![images[1].jpg](https://i.imgur.com/Khw1PG0.jpg "images[1].jpg")
I'm looking to stablize a flat platform using 4 motors, what sensors should I use? I'm using an arduino, is it safe to assume a couple of accelometers will do the job?
You can get a 6 axis accelerometer (often called a gyro but it's not) on a single chip.
Then you need a servo shield. Then you need to copy paste some code you can google.
>>916779
>often called a gyro but it's not
But it is.
It's a 3-axis gyro and a 3-axis accelerometer on the same chip. Either way, yes, that's what you want.
>>916775
Imagine the top of the cup being your platform, and the filled (and heavy) bottom of the cup some suspended weight.
You're thinking way too complicated.
>>916775
Just put the platform on 4 rigid legs. A table doesn't need all that to stay stable.
>>916783
My russian LOMO (actually soviet) microscope had a table that was heavy as shit and it had rubber pads between the drawers that doubled as legs and the tabletop. If it is enough to work with a 200-fold magnification then it might be for you also?
Yes they will, but how good are you at PID control?
>>916779
>Then you need a servo shield.
>Not driving the sevos off the pwm pins like a man
Normies...
>>916775
>I'm looking to stablize a flat platform using 4 motors, what sensors should I use? I'm using an arduino, is it safe to assume a couple of accelometers will do the job?
Define "stabilize." It might be enough or it might not, depending on your application.
tape it to a chickens head..
https://www.youtube.com/watch?v=_dPlkFPowCc&feature=iv&src_vid=UytSNlHw8J8&annotation_id=annotation_132186
>>916779
What does a shield add? Can't I use transistors and build an h bridge? what does the gyroscope add in terms of accuracy(?) to the accelerometer?
>>916891
>>916785
>>916783
>>916782
Actually the table is going to be floating aproximetly 50meters underwater, and going to have to function correctly while carrying fish that is still trying to escape even with a gaping hole. Having infinite energy (cable to generator above the water) I don't care about the motors doing this job.
>>916813
As good as someone who searched the term in wiki just now and generally understood the equations. Any tips? I don't mind reading before monkeying code.
>>916904
>Actually the table is going to be floating aproximetly 50meters underwater, and going to have to function correctly while carrying fish that is still trying to escape even with a gaping hole.
So it's moving, then? And what kind of outputs are you going to have?
A gyro measures rotation directly. An accelerometer measures acceleration and gravity, and can indirectly find tilt angle for a platform which isn't accelerating (by measuring gravity itself). Depending on your goal, accelerometers might even work for platforms that ARE accelerating (i.e. the yaw dampener on an airplane in a coordinated banked turn - doesn't keep the plane level/straight, but it will keep the passengers from spilling their drinks as if they were flying straight). There are 6-axis IMU chips/modules which can do all at once, and even 9-axis IMU+magnetometer units as well.
I can help you determine what inputs you need to control your platform properly, but I'm going to need more details about what you need your platform to do.
>>916904
>50 meters underwater
hoo boy that's a lot of water pressure to deal with.
What exactly are you trying to do.
>>916904
Shield is just what they call it when you buy the part for your Arduino. Its like $10 w shipping so no real reason to DIY the accelerometer part as long as one off the shelf is suitable for your application.
>>916924
>>916928
It's basicaly a submarine controlled by a wire above sea, it should be able to move in the x and y axis and even tilt and move while staying in that degree (give or take) in case entering a tight entrance.
The main target is spear fishing large fish which is in this general depth, this fish is also extremely calm and "friendly" for lack of other dangerious interactions, there is going to be a gun module attached with a camera, the secondary target is just exploring.
It's a large project but I have the time and money and even manpower, most of the education (CompE) but very little experience, my focous is software but hopefully I can manage with you guys on board.
>>916930
The thing is that the lab I'm stealing from has plenty of transistors and plain sensors, but nothing arduino oriented, I don't mind buying the stuff but if it's possible to implement then why not, unless it's inferior, I'm also not a fan of black boxes spoon feeding me (to some extent).
>>916928
Also forgot to mention, pressure wise, sealing the whole deal with mineral oil should fix most of the problems, but can the accelerometers handle pressure? They use crystal reactions to pressure for electrical feedbak, 50m is enough to affect those?
how about the camera for the arduino?
>>916775
One flywheel will do it perfectly just work mass and sniping speed. level it once and forget it will stabilize itself.
you could just get a kk2 board and some servos or motors. easy job.
>>916932
Okay. In that case, some degree of righting moment can be achieved simply by buoyancy/mass distribution (i.e. keep the heavy bits at the bottom and floaty bits at the top, or even consider placing a weighted external keel below the submersible). But if you need active stabilization and control, then gyro stabilization would probably be ideal and even accelerometer stabilization would probably be workable. You COULD homebrew it, but as >>917058 suggested, the easiest and most practical approach is probably to use a hobby-grade stabilizer board (something like a Naze32 will only cost you like $25, is purpose-built for the job with a 6-axis IMU and can be programmed with custom mixes with relative ease, unlike the KK2).
Obviously radio gear isn't going to work well through 150 meters of water, but you could generate PWMs or serial inputs (which are what drive the stabilizer boards) using an Arduino or whatever.
So anyways... let's talk controls. Given that you want to navigate slowly through confined spaces, I assume you'll want 3-axis translational control... you probably don't NEED all 3 axes of rotational control, but it might help. You should probably look at other ROV examples, and then sketch out your ROV including where all the thrusters will be located and what each will be used to effect (i.e. two side-mounted longitudinal reversible thrusters could be used to effect both longitudinal translation and yaw).
I'd say the gravitational stabilization sounds best. And then forget about the rest, just learn to drive your new rov, I bet it won't be to hard. Put a diving compass in the field of view of the camera. And don't drive your rov into tight spaces as the tether will get stuck.
My friend drives scientific rovs that they use to survey the sea floor. I'll ask him if they have some active stabilization.
God damn /diy/, human advancement moves full circle. The reincarnation of Captain Ahab is trying to catch his whale using an arduino. My sides are in orbit.
>>917637
>I'd say the gravitational stabilization sounds best.
It is what the majority of ROVs use. See pic related; large yellow buoyancy chamber up top, dense (probably ballasted) chamber at the bottom with all the equipment, and a handful of thrusters which provide 3 axis translation as well as yaw only (pitch and roll being passive via buoyancy/gravity). It works fine, no stabilization needed really.
>>917681
>My sides are 20,000 leagues under the sea
>>918755
I just drove home from spearfishing. Got none, didnt even see any(worthwhile) fish.
I was chatting with my friend about their rovs for sea floor surveys. They have gravitational balancing.
Anyway I'd say the first problem will be waterproofing the thing. 50m is a lot. The thrust system is always tricky, lots of moving parts that needs to be waterproof. Another tricky thing would be the spear itself. How to reliably reload the gun? Can't expect to hit each time... After the movement and the actual speargun is okay then adding the stabilization is just some programming, which is indeed a problem of its own but not the first thing to consider. Just have that in mind when building it so it actually can rotate on very direction.
So I'd say just build the damn thing, make it work manually and add the automation later!
An idea for fighting the fish would be a CO2 balloon you could fire that would surface the rov when catching prey. If the fish has a swim bladder then surfacing will kill it anyway and the fight is over.
As for driving it in confined spaces, I'd avoid. You will have serious trouble with the tether, it WILL get stuck and diving down to recover it will not be a trivial task.
>>916928
50 meters? You might be able to do that with standard glued PVC fittings.
>>916775
MPU-6050
