[0.90] Magic Smoke Industries Infernal Robotics - 0.19.3

[sirkut]

I've installed IR for the first time and I'm getting this showing up in the ksp.log

I couldn't see anything in the threat that looks applicable...any ideas on how I can diagnose it (I suspect its a mod incompatability)

I need to see the output_log.txt file.

[Revenant503]

I need to see the output_log.txt file.

How do I get a file into the forum?

[Dr_Goddard]

Hey Sirkut,

I am making a six legged "athlete" style rover.

i'm using rotatrons at the base of each leg, allowing me to stow/unstow, and perhaps eventually insect walk.

using the new Kos code to control the parts by addressing them directly.

One problem. Rotatrons on opposite sides move in opposite directions, so when they are stowed by rotating them in the VAB they are 180° out of sync at launch.

reversing the direction on one side fixes the problem, but then the legs all rotate in the same direction rather than opposing directions.I don't know right now if I can describe it better, as I'm pushing a 101° fever.

I think I want to be able to adjust the mid-point for rotation.

I'll post a video later.

[Herra Tohtori]

I have two little features that I think could be useful.

Is it possible to tie crossfeed status of a part to the phase angle/position of a robotic part? Like this:

It could be either boolean on/off switch as described here, with an adjustment option for the angle interpreted for "crossfeed on" and the rest being "off", or it could use a smooth threshold to reduce fuel flow to an engine.

As to what possible use could this be, well...

Another feature request: Servos. Basically, assigning a target position to a robotic part based on a controller axis input. For example, making a rotatron that would have a ±15 degree movement range, and would track position of pitch axis and try to match its position to the axis input.

This would make it possible to build manually articulated control surfaces that respond to control inputs, which would be rather useful in certain cases... like this, but better.

(If it's impossible to tie variables together like that, that's all right... disappointing, but I would be OK with it)

[Dr_Goddard]

I have two little features that I think could be useful.

Is it possible to tie crossfeed status of a part to the phase angle/position of a robotic part? Like this:

http://i.imgur.com/79nt7Kh.png

It could be either boolean on/off switch as described here, with an adjustment option for the angle interpreted for "crossfeed on" and the rest being "off", or it could use a smooth threshold to reduce fuel flow to an engine.

As to what possible use could this be, well...

http://expresslane.idrivesafely.com/wp-content/uploads/2013/03/internal-combustion-engine-wiki.jpg

Another feature request: Servos. Basically, assigning a target position to a robotic part based on a controller axis input. For example, making a rotatron that would have a ±15 degree movement range, and would track position of pitch axis and try to match its position to the axis input.

This would make it possible to build manually articulated control surfaces that respond to control inputs, which would be rather useful in certain cases... like this, but better.

http://i.imgur.com/C9JRJMQ.gif

(If it's impossible to tie variables together like that, that's all right... disappointing, but I would be OK with it)

It's possible to do both with IR and Kerbal Operating System.

Cool cyclic pitch BTW.

[Master Tao]

@Revenant503:

You can't – no attachments on the forums. Upload the output_log.txt to a website like Dropbox or Google Drive and then link to it here.

[BadLeo]

I'm sorry if it is a redundant report, but it seems the gantry rail won't operate normally if it's initial position is changed in the hangar. I made an arm for a space plane and I use the rail to make it hold satellites in position and deploy it when in orbit. for that I change the starting position of the gantry, since satellite's sizes can vary. It works as it should on the hangar, but not in flight, as it kind of assumes it's initial position as the maximum for one side and moves more than it should to the other. Is it a known behaviour already?

[Dr_Goddard]

Here we go.

What I want is for the left and right sides to swing in opposite rotations.

It is doable in code, but very complicated.

Would prefer not to have to code it every time for every rotation.

Problem is, with all the legs stowed. the legs on one side are 180º out of synch with the legs on the other side.

So, I am faced with making different checking routines for each side.

I think It might be solvable by rebuilding the rover in the spaceplane facility and avoiding symmetry.

I'm going to give that a try later today.

[sirkut]

Wouldn't invert axis work in this case?

[Dr_Goddard]

Here is the result of not inverting the axis. One side is rotated clockwise. The other counterclockwise.

Notice they end up at different rotations.

If you invert the axis on one side, you get rotation all in one direction per the video. But they end up rotated the same number of degrees.

I've done a few tests and it comes out the same no matter how you build it.

It seems like I need an actual clockwise rotating and counter-clockwise rotating version. The method used is like the device is flipped over to reverse it. I want the rotational scales to count in opposite directions.

[Dr_Goddard]

Ok, I got it.

After monkeying around with the .craft file for a while I was able to see it was a matter of symmetry. But ONLY SPH symmetry will do it.

If you want the rotations to move in opposing directions in a MIRRORED fashion you have to install them as pairs in the SPH editing mode. I used editor extensions to do it in the VAB.

[Dr_Goddard]

I have discovered there is a benefit to manually editing the .craft files when building robotics. Especially when you want mirrored or coordinated movements.

I tried to place all the legs and joints accurately in the VAB, but there is usually a bit of error.

If you look at the first video above, you can see the legs extend in an uncoordinated fashion. This is due to less than .02 degrees in positioning error before beginning the acceleration algorithm.

In the latest video, I cleaned up the "module" sections of the .craft file so all the legs were the same, and the resulting change in coordination is evident.

I'm trying to get my programming to a point where I can demonstrate some more advanced movements. This is a bit difficult since kOS has pretty limited support for keyboard input to programs.

I have a plan to fix this, so I'll trot out some better demos soon.

[Maxwell Fern]

I'm giving a try to this legendary mod and after two hours, my brain is starting to fry.

First of all: is there some documentation out there on the parts, or at least some good up-to-date tutorials ? I can't seem to find any.

Now for the main problem: as a first exercise, I'm trying to make foldable wings. For each, there's an open hinge attached to the ship's body, on wich a rotating part is attached, on wich a swept wing is attached.

After some trial and errors, I managed to make the hinges work as intended (symmetric, locked at certain values, assigned to a group).

But the rotating part is driving me nuts. I couldn't get anything symmetric with the rotatrons, and using the "Washer", I actually managed to get the exact behavior expected... In the SPH only. Once on the runway, one of the washer rotates in the wrong way. Is this a bug or am I doing something wrong ?

Thanks guys.

[Jivaii]

I'm giving a try to this legendary mod and after two hours, my brain is starting to fry.

First of all: is there some documentation out there on the parts, or at least some good up-to-date tutorials ? I can't seem to find any.

Now for the main problem: as a first exercise, I'm trying to make foldable wings. For each, there's an open hinge attached to the ship's body, on wich a rotating part is attached, on wich a swept wing is attached.

After some trial and errors, I managed to make the hinges work as intended (symmetric, locked at certain values, assigned to a group).

But the rotating part is driving me nuts. I couldn't get anything symmetric with the rotatrons, and using the "Washer", I actually managed to get the exact behavior expected... In the SPH only. Once on the runway, one of the washer rotates in the wrong way. Is this a bug or am I doing something wrong ?

Thanks guys.

Could right click it and invert axis.

[Maxwell Fern]

My main problem is that the parts do not behave the same in the SPH as on the runway.

For example, I'll place a washer-thingy on the side of my plane using symmetry, and on top of that I'll put a swept wing so that it could rotate to the back. Now if i use the group in the SPH, both washers turn the same way, meaning one wing folds back, the other folds ahead. Now if I load my plane on the runway and do the exact same thing, both wings fold back or ahead at the same time.

Rotatron are even more crazy: in the SPH, on top of that problem two rotatrons will never rotate at the same speed, and if I try directly on the runway they'll spin like crazy, sending all parts attached to them flying. Which is fun, indeed, but not very useful. I also have a lot of other problems too long to list here, but basically if at one point something goes weird, I'd better start over than try to fix it. Which is pretty frustrating.

[TheRubberJonny]

Anyone know where I can find a download of the cargo bay from wolfpack aeronautics?

Cheers

[sirkut]

Anyone know where I can find a download of the cargo bay from wolfpack aeronautics?

Cheers

One day I will upload the pack as I'm now the sole person in control of them.

[JeffreyCor]

Any sort up update on those cargo bays would be awesome! Overall still find them to be some of the best for non spaceplane use. The Firespitter bomb bay seems to be the only other good alternative but the sealing / shelf reduces the effectiveness of them.

[dcompart]

@sirkut Is there any way to increase the rigidity of the parts? I made my own robotic arm out of rotatrons and hydraulic cylinders, but it's so heavy it is drooping down under it's own weight. I would like to make everything more rigid. How do I do that? Reduce the mass of the parts I'm using? Thank you for the mod and let me know if you have a solution.

[MartGonzo]

Any chance of a move and lock option in the action groups?

it kind of bugs me that the motors don't stop when using action groups.

[sirkut]

There is nothing that can be done to make the joints more rigid. Blame that on unity and it's joint system.

[gkorgood]

@sirkut

the other day on DasValdez's stream, he was building a replica of something called the Discovery II, out of this NASA paper: NASA Paper. The view of the ship he was using as reference is at the top of page 5. Those 3 arms spin to generate artificial gravity, and he tried using your free-spinning docking washer to accomplish this. However, the command module before the centrifuges is supposed to be linked to the rest of the ship below them. This was not doable without some Kraken-enticing strutting. I, along with others in chat, suggested a skateboard-bearing type part, where there is an inner race, and an outer, the inner remains stationary, while the outer spins. I attempted to make one, and successfully modeled it in blender. However, the two meshes are structurally fixed to each other whenever I test it in-game. I don't totally understand how to fix it, any help would be awesome.

[sirkut]

You mean something like this?

http://forum.kerbalspaceprogram.com/threads/84566-WIP-Magic-Smoke-Industries-Station-Parts-%28Infernal-Robotics-required%29

[gkorgood]

You mean something like this?

http://forum.kerbalspaceprogram.com/threads/84566-WIP-Magic-Smoke-Industries-Station-Parts-%28Infernal-Robotics-required%29

yes but a free spinning one so you could spin it up to centrifugal speeds. also, making it one part would be nice, and i just got into modding and wanted to try my hand. any ideas?

[sirkut]

yes but a free spinning one so you could spin it up to centrifugal speeds. also, making it one part would be nice, and i just got into modding and wanted to try my hand. any ideas?

Making it one part isn't going to be possible which is why this is in two pieces. Free spinning is just a matter of changing parameters like the free spinning docking washer.