Jump to content

Shuttle Challenge v5 - The STS thread [Stock and Mod Friendly] - MAJOR CHALLENGE ANNOUNCEMENT! - 30.3.2020


Recommended Posts

^ Interesting node on why I use mechjeb for ascent guidance. I initially planned on having MechJeb just handle the flying entirely, but the way MechJeb works it can't actually fly the stack properly. It overcorrects too much and doesn't utilize the control the stack has very well.  So I use stock SAS to lock itself to the target MechJeb tries to follow and it in turn flies as its supposed to. 

Link to comment
Share on other sites

@michal.don I think that the initial flight should show that it is manually fly-able and you can then use autopilot for that shuttle design unless the mission is like STS-1b or 2b, which tests your and your shuttle's flying abilities. Also, if you choose another design, you should fly the initial flight of that design manually.

Link to comment
Share on other sites

9 hours ago, Bottle Rocketeer 500 said:

@michal.don I think that the initial flight should show that it is manually fly-able and you can then use autopilot for that shuttle design unless the mission is like STS-1b or 2b, which tests your and your shuttle's flying abilities. Also, if you choose another design, you should fly the initial flight of that design manually.

This is a fair idea actually. +1

Link to comment
Share on other sites

@Alchemist, @Bottle Rocketeer 500, @G'th, thank you for your opinions. I thought about it a bit as well, and I decided to change the rules regarding autopilots - they are way too strict. As Alchemist said, the rule should encourage manual-ish piloting, so pressing space, going for a coffee and returning to a shuttle in orbit is not possible. However, I'm not sure where to draw the line, and how to exactly phrase the rule. I'll take some time to think about it a bit more, but if you have a suggestion how the rule should be, your advice is much appreciated.

Thanks,

MIchal.don

Link to comment
Share on other sites

11 hours ago, michal.don said:

@Alchemist, @Bottle Rocketeer 500, @G'th, thank you for your opinions. I thought about it a bit as well, and I decided to change the rules regarding autopilots - they are way too strict. As Alchemist said, the rule should encourage manual-ish piloting, so pressing space, going for a coffee and returning to a shuttle in orbit is not possible. However, I'm not sure where to draw the line, and how to exactly phrase the rule. I'll take some time to think about it a bit more, but if you have a suggestion how the rule should be, your advice is much appreciated.

Thanks,

MIchal.don

Perhaps make the distinction between manual and (semi)autopilot flyers, with the autopilot option obviously being the lesser distinction. The way I see it the extent that I used mechjeb should really be the end of the line as far as autopilot goes, as the only point of even using it is just to smooth out the gravity turn. (who likes looking at super gimballed engines gimballing back and forth a thousand times a minute lol) But at the same time the only point I put my hands on the controls during launch is to stage or complete a roll. Not quite full manual but it also wouldn't make it to space if I just let it go.  I mean we could almost call what I did fly-by-wire really. 

It should also be noted that if someone piped up along the line that designed their shuttle as such that it does its gravity turn automatically, so he still doesn't have to put his hands on the controls, do we still call that manual flying? And if we don't, iit'd be a shame not to recognize the skill and expertise it'd take to get a shuttle that has that behavior. 

Link to comment
Share on other sites

d7ieXvT.jpg

Hiya michal, managed to complete STS-2a, though I do have to say for full disclosure, an incident in my reentry brought me down between the mountains and KSC instead of KSC as intended (based on reentry behavior from STS-1b), so the jets had to be fired to make it to the runway and I'm not sure if this doesn't count as a precision reentry-to-landing for the Commander badge? And I had the spin stabilization set so kindly for a shuttle noob, too,  it wasn't the hardest part of the mission I feared it to be.

Re, the use of autopilots: from what I observe, there is a qualitative difference between letting the computer run it all for you and just letting it assist you. It's more of a sliding scale of human intervention in computer-aided flying, in fact. G'th's, for example, clearly exploits the same tactic that some other shuttle pilots do in setting prograde-follow on SAS, only instead of prograde he's tracking a target being moved by MechJeb. He's probably right in saying that it's the limit in terms of computer aids, because the only difference between what he does and what MJ really does is that MJ includes some degree of corrective steering, and G't still has one critical control input available, thrust modulation (which is what Gravity Turn does instead of MJ's corrective steering). Alternatively, one of the earlier qualifiers here who suggested using SmartASS to control pitch angle still retains some degree of pilot intervention: plugging the numbers into SmartASS (plug the wrong numbers, and it's game over), though admittedly it does eliminate one of the challenges of reentry, hypersonic flight control (which I had trouble with handling manually during -2a, as explained in the Mission Report therein). But there is still that avenue of pilot intervention.

On the other end of the scale, you have Atmopshere Autopilot, effectively a fly-by-wire allowing for stability-relaxed designs (though not sure about its behavior in hypersonic or high-alt regimes), which does ease up on the design process, or TCA's ability to modulate engine power independently to minimize or eliminate torqued thrust, which allow the full range of pilot input but nonetheless actively control certain elements to prevent departures from controlled flight. And like G'th pointed out, there are kOS scripts and even Smart Parts/Remotetech programming which can make a craft launch a one-button press affair. Though admittedly, asymmetrical designs may pose a unique challenge to such efforts. (Not that it's stopped MechJeb from being used on KSOS before, though I's not sure for its feasibility after 1.0)

If only to keep the spirit of the Challenge predecessors, I'd qualify the use of computer aids as requiring that there still be some significant pilot input. Exactly what that pilot input is (whether having one control element or axis not under computer command, or the exercise of punching numbers into SmartASS, or programming the launch script) may still be up for discussion, and it could even be a "I'll know it when I see it" scenario. And like G'th said, what he did might have to be the floor for minimum allowable pilot input. In either case, I'm okay with some assistance being used so as long as it's clear that the player is not brute-forcing it. 

Edited by B-STRK
Link to comment
Share on other sites

Hello again my friends, sorry for the slight delay - I'm afraid a busy ime is coming, so I might not be able to check this thread as often as I would have liked.

 

On 14. 10. 2017 at 9:30 PM, B-STRK said:

Hiya michal, managed to complete STS-2a, though I do have to say for full disclosure, an incident in my reentry brought me down between the mountains and KSC instead of KSC as intended

Great to see another of your reports - I have to admit I absolutely love the story-telling, and your style when most of the kerbals running the space program are a bit incopetent and problems are to be expected :) Also, thumbs up for the Hubble mirror reference :D

So, a well earned badge, commander :wink:

E6Sskam.jpg?1

 

Regarding the autopilot rules:

I still did not figure out a way to phrase the rule exactly. So, for the time being, I'll decide each case as it comes. Generally, pilot aids like mechjeb target for gravity turn are A-OK, complete autopilot to orbit is a no.

 

Michal.don

 

 

Edit: 

that being said, I correct my previous verdict, and award @G'th a badge for his great mission:

nTYtbyq.jpg?1

And I hope to see more!

 

Edited by michal.don
Link to comment
Share on other sites

Thanks once more @michal.don! And geez, I know the feeling: I have a licensure exam all next month and my KSP time is already feeling the squeeze (yet it's one of the only few ways I can decompress from all the studying, this isn't healthy for me at all :D). Anyway I think I figured out how to deal with the instability I experienced in -2a, just need a verification flight or two to confirm it, and then (presuming I can squeeze out the free time to do so) -3. :)

Link to comment
Share on other sites

Here is my entry for the STS-1a (commander) challenge. (gotta start somewhere right? :)) I will be trying the other challenges as well. In the image below is a complete list of all mods I use, most of them don't add parts but are there to enhance the game a little. Mechjeb was only used for orbital procedures.

hNi6TBl.png

And the video log of the flight:

 

To be honest I could have flown it a bit better, but after the boosters drop it kinda starts jiggling (anyone have a remedy for this?) As for the design I went for a more classic approach. :D

Edited by Kaname
Link to comment
Share on other sites

Completed the STS-2a mission.

Out of curiosity, do I swap my current badges with their equivalent of the newer, fancier badges, or should I keep the old ones?

All parts stock, 360km shuttle orbit, 3 spin stabilized satellite deployments, runway landing. I believe this qualifies for commander badge 

Edit: I hope my self-written launch and partial reentry control scripts are ok. The launch script was accepted in the previous thread (although it has seen  pretty much a total rewrite since then to use vector math instead of headings). I wrote the 12 line partial reentry control script just before I reentered, simply to reduce the amount of babysitting of the attitude I would have to do. Was pretty much just "hold nose Pilot_input_x degrees above horizon."

@Kaname not sure how to make that more stable. Perhaps a less aggressive pitch-over after booster separation? Might be that the high angle of attack is resulting in stalls on the wings.

Edited by EpicSpaceTroll139
Realized I actually hadn't done them that out of order, just misread
Link to comment
Share on other sites

14 hours ago, EpicSpaceTroll139 said:

 

Out of curiosity, do I swap my current badges with their equivalent of the newer, fancier badges, or should I keep the old ones?

I think it's ok either way.  Some people line up all their badges, some only keep the badge for the highest/furthest mission they've completed.

Link to comment
Share on other sites

Figured I'd give this a try. Below Missions 1 and 2 including lugging the 40t tank around.

Mission 1:

Album https://imgur.com/a/JYdFv will appear when post is submitted

 

Mission 2: (I moved the fuel out of the tank in case I needed it on descent, but better piloting ultimately meant I didn't need it. It's still there, just not in the right tank.)

Album https://imgur.com/a/Yy1fT will appear when post is submitted
 
I'm not certain I completely understand mission 3, can anybody post a link to a log?

Link to comment
Share on other sites

1 hour ago, hoioh said:

<snip>

I'm not certain I completely understand mission 3, can anybody post a link to a log?

If by mission 3 you mean the STS-2a mission, in my previous post (the second post above yours), I completed the mission in a manner which qualifies for the commander badge (I hope). I went fancy and used carefully fueled solid kick motors to send my satellites to roughly the right orbit after deploying the from the shuttle, but I don't think there's anything that says you can't do the orbit boost using liquidfuel rockets, or even take the whole shuttle up to geo orbit.

Link to comment
Share on other sites

8 hours ago, EpicSpaceTroll139 said:

If by mission 3 you mean the STS-2a mission, in my previous post (the second post above yours), I completed the mission in a manner which qualifies for the commander badge (I hope). I went fancy and used carefully fueled solid kick motors to send my satellites to roughly the right orbit after deploying the from the shuttle, but I don't think there's anything that says you can't do the orbit boost using liquidfuel rockets, or even take the whole shuttle up to geo orbit.

I meant STS3, but you're right I didn't perform STS2a up to spec either.

What is meant by "no spin stabilization" for the commsats? otherwise I think I get it and all I have to do now is move my sattelites to a higher orbit. (may still be salvageable).

Link to comment
Share on other sites

2 hours ago, hoioh said:

I meant STS3, but you're right I didn't perform STS2a up to spec either.

What is meant by "no spin stabilization" for the commsats? otherwise I think I get it and all I have to do now is move my sattelites to a higher orbit. (may still be salvageable).

Maybe Not having any Reaction wheels on on the shuttle cockpit aswell and rcs can be used?

Link to comment
Share on other sites

3 hours ago, hoioh said:

I meant STS3, but you're right I didn't perform STS2a up to spec either.

What is meant by "no spin stabilization" for the commsats? otherwise I think I get it and all I have to do now is move my sattelites to a higher orbit. (may still be salvageable).

 

1 hour ago, Kerbal Tween said:

Maybe Not having any Reaction wheels on on the shuttle cockpit aswell and rcs can be used?

Actually the "spin stabilization" requirement refers to what you do to the satellite after detaching (staging if you loaded the satellites using a decoupler, decoupling node if by docking port) the satellite from the shuttle cargo bay before igniting its engine the first time. It has nothing to do with the shuttle SAS or RCS.

To keep a craft on course during a rocket engine burn, it needs to be stabilized or any rotational movement will cause the vessel (satellite in this case) to turn around while under burn (and the engine torque can make it worse); that's why we activate SAS to keep a certain vector (e.g., prograde, maneuver node) to keep the vessel steady. For the Pilot level STS-2a, you can do that for launching your satellite from your shuttle: just stage, activate the sat's SAS (if it's not on) so the reaction wheels keep it steady, and take it to its target orbit (geostationary, or a transfer orbit, depending on how you do it). Without this control by SAS (or Mechjeb or another autopilot), you can "topple over" a rocket while under burn and cause it to burn off course (e.g., turning towards retrograde while you want to burn prograde); you'll need to make a WSAD control input to get it back on course.

"Spin stabilization" works differently. No SAS, reaction wheel torque, or RCS is used to keep the satellite on course during the burn. Rather, upon the launch of the satellite but before (or during, but this is harder) the satellite's engine is lit, it is spun as fast as possible, while the reaction wheels are OFF (either turn off SAS, or disable reaction wheel torque) so that it keeps spinning, so that like a top, it's hard to "topple over" the satellite as you burn the engine to take it to geostationary/transfer orbit. Just use the throttle; no need to touch WSAD until the burn's done. One way to do this as suggested was using separatrons set so they spin the satellite, and activated before firing the main engine. It also only has to be for the first burn; any course correction burns after that can be done SAS-on. 

Based upon the fact that you launched two satellites, you're qualifying for the Pilot STS-2a badge (plus attempting the STS-2b Flight Director badge bonus), so you don't have to worry about spin stabilization. Just get the satellites to geostationary orbit. As for moving fuel out of the fuel pod, even if unconsumed by your maneuvers, you might want to check with @michal.don on that. 

Link to comment
Share on other sites

6 hours ago, hoioh said:

I meant STS3, but you're right I didn't perform STS2a up to spec either.

What is meant by "no spin stabilization" for the commsats? otherwise I think I get it and all I have to do now is move my sattelites to a higher orbit. (may still be salvageable).

Pretty sure by "no spin stabilization" it means just like normal. With spin stabilization would be setting it spinning it like a top to keep it stable during the burns like I did.

Edit: @hoioh my bad on not seeing you'd done STS 2a. I hadn't had the chance to look at you logs, and I misread the bit about bringing the tank down, so I thought your latest one was STS 1b

Edited by EpicSpaceTroll139
Ugh autocorrect
Link to comment
Share on other sites

Did a retry of STS 2a in order to make the sattelites spin on their axis while racing off towards KSO, I just angled the litte engines a bit to provide spin and off they went whithout SAS control. But Jeb had been gnawing on the fuel lines out of boredom again and while launching nr 2 the ship exploded... 1 of the remaining sattelites could be salvaged, but the fourth was unfortunately lost and the crew had to be resqued by a spare STS vessel with a claw. The mission was a succes with just 3 sattelites in equally spaced KSO orbits, the resque went well too (only the claw burned off during re-entry without damaging the cockpit) and everyone safely touched down on the KSC runway in the end.

Album https://imgur.com/a/0fEMt will appear when post is submitted
Edited by hoioh
Link to comment
Share on other sites

Finally put in the effort to start with this challenge! So, as is necessary I'm starting with STS-1a, quite simply getting into orbit.

I did design my own shuttle, would be a bit boring without the challenge of that imo, so here it is: the Ubehebe Shuttle.

L95d8nr.png 

It's a MK2 shuttle, so not a lot of room for cargo in it, but it has got a cargo bay. The big tank that it's attached to is powered, to make it have a tad easier flight characteristics. Got into orbit pretty much in the first attempt I made at it. 

I'll leave you a nice imgur album with all the images in it right here, cuz otherwise i'd be absolutely flooding this chat with images probably.

Reentry was quite a poodle with this shuttle, but her excellent glide path made sure I could make it to the runway without too many issues regardless. I hope this qualifies as a success for STS-1a, cuz I'm already looking forward to launching off for STS-2.

Link to comment
Share on other sites

I have an issue moving onto STS-2 mission. In my first mission I placed the MUD tank perfectly, and even docked with the tank after to complete systems set-up and place it into a very precise orbit.

Since then, my attempts to send the sats up (got WAY too many geosync sats now!) have been successful, and in trying to get the fuel pod, I have crashed into it so many times the orbit is no longer perfect.... I have however shot the pod down using an ICBM, landed it using a chute probe carried up by the shuttle etc. etc. Must I have the pod back inside the shuttle to qualify for landing it?

 

Link to comment
Share on other sites

11 hours ago, Andetch said:

I have an issue moving onto STS-2 mission. In my first mission I placed the MUD tank perfectly, and even docked with the tank after to complete systems set-up and place it into a very precise orbit.

Since then, my attempts to send the sats up (got WAY too many geosync sats now!) have been successful, and in trying to get the fuel pod, I have crashed into it so many times the orbit is no longer perfect.... I have however shot the pod down using an ICBM, landed it using a chute probe carried up by the shuttle etc. etc. Must I have the pod back inside the shuttle to qualify for landing it?

 

You Have to its A Space shuttle challange 

Link to comment
Share on other sites

@Andetch   Yes, i think you need to land with the pod inside the cargo bay.   A warning for the landing:  the pod likes to break when you touch the ground... or, like in my case, once landed, when a kerbal dares to get out of the cockpit  :D
 

Spoiler

OwdogZ4.png

XJfFcim.png

0MdHLVl.png

1YaAq67.png

 

Link to comment
Share on other sites

On 10/14/2017 at 3:30 PM, B-STRK said:

d7ieXvT.jpg

Hiya michal, managed to complete STS-2a, though I do have to say for full disclosure, an incident in my reentry brought me down between the mountains and KSC instead of KSC as intended (based on reentry behavior from STS-1b), so the jets had to be fired to make it to the runway and I'm not sure if this doesn't count as a precision reentry-to-landing for the Commander badge? And I had the spin stabilization set so kindly for a shuttle noob, too,  it wasn't the hardest part of the mission I feared it to be.

Re, the use of autopilots: from what I observe, there is a qualitative difference between letting the computer run it all for you and just letting it assist you. It's more of a sliding scale of human intervention in computer-aided flying, in fact. G'th's, for example, clearly exploits the same tactic that some other shuttle pilots do in setting prograde-follow on SAS, only instead of prograde he's tracking a target being moved by MechJeb. He's probably right in saying that it's the limit in terms of computer aids, because the only difference between what he does and what MJ really does is that MJ includes some degree of corrective steering, and G't still has one critical control input available, thrust modulation (which is what Gravity Turn does instead of MJ's corrective steering). Alternatively, one of the earlier qualifiers here who suggested using SmartASS to control pitch angle still retains some degree of pilot intervention: plugging the numbers into SmartASS (plug the wrong numbers, and it's game over), though admittedly it does eliminate one of the challenges of reentry, hypersonic flight control (which I had trouble with handling manually during -2a, as explained in the Mission Report therein). But there is still that avenue of pilot intervention.

On the other end of the scale, you have Atmopshere Autopilot, effectively a fly-by-wire allowing for stability-relaxed designs (though not sure about its behavior in hypersonic or high-alt regimes), which does ease up on the design process, or TCA's ability to modulate engine power independently to minimize or eliminate torqued thrust, which allow the full range of pilot input but nonetheless actively control certain elements to prevent departures from controlled flight. And like G'th pointed out, there are kOS scripts and even Smart Parts/Remotetech programming which can make a craft launch a one-button press affair. Though admittedly, asymmetrical designs may pose a unique challenge to such efforts. (Not that it's stopped MechJeb from being used on KSOS before, though I's not sure for its feasibility after 1.0)

If only to keep the spirit of the Challenge predecessors, I'd qualify the use of computer aids as requiring that there still be some significant pilot input. Exactly what that pilot input is (whether having one control element or axis not under computer command, or the exercise of punching numbers into SmartASS, or programming the launch script) may still be up for discussion, and it could even be a "I'll know it when I see it" scenario. And like G'th said, what he did might have to be the floor for minimum allowable pilot input. In either case, I'm okay with some assistance being used so as long as it's clear that the player is not brute-forcing it. 

Btw I Checked atomsphere autopilot and it was better then SmartASS Pitch control :) And It does work in hypersonic flight like in subsonic flight 

Edited by KerbalTween
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...