I'd love to keep improving this and add more features in the futures. And who knows; maybe this is the start of a really cool open source cable camera (still need a name for that). So, subscribe to my channel, if you'd like to be updated on future upgrades :)
Haha :D Glad you enjoyed it. It's very interesting working together like this. I was really nervous and had to prepare a lot in order to use our time efficiently. Would definitely like to do that again sometime.
Changing the motors is the only way to improve this. BLDC and the right control would be ideal for this application. Additionally, you can examine the subject of cable release during an accident and how to make sure cables are always tight for safety reasons. Sensors can also be added to measure the distance to the ground.
A good friend of mine told me to make something like this after I finish my current project! This is pretty cool I 150% want to take a look at your move logic :)
Uh nice! I’d still love to improve this, but can’t find any time at the moment. In case you write any code, I would be very interested to see how you approach it :)
Very cool! Time to put the pulleys in the four corners of the room and have a flying camera that can go anywhere! I bet a simple pendulum type steadicam setup could be relatively cheap and easy to setup and it would make the camera movement smooth enough for internal stabilization on the camera to really smooth it out. I'm imagining a heim joint in the center of the mounting plate, then the camera mounted below the plate and a counter balance mounted above the plate. (Though, I've never actually tried that, so I could be way off mark here, haha). Either way, excellent project and looking forward to more! Oh, for a name, how about "The MFC: Matou Flying Camera"!
Yes I’d love to put it in the room corners. However: I then realized that the LED panel on the light stand would always be in the way. That’s why I’m currently inclined to have a movable setup. The steadicam idea is another very interesting suggestion with interesting consequences. Again my initial thought was that it would be a really good idea. But then I watched scanlime’s video on the Tuco Flyer and she explains that it’s better to have the cables attached close to the center of mass. Because otherwise the camera would swing around a lot. I will definitely experiment with different approaches. And thanks for the name suggestion. I’ll keep that in mind. Best I came up with so far was: Cool Cable Camera 😅
This is a awesome project. I would try smal brushless motors with a encoder for this project. You can stil position the camera where you want, bruschless motors run smoother and you can use a belt drive to make it even more smooth. Great project annyways tho, I bought a arduino to make a camera crane, but put the project on a hold because the coding is to hard for me atm as I never had. A arduino before. LoL
Yeah, a camera crane isn't the easiest project to start out on an Arduino :) Regarding the brushless motors: I think they might have too little torque. Although I don't know much about motors. I will definitely explore more options in the future.
I checked for your code. I'm deeply impressed, you didn't use any ready to use library. On the other hand such libraries can help because they are very advanced. I used AccelStepper in the past. This enables to define some acceleration instead of moving with constant speed. This should avoid swinging camera. As far as I understood your code, you don't compensate the offest of the mounting at the camera. (the four cables are not fixed at the same point of the camra). Depending on the actual cable angle this creates different offsets. In addition it seems that you move the four motors with the same speed, this will lead to the problem that some motors reach the final cable length faster than others? If so, this will lead to a tilt of the camera during the movement.
One goal of mine was to also learn about the components, such as the motors. So I started out coding it from scratch with the goal to research suitable libraries in the future. Thanks for the hint to AccellStepper. However, I don’t think smooth acceleration of the motors will completely get rid of the issue. I believe a lot of the instability comes from the mechanical side, such as the pulley design and camera mounting. I will keep working on that as well. Actually, the motors are not moving at the same speed. I first check which cable needs to move the most. That motor goes full speed. The others are adapted in relation to that.
Awesome project! I’m currently building a shed/workshop for my boat building and video project. I would really like to build a spider cam system for the shed to film the process easily and from interesting angles. I could try to implement your code for it at some point. :)
That sounds like a great application for a spider cam. I really hope I can continue building mine soon. Really didn’t have time for it these last few months
@@matoumakes Yeah. I think that's the easiest way to accomplish it as well. If you think alternatives, they all include some kind of bigger structures to move across the space. (like cnc machine like system with three axis) but may be easier to control with ready-to-go systems. The biggest problem I think is to reliably spool the wire reliable and without mess to the rollers. In my case the space is about 16m x 10m x 5 meters. So there's quite a bunch of wire needed to roll. Another idea is to have only one wire across the center and on to that there's longish rotating arm with counterweight to mount the camera. BTW the idea of having 360 camera is awesome! No need to fancy camera controlling doodads!
I really like your project, i was also planning for the same project but for irrigation system. I haven't started the project yet, but if i get some help from you that will be great for me.
Nice to hear that other people are trying as well. I think this really has potential. I’m a computer scientist, so the coding part was okay for me. I’m still not convinced that the math part is as good as it can be. For example: I still have no idea how to get changing speeds in there in a good way.
@@matoumakes I'd start looking around for functions like easeInOutSine and whatever Stackoverflow/Google spits out, e.g., stackoverflow.com/questions/23038569/c-trouble-understanding-easeinoutsine (not sure if you meant that, and also ... I have no clue :D)
There has to be an algorithm that does this a bit smoother. How could it be done without breaking it into chunks? I am thinking of this exact problem right now as well. There are curves I believe. I also wonder if you had 4 cables coming from down to up, would it lock the camera into position more rigidly.
Yes there definitely has to be a better algorithm. I think there is so much possible improvement. And 4 additional cables would most likely make it more rigid. Searching for cable robot on RUclips gives some impressive videos about such machines. But it’s also more complicated and expensive
I think it would work. I thought about using an arduino gcode library. However my goal with this project is also to learn as much as possible by implementing most things myself. But maybe I should start with a library that’s available.
I think I figured it out. I am not sure but I think you need to also calculate for time. Make it so the speed of the motor spins at the speed that gets point a to point b in a specific time. If all 4 points have the same start and stop time then the movement should be smooth.
Nice project! I noticed that you assumed that every revolution of the motor has a fixed cable/thread length but that isn't true if you ask me. Thats because on every revolution the cable/thread stacks up on each other. Thoughts? :D
Yeah you’re right. I thought this would be a good first approximation. But it definitely showed that some cables got more slack while others tightened over time. I think including a sensor for rope tension would be best to keep the cables nicely even. And for positioning I am not sure. I would love to actually measure where the camera is in the room. Gotta look into indoor positioning systems for that.
I'd love to keep improving this and add more features in the futures.
And who knows; maybe this is the start of a really cool open source cable camera (still need a name for that).
So, subscribe to my channel, if you'd like to be updated on future upgrades :)
Thanks for letting me help you, turns out it's more fun to help you work on your videos than it is to work on my own :D
Haha :D
Glad you enjoyed it.
It's very interesting working together like this. I was really nervous and had to prepare a lot in order to use our time efficiently. Would definitely like to do that again sometime.
Changing the motors is the only way to improve this. BLDC and the right control would be ideal for this application. Additionally, you can examine the subject of cable release during an accident and how to make sure cables are always tight for safety reasons. Sensors can also be added to measure the distance to the ground.
A good friend of mine told me to make something like this after I finish my current project! This is pretty cool I 150% want to take a look at your move logic :)
Uh nice! I’d still love to improve this, but can’t find any time at the moment.
In case you write any code, I would be very interested to see how you approach it :)
Very cool!
Time to put the pulleys in the four corners of the room and have a flying camera that can go anywhere!
I bet a simple pendulum type steadicam setup could be relatively cheap and easy to setup and it would make the camera movement smooth enough for internal stabilization on the camera to really smooth it out. I'm imagining a heim joint in the center of the mounting plate, then the camera mounted below the plate and a counter balance mounted above the plate. (Though, I've never actually tried that, so I could be way off mark here, haha).
Either way, excellent project and looking forward to more!
Oh, for a name, how about "The MFC: Matou Flying Camera"!
Yes I’d love to put it in the room corners. However: I then realized that the LED panel on the light stand would always be in the way. That’s why I’m currently inclined to have a movable setup.
The steadicam idea is another very interesting suggestion with interesting consequences. Again my initial thought was that it would be a really good idea. But then I watched scanlime’s video on the Tuco Flyer and she explains that it’s better to have the cables attached close to the center of mass. Because otherwise the camera would swing around a lot. I will definitely experiment with different approaches.
And thanks for the name suggestion. I’ll keep that in mind. Best I came up with so far was: Cool Cable Camera 😅
very nice! would love to see an update with the pulley wobble fixed!
Thanks :)
I’ll definitely keep working on this. Only currently I don’t have a lot of time 😅
This is a awesome project. I would try smal brushless motors with a encoder for this project. You can stil position the camera where you want, bruschless motors run smoother and you can use a belt drive to make it even more smooth. Great project annyways tho, I bought a arduino to make a camera crane, but put the project on a hold because the coding is to hard for me atm as I never had. A arduino before. LoL
Yeah, a camera crane isn't the easiest project to start out on an Arduino :)
Regarding the brushless motors: I think they might have too little torque. Although I don't know much about motors.
I will definitely explore more options in the future.
I checked for your code. I'm deeply impressed, you didn't use any ready to use library. On the other hand such libraries can help because they are very advanced. I used AccelStepper in the past. This enables to define some acceleration instead of moving with constant speed. This should avoid swinging camera.
As far as I understood your code, you don't compensate the offest of the mounting at the camera. (the four cables are not fixed at the same point of the camra). Depending on the actual cable angle this creates different offsets.
In addition it seems that you move the four motors with the same speed, this will lead to the problem that some motors reach the final cable length faster than others? If so, this will lead to a tilt of the camera during the movement.
One goal of mine was to also learn about the components, such as the motors. So I started out coding it from scratch with the goal to research suitable libraries in the future. Thanks for the hint to AccellStepper.
However, I don’t think smooth acceleration of the motors will completely get rid of the issue. I believe a lot of the instability comes from the mechanical side, such as the pulley design and camera mounting. I will keep working on that as well.
Actually, the motors are not moving at the same speed. I first check which cable needs to move the most. That motor goes full speed. The others are adapted in relation to that.
@@matoumakes Thanks for the explenation. Very impressive work.
Awesome project! I’m currently building a shed/workshop for my boat building and video project. I would really like to build a spider cam system for the shed to film the process easily and from interesting angles. I could try to implement your code for it at some point. :)
That sounds like a great application for a spider cam.
I really hope I can continue building mine soon. Really didn’t have time for it these last few months
@@matoumakes Yeah. I think that's the easiest way to accomplish it as well. If you think alternatives, they all include some kind of bigger structures to move across the space. (like cnc machine like system with three axis) but may be easier to control with ready-to-go systems. The biggest problem I think is to reliably spool the wire reliable and without mess to the rollers. In my case the space is about 16m x 10m x 5 meters. So there's quite a bunch of wire needed to roll.
Another idea is to have only one wire across the center and on to that there's longish rotating arm with counterweight to mount the camera.
BTW the idea of having 360 camera is awesome! No need to fancy camera controlling doodads!
Fantastic work. Love this project!
Thanks :D
Excellent
love the project !
Thanks 😊
I really like your project, i was also planning for the same project but for irrigation system. I haven't started the project yet, but if i get some help from you that will be great for me.
As useless as this is for my rig, thats only because it weighs 12-15 pounds. This is a brilliant cable cam for this setup. Well done!!
I've been trying to do a similar project, but you've got further with the "hard part" - the math and coding.
Nice to hear that other people are trying as well. I think this really has potential.
I’m a computer scientist, so the coding part was okay for me. I’m still not convinced that the math part is as good as it can be. For example: I still have no idea how to get changing speeds in there in a good way.
@@matoumakes I'd start looking around for functions like easeInOutSine and whatever Stackoverflow/Google spits out, e.g., stackoverflow.com/questions/23038569/c-trouble-understanding-easeinoutsine
(not sure if you meant that, and also ... I have no clue :D)
@@EggCess good idea. Thanks 😊
Cool as heck
Thanks 😁
There has to be an algorithm that does this a bit smoother. How could it be done without breaking it into chunks? I am thinking of this exact problem right now as well. There are curves I believe. I also wonder if you had 4 cables coming from down to up, would it lock the camera into position more rigidly.
Yes there definitely has to be a better algorithm. I think there is so much possible improvement.
And 4 additional cables would most likely make it more rigid. Searching for cable robot on RUclips gives some impressive videos about such machines. But it’s also more complicated and expensive
@@matoumakes Have you heard of gcode for cnc? Do you think that would work?
I think it would work. I thought about using an arduino gcode library. However my goal with this project is also to learn as much as possible by implementing most things myself.
But maybe I should start with a library that’s available.
@@matoumakes Well I believe the speed of the motors will be a curve. I too am trying to figure out this exact problem. Hard one to crack.
I think I figured it out. I am not sure but I think you need to also calculate for time. Make it so the speed of the motor spins at the speed that gets point a to point b in a specific time. If all 4 points have the same start and stop time then the movement should be smooth.
The NFL (American football) uses this system. Of course, they've taken it to a whole different level.
Nice project! I noticed that you assumed that every revolution of the motor has a fixed cable/thread length but that isn't true if you ask me. Thats because on every revolution the cable/thread stacks up on each other. Thoughts? :D
Yeah you’re right.
I thought this would be a good first approximation. But it definitely showed that some cables got more slack while others tightened over time.
I think including a sensor for rope tension would be best to keep the cables nicely even. And for positioning I am not sure. I would love to actually measure where the camera is in the room. Gotta look into indoor positioning systems for that.
BirdCam is cool 😎 😉
Oh I like that as well 😃
Can anyone help me get the circuit diagram
"cable camera" ? ... is a SpiderCam!
Why not call it the Vector Camera rig
VCR for short? I like it 😆
I sacrifice this comment in the name of the algorithm.
😁