patent thing works as a positive signal to a normal prospective "investor".

Working in software just has me take a big gulp when I read the section about a patent here.

This stands out not only for the great design, but also for being an extremely well put together Kickstarter campaign.

Here's a startup idea: create a flashy Kickstarter campaign for your idea/product.

There's already a startup that takes care of printing T-shirt rewards.

It looks awesome ! It's an impressive piece of engineering.

The price tag is a still a bit high for what I would use as a toy, but if you consider the potential "drone" spinoff that's a steal.

The battery life is a bit limited "At the moment flying 11 min / driving up to 18 min, combined performance 15 min. "

At that price I think he'd barely scrape by with the BOM cost. R&D costs serious money to be very honest and (in my present experience) you burn through electronics and mechanical parts very quickly thanks to mistakes even after you've modeled it all up on the computer. Oh, and for an individual who cannot afford instant shipping it costs plenty of time too. I'm honestly surprised with his claim of putting it all together in 6 months actually if I assume he has a full-time job apart from this.

The price is actually decently competitive for a quadcopter with the bells and whistles he's including.

Battery life is about what I'd expect from a quad. I have a sailplane that I'm building into an FPV platform; using a 2200mAH battery like this thing is specced for, I generally get around half an hour of flight time out of a battery charge, but that's with a single motor and having the option to cut the engine and glide intermittently.

For reference, the Parrot.AR drone is about $350, and gets 15 minutes of flight time. As far as I know that's among the more popular low-end UAVs.

I'm really surprised that an innovation which is seemingly very 'simple' has been overlooked by the RC community for so long, with that said though, I'm glad someone he's taking the initiative to actually build it. I've reserved mine.

It might be worth noting for anyone who thinks this is intriguing, there is actually quite a community for FPV / RC copters. One of my favorite YouTube channels (once you get over the username) is http://www.youtube.com/user/nastycop420 which is 3 guys who go to different destinations around the world, fly quadcopters, and post the videos on YouTube.

Interesting and original-ish. But practical? Can't say.

Here's a suggestion I'm going to throw out there. Instead of driving the wheel(s), use the propellers. When landing, at least two of the propellers pivot up and drive it around.

(You could also be clever and pivot the motors up about 60-degrees, so that there is forward force and an upwards force that 'lightens' the unit. And then you could drive the two stationary motors at slow speed to further 'lighten' the unit.)

That way, you don't need another motor to power it on land, and also you could put floats on it, so it would work in the water as well as on land.

That'd be incredibly inefficient. It uses a single brushless motor to drive the wheels which you're driving directly. You'd have to run at least two motors for the props and it's unlikely you'd generate enough thrust to move nearly as well as a single direct drive.

A 2200mAH battery is already going to have limited run times in a quadcopter; trying to use the props for the wheels too would just make that worse.

That is true -- propeller-driven cars are not so efficient, but for something this small and light I don't think it would make much of a difference. The thrust to move it on wheels must be a tiny fraction of the thrust needed to lift it.

On the plus side, you could do away with those goofy wheels. Too many parts (18? 18!) and I see tiny little gears prone to binding with the slightest bit of dirt. And just a single wheel is driven.

Also, the wheels being directly in the airflow of the propellers can't be a good thing.

Of course, to use the propellers on the ground, you'd need a mechanism to rotate the 'propeller pods' up. I think using the thrust of the props themselves to do this, with a micro servo to lock them in place, would accomplish this without adding much weight.

And if the propeller motors can be run backwards (I don't know offhand) then you can get rid of the steering mechanism as well.

If that could scale (and unfortunately my intuition says that it would not), that would be the first reasonable layout for a air/road vehicle that I've seen. Pretty wild.

It would scale just fine if you wanted to drive on the highway you would be stuck with a single seat aircraft because there would not be a lot of room between the blades. Also there is no need for huge wheels if you scaled it up. Worse, a helicopter is more efficient and simpler design and yet there a extremely expensive to fly relative to a fixed wing aircraft.

IMO the best compromise is a gyro copter. http://www.blessthisstuff.com/stuff/vehicles/misc/gyrocopter...

That's not exactly what I'd call a reasonable thing to drive around. At the very least I'd want ducted rotors.

That specific design is not road worthy. Ideally you would want to be able to fold the top blades and add a duct for the pusher but that's not to bad. However, you can see the design is structurally fairly close to a car. Many designs have powered wheels simply for added maneuverability. Safety in a collision is going to be an issue due to weight concerns, but they can easily be safer than motorcycles which many people drive.

http://en.wikipedia.org/wiki/Autogyro

Advantages, unlike quad-copters loss of electronics or engine power at altitude does not equal death. They can do unpowered landings far more easily and safely than helicopters. Unlike fixed wing aircraft they can do vertical take offs and landings. Also there constuction and maintenance costs are well below that of a helicopter.

Downsides they can't hover, they have slightly lower top speeds than small airplanes, and unlike quad-copters they have a fairly large wingspan on takeoff. However, quad-copters are the only aircraft with this feature and there incredibly unsafe.

EX: A 4 seat 130mph cruse version is just a little to big for the highway. http://en.wikipedia.org/wiki/Groen_Hawk_4

I'm sure it'll come but I'd love to see a petrol powered version of this with a USB plug for an android phone. You can add all the sensors you want (LTE, gyros, gps) via the phone, and the range would be amazing by using mostly the wheels, then the rotors just to get over obstacles.

Would actually be pretty good for park rangers, border control, protecting large industrial complexes (unpredictable camera points with computer vision to locate human movement)

I'm not sure this would work; the phone will introduce a delay in the feedback, and this can be bad, and also, the sensors need to be placed in fixed positions for the best response (the gyro is usually placed right in the center).

Also, petrol/nitro engines are not the best for this kind of job, as it needs a very fast response in rotation speed and fuel engines have a high latency. What you can do is use a fuel engine as an electrical power generator or use variable-pitch propellers to get a faster response.

Sensor position is, mostly, irrelevant. You obviously get better performance with they sensors at the crossing point of all the motors, but it's entirely possible to put the sensors on one motor pod and calibrate it to fly with them in that location.

Fast on the ground, agile in the air.

It's a radio-controlled helicoupeter!

Could somebody explain why he insists on using bullet proof material to make the body? He could have picked simpler and cheaper components/materials. Is this a common practice for higher end quad-copters?

Polycarbonate is used for many things.

This is the equivalent of saying he made it with a (commonly used plastic) * 10 * $fluff.

Concrete's bulletproof too and you don't see people calling it complex or expensive.

You have a point. Let me rephrase it. There are $40 quadcopters which use weak, yet easy to replace parts. More expensive parts and materials are usually easier to work with as well, so for a prototype they are well justified. But for a production model like this one, is it a product market choice to go with expensive parts, or does the cost of designing for other materials outweigh the risk of going to the customer with a more expensive product?

CDs are polycarbonate and are sold for $20 for a spindle of 100. At a pretty good profit. "Bullet proof plastic" is just marketing buzzwords.

FWIIW, polycarbonate has to be quite thick to stop a bullet and is generally used in conjunction with glass. Looking at the chart[1], it is clear it will only stop relatively lower energy handgun bullets, not high power rifle bullets. For that, it requires a glass + plastic laminate.

[1] https://en.wikipedia.org/wiki/Bulletproof_glass

Being a mechanical engineer, i really wonder what was the path he took and tools he used (being software and hardware) in designing, verifying and creating this :)

If I was procuring military reconnaissance kit I'd be all over this. Efficient and quiet, but able to cross streams and obstacles way bigger than it.

Very cool and ambitious.

I expect the price point and lack of the streamlined Amazon checkout for US backers is going to hurt this one.

The video doesn't show, how it will turn in air? It seems that it only moves to left and right, doesn't turn to left or right.

If I saw a RC car flying like that I would freak out.

"Roads? Where we're going, we don't need roads."

The first thing I thought of was with all these bridges collapsing there may be a market for a full scale version to be able to drive up to where a bridge used to be and just fly over the gap.

I'm getting a 403 error right now. . . ?

I can't believe there's not a pledge level like this:

"If you pledge $5, we'll send you a GI Joe to put inside your B"

Kickstart all the functioning GI Joe vehicles of my childhood.

Reply on news.ycombinator.com