Building a drone copter

[cobito]

Yesterday I saw a video recorded from a remote-controlled drone copter and I was amazed. So, I started looking at prices and below €250-300 there is nothing.

I've been thinking about how I could build a decent one and looking at electronic components, I've seen that the price mounts up to more than what the pre-assembled ones are worth. So I've thought about assembling the electronics myself.

The idea is to use a microcontroller (PIC16Fxx which is the one I have some experience with) and with it control the speed of the motors by pulses to a thyristor. In turn, the device would be stabilized with an accelerometer connected to the PIC and the PIC would perform the appropriate calculations to keep it stable and make the drone obey orders coming from the outside.

The problem is that the experience I have with PICs is only at the simulation level, so I would need a programmer (how do you buy that?).

On the other hand, there would be the issue of the radio link: I thought that with a wifi module I could control the drone from a laptop.

This is very roughly what I would like to do (and I say would like because I have to see if it is economically viable).

Do you think that in this way the savings in construction are considerable or not worth it and it is better to buy the prefabricated boards they sell?

Thanks.

[CNCBCN]

good looking and cheap Juguetronica | Gadget and toy store - UFO SOLAR.
they sell kits on the boards (as you well know) that are expensive, but if you look around a bit you're sure to find the schematics for one and you can make it yourself (that is, if you're willing to use hydrogen peroxide and all that hassle)
edit:
i don't know how long it will take the guy to finish it, but it promises http://aeromodelismovirtual.com/showthread.php?t=13001

[cobito]

The model that is assembled is fine. Let's see how it progresses. I have continued with my research and it seems that the best solution for the electronics would be the Arduino platform (which I did not know about and it looks pretty good). I would already know more or less what parts I would have to buy (although I still have some doubts). My model would not be less than 300€. I see it getting more complicated every time although I will continue researching to see if I find a way to make the design cheaper. If anyone has knowledge of model aircraft, their experience would be helpful because in other forums I have not been able to get answers to my questions.

[Tecnho]

It's that for 300e you have the Parrot ARdrone, which I like for a while, I was going to tell you that you have mini helicopters and spare parts in Chinese stores like miniinthebox and to play it's fine priced around 30-40€, but I suppose you want these drones that are very stable.

[cobito]

@Tecnho:

It's that for 300e you have the Parrot ARdrone, which I already like for a while, I was going to tell you that you have mini helicopters and spare parts in Chinese stores like miniinthebox and for playing it's well priced around 30-40€, but I suppose you want these drones that are very stable.

Sure, I don't want a toy, I want something real, where I can mount a video camera or a reflex, that has stability and some autonomy in addition to a considerable range. Getting all that costs a lot of money.

The ARdrone from Parrot also appeals to me, but I think that for 300€ you can make something better (bigger, that can carry more weight).

Anyway, for now I don't want to spend those amounts on a whim of these characteristics, although in the not too distant future maybe I will.

Besides, it promises to be a very interesting world from the point of view of making it yourself.

For now I will continue to soak up everything (the Arduino platform has seemed really interesting to me and has many applications) so that when I see myself with funds and time, I do it well and with knowledge of the matter.

[Tecnho]

Well, at home, I encourage you to do a full worklog as it would be of great interest to me personally and other forum members, it would be interesting and I would like to have the morning and patience to do it XD.

Regards

[CNCBCN]

I hope it works out well and they don't give you these things

by the way, can these bugs rotate 180 or 360 degrees on the X Y axes? like a helicopter, if you will

[cobito]

That could have been avoided with a GPS module and programming the electronics so that the drone goes to some coordinates in case of losing the signal.

@CNCBCN:

by the way can these drones rotate 180 or 360 degrees on the X Y axes? like a helicopter to make ourselves understood

With a gyroscope they can rotate around the xy plane. A gyroscope adds about €130 to the design. It should not be confused with the measuring gyroscope (which informs the electronics of what the drone is rotating in the 3 axes, but which does not modify its position.

I would like not to have to use gyroscopes but servos that modify the angle of the motors. I think it would make it much more manageable and agile, although the design would become considerably more complicated.

[CNCBCN]

I've been thinking for a while about how those bugs work and I've come to the following conclusions (most likely erroneous) the rotors are numbered as follows:
R1 R2
R3 R4
if you increase the power of R3 and R4 you modify the Y axis if you then equalize the power of all four the bug moves forward. For backward it's the same but inverted and for the sides more of the same but with R1 and R3 (or R2 and R4) but to turn I think it has to be "headbutting" this already takes a while: R1 at 50% R2 40% R3 60% R4 45% is it more or less like that?

[cobito]

@CNCBCN:

I've been thinking for a while about how these things work and I've come to the following conclusions (most likely incorrect). The rotors are numbered as follows:
R1 R2
R3 R4
If you increase the power of R3 and R4, you modify the Y axis. If you then equalize the power of all four, the thing moves forward. For backward, it's the same but inverted, and for the sides, more of the same but with R1 and R3 (or R2 and R4). But to turn, I think it has to be "headbutting". This already takes a while: R1 at 50% R2 40% R3 60% R4 45% Is it more or less like that?

I don't know if your method would work. They usually use (or so I think) gyroscopes. They are devices that have a disk spinning inside at high speeds so that when the disk is rotated on its axis, a force is exerted on the supports. This is what segways use to keep their balance.

[Sylver]

And a tail rudder with horizontal propeller(s)? To propel it forward and backward, and with the tail rudder it would already turn… You would have to place these elements well, but it would work according to my calculations, airplane style...

Edit: Even a single rotating propeller would be feasible, like in hovercrafts, it's all about getting a good design.

[cobito]

@Sylver:

¿Y a tail rudder with horizontal propeller(s)? To propel it forward and backward, and with the tail rudder it would already turn… You would have to place these elements well, but it would work according to my calculations, airplane style...

Are you referring to a fifth engine that turns the machine? The idea is to include the fewest number of elements possible, because every extra gram reduces efficiency.

@Sylver:

Edit: Even a single rotary propeller would be feasible, like in hovercrafts, it's all about getting a good design.

It must be taken into account that a single engine with a single propeller does a quarter of the effort. To save costs it's fine. You would have to calculate if it could lift everything.

[CNCBCN]

this is the design of Sylver
http://1.bp.blogspot.com/_8IhgYlS934s/TTNeeTONBHI/AAAAAAAAABs/IF4rbVa2BYc/s1600/gorro-helice.jpg

[defaultuser]

The truth is that the method of playing with the engine regimen through electronics is the one that will make a lighter design.
Lighter, and with more load capacity consequently, and initially also more economical.

[incrediboy]

@defaultuser:

The truth is that the method of playing with the engine regimen through electronics is the one that will make a lighter design.
Lighter, and with more load capacity consequently, and initially also more economical.

and also much more difficult to work well considering the air currents.

[Sylver]

@CNCBCN:

this is the design of Sylver
http://1.bp.blogspot.com/_8IhgYlS934s/TTNeeTONBHI/AAAAAAAAABs/IF4rbVa2BYc/s1600/gorro-helice.jpg

Exactly! How did you know? Did it remind you of your childhood? :D:D:D

[CNCBCN]

@Sylver:

¡Justamente! ¿Como lo supiste? ¿Te recordó a tu infancia? :D:D:D

JODER como lo sabes :D:D:D como disfrutaba dándole con la mano abierta :D:D:D:D:D

[cobito]

@incrediboy:

and also quite more difficult to make it work well considering the air currents.

I think that playing with the engine settings is the ideal way. The user does not control the revolutions of each engine, but rather the microcontroller does it totally automatically adapting to the conditions with a measurement gyroscope and accelerometer.

In this way, the user just tells the "avionics" what they want the drone to do (for example, to move forward at a certain speed) and it will make the appropriate calculations to offer the right setting for each engine and that what is desired is fulfilled, whatever the environmental conditions (within certain limits).

[Sylver]

What if you place the motor that acts as a rudder right in the center of the structure? Under the body of the drone right in the middle, so the weight is balanced, and to lift the entire structure you will only need a little more power in all the rotors, but if the motor is not very heavy it will only take a few revolutions.
I see it as less complex and if done right it would be very effective.

Regards

[CNCBCN]

Sylver …. it's not for nothing but the "Rudders" always get put behind, they could be put in front, but not in the middle .
the only more or less viable and cheap thing would be to do the three motors and a servo of ¿180 degrees? two motors in front and one behind connected to the servo, it would turn on the Y axis to give direction and the rest as I commented before