Jump to content

The Elcano Challenge: Ground-Based Circumnavigation (4th)


Recommended Posts

Thresomin is a very, very small world in the whirligig world planetary pack. It's only got a radius of 2 km, plus, some 700 m of terrain elevation. Less than 20 km to circumnavigate. So, as I landed there, I decided to circumnavigate it.

Spoiler

SYp7WGd.png

Thresomin, as it's being approached

0VzUtCs.png

For this mission I'm using the lander Cigar. No kerbalism this time - it's incompatible with the planetary pack - so I get to use the high visibility cockpit.

Thresomin is very flat, especially for its small size. it's oval, rotating at a significant speed, but it's only got a few hundred meters of elevation difference across the equator. this means that even in this diminutive gravity, I can pick up some speed. I got as far as 7 m/s, which on such a world is huge. To avoid jumping too much, I decided to circumnavigate against the moonlet rotation - else I would risk reaching orbital speed by accident.

On low gravity worlds, you have to deactivate reaction wheels, accelerate; soon the rover will start to tip up, and then you have to reactivate reaction wheels and stop this rotation. Deactivate the wheels again, and when you touch the ground you can pick up some more speed. Cigar is light and pretty responsive, so I could accelerate decently.

q5Aml5d.png

etT2icQ.png

In the sky is Mesbin, your home in this mod. Mesbin is as massive as Jool, sporting a crazy 11 g on the surface; however, it also rotates extremely fast - 1 round in 15 minutes or so - therefore you only need a small speed on the equator to reach orbit. I actually fancied a polar circumnavigation, but I couldn't make any rover capable of resisting 11 g; the wheels just broke. The fast rotation also gives it the oblate shape.

TbYjSVg.png

Picking up some speed

Dc94ABb.png

3SNcdaD.png

wUs2zdp.png

The sun is going below the horizon, but I already have speed, I only need a tiny amount of battery to keep the reaction wheels pointed. I decide I'll try to go on and not pause and wait for the day. To plant flags, I had Bob perform an EVA while the rover was still moving, so I wouldn't have to lose speed. I'd then catch up with the jetpack. On Priax it didn't work, it was too irregular, but here it's working nicely. I forgot to take a pic, though.

Anyway, the battery run out too fast. While I could cover several km like that, and I came close to the first flag, I run out of battery.

FSHufFu.png

And I subsequently spun out of control.

Nmca3Rg.png

I felt safe enough in my low speed, but scraping on the ground was enough to break all three solar panels. And I have Bob on board, can't fix them.

koHyRZU.png

From its taxi in orbit, Bill came down with the jetpack to rescue the rover

x1iLlfQ.png

He fixed the solar panels, then hitched a ride on the ladder

84skZDa.png

Obligatory IVA view of Bill on the ladder

y2htV87.png

First flag reached

Dsh1Xk2.png

Thresomin circumnavigated.

I never needed to use the rocket on Cigar; on such a small moon, the distinction between "giving yourself a push" and "skipping ground with a suborbital jump" is too small.

Whirligig world offers a couple of very interesting challenges.

The starting world, Mesbin, is a giant planet that rotates super fast. One day lasts 15 minutes. there are 11 g on the poles, but only 1 g on the equator due to centriphugal force (yes, I know, it's not a perfectly accurate explanation, let's just be short). It would be a unique challenge to make a polar circumnavigation of Mesbin; a rover would have to resist 11 g, and be able to move uphill. I tried a bit, but there was nothing I could build that would survive the poles.

 

Statmun is the inner moon of Mesbin, floating in geostationary orbit, very close to the surface. it's very small (3 km) and it rotates fast, so on the equator you cannot stand still; centriphugal force is higher than gravity, so you just fly away. to make an equatorial circumnavigation of Statmun one could make a rover with rockets pointing downwards (boring). Or, one could make a robot with ground anchors on the leg, to take one step at a time while staying anchored to the terrain (super cool!). Would be a nice challenge, but I don't want to learn robot making from scratch.

However, maybe someone else will pick those challenges

 

Link to comment
Share on other sites

3 minutes ago, king of nowhere said:

On low gravity worlds, you have to deactivate reaction wheels, accelerate; soon the rover will start to tip up, and then you have to reactivate reaction wheels and stop this rotation.

I'm sorry if this is Captain Obvious, but this sounds like you are using the default controls where W is both accelerate and pitch-down.

Link to comment
Share on other sites

47 minutes ago, damerell said:

I'm sorry if this is Captain Obvious, but this sounds like you are using the default controls where W is both accelerate and pitch-down.

I do, but i deactivate reaction wheels before pushing W. in fact, it's got nothing to do with pitch control, because the rover pitches up, not down.

in a realistic simulation, I surmise it's because the wheels are spinning in a sense, and conservation of angolar momentum imposes an opposite rotation on the rover, and in low enough gravity that's enough to pitch up. I don't know how realistic ksp is - as I tried to learn better planes, I was flabbergasted at how much stuff doesn't work realistically - but either way, I've done several tiny worlds with different rovers, and always my rover pitch up. I'm surprised you didn't experience the same on Gilly. then again, from what I understand your gilly circumnavigation was done at orbital speed with rockets pointing downwards, so it was an entirely different experience.

Edited by king of nowhere
Link to comment
Share on other sites

@damerell and @king of nowhere, one of my frustrations with KSP2 has been the lack of ability to re-map rover controls.  I personally like to re-map rover controls so they do not overlap the SAS controls (Pitch, Roll, Yaw).

There are a lot of valid solutions for dealing with 'rovering' on low-gravity CBs.  But using powered wheels can be quite frustrating (in my opinion).  You both have come up with innovative and useful solutions to the problem!  

Link to comment
Share on other sites

4 hours ago, 18Watt said:

@damerell and @king of nowhere, one of my frustrations with KSP2 has been the lack of ability to re-map rover controls.  I personally like to re-map rover controls so they do not overlap the SAS controls (Pitch, Roll, Yaw).

from a purely effectiveness perspective, that's the best option. if we were rover racing, we'd all do it. you can steer and use the reaction wheels to stabilize simultaneously, that's a huge advantage.
on the other hand, keeping the same controls and turning the reaction wheels on and off has the advantage of being easier to learn and perform, and to require only one hand - leaving the other free to turn around the camera with the mouse. it's a more "touristic" approach

Link to comment
Share on other sites

20 hours ago, king of nowhere said:

in a realistic simulation, I surmise it's because the wheels are spinning in a sense, and conservation of angolar momentum imposes an opposite rotation on the rover, and in low enough gravity that's enough to pitch up..

I think it's because applying a force at the base of the rover creates a torque about the centre of mass, just as IRL dragsters sometimes pitch up before things go horribly wrong.

Link to comment
Share on other sites

I think I'll really start my Elkano this time...

Quick math: rover has top seabed speed of around 7 m/s. a fourth of Kerbin's circumference is 300*pi ~= 900 km. The time the water part will take will be 900,000/7 ~= 128571 s which equals... 35 HOURS??
OK, I'll still do it, I was expecting worse :)

Link to comment
Share on other sites

8 hours ago, Nazalassa said:

Quick math: rover has top seabed speed of around 7 m/s. a fourth of Kerbin's circumference is 300*pi ~= 900 km. The time the water part will take will be 900,000/7 ~= 128571 s which equals... 35 HOURS??

While I'm reasonably sure I was over that on Eve (at 40 m/s it would take over 30 hours, and I was not running at top speed any time I was going uphill), I think at that point I would consider a bit more time in the VAB might pay off.

I've a rules question, or rather, a request for an exemption. RSS Earth is really big. If I could sail directly around the Equator at the 155 m/s of the water speed record IRL, it would take over 70 hours. In view of that, I'd like to ask if it is permissible to use an autopilot that can only hold a steady course ("steer 275 degrees, limit speed to 150 m/s"), with no ability for the autopilot to adjust that figure or stop the boat - any change to planned heading, maximum speed, or maximum throttle setting would have to be done by human intervention. I'd write the autopilot myself in kOS, no use of MechJeb (not even the stabilisation at sea used on Laythe). It's fine if the answer is "no".

Edited by damerell
Link to comment
Share on other sites

On 5/21/2023 at 7:47 PM, damerell said:

While I'm reasonably sure I was over that on Eve (at 40 m/s it would take over 30 hours, and I was not running at top speed any time I was going uphill), I think at that point I would consider a bit more time in the VAB might pay off.

I've a rules question, or rather, a request for an exemption. RSS Earth is really big. If I could sail directly around the Equator at the 155 m/s of the water speed record IRL, it would take over 70 hours. In view of that, I'd like to ask if it is permissible to use an autopilot that can only hold a steady course ("steer 275 degrees, limit speed to 150 m/s"), with no ability for the autopilot to adjust that figure or stop the boat - any change to planned heading, maximum speed, or maximum throttle setting would have to be done by human intervention. I'd write the autopilot myself in kOS, no use of MechJeb (not even the stabilisation at sea used on Laythe). It's fine if the answer is "no".

do you need an autopilot to keep a steady course?

For my laythe circumnavigation, I just pointed my "boat" east and alt-tabbed out. For speed limit, speed is set by propeller angle; it kept 53.2 m/s (just below the speed that plane needed to take off from water) without any other input. As for direction, it would steer north or south with time, but I only had to check every few minutes. and I would find in those minutes the "boat" had started pointing 20 degrees north or south, it's ok, it's still going mostly east. So I point it back east, and alt-tab again. no need for any autopilot.

 

As an alternative, you can make a boat that has a lot of lateral control, but no up/down control, and set as target a planet. Choose it so that a boat pointing at it will point roughly east (or west, if you prefer). sure, it will be a bit below or above the horizon, and the boat will try to sink its head underwater, but since you have no up/down control, it won't be able to. and it will keep the east/west direction.

as a fancier version of this, you could set up a satellite constellation to use as targets to keep pointing east. You'd have to change target every once in a while, but you may still be able to go a couple hours without inputs.

 

P.S. @18Watt I see my thresomin circumnavigation hasn't been added to the leaderboard yet

Link to comment
Share on other sites

On 5/21/2023 at 12:47 PM, damerell said:

I've a rules question, or rather, a request for an exemption. RSS Earth is really big. If I could sail directly around the Equator at the 155 m/s of the water speed record IRL, it would take over 70 hours.

I'll get to this discussion towards the end of this post..

2 hours ago, king of nowhere said:

P.S. @18Watt I see my thresomin circumnavigation hasn't been added to the leaderboard yet

It's on there now!  And thank you sincerely for reminding me!  I sometimes get busy with work, and forget to do things I need to.

7 minutes ago, damerell said:

Target-following would, to me, be included in the usual prohibition on having anything steer for you.

Ok, back to the autopilot question.  In my opinion, autopilots are not valid for Elcano runs.  That becomes a different sort of challenge than this challenge is intended.

However, targeting is absolutely allowed.  Feel free to use any of the 'stock' SAS modes available- Pro/Retro, Normal/Anti-Normal, Radial In/Out, Target Lock.  In fact, if you want to get really creative, you can build a functional autopilot with a probe-core mounted on a servo, and 'steer' the rover/boat by locking on to various stock SAS modes.

Here's a link to a challenge I participated in a year or so ago:

The challenge was to build a rover which could drive from KSC to the North Pole, without any manual control after departing KSC.  Oh, and you could only use Stock + BG + MH parts, no autopilots.  We managed to make some rovers which could be navigated to the North Pole using only servos and KAL controllers for 'steering'.  All turns were time-based.  There was a lot of trial-and-error..

Again, my logic is yes it is much easier to do Elcanos with an autopilot, they can be set up (not that you would do this) so you can walk away from your computer, come back in 12 hours, and the run is complete.  The Elcano has always discouraged autopilot use, from day 1.

However, using the stock SAS targeting modes is absolutely allowed.  If you are targeting waypoints or flags, you will need to manually go fly a mission to drop your 'targets' out there- all support missions also have to be flown from KSC too, 'editing' support missions into position is not allowed.

Using 'targets' in KSP1 is probably the best way to go.  In (stock) KSP1 there is no SAS mode which can target relative to 'north' or any other compass heading.  If you are near the equator (say within 30 degrees of it), then referencing the Normal/Anti-Normal vectors provides a fairly good North reference.  However, as you get closer to the poles, Normal/Anti-Normal no longer points North/South, it begins to have a vertical component, and becomes nearly completely useless beyond about 70 degrees N or S.

One interesting feature in KSP2 is that when your frame of reference is 'Surface' mode (as opposed to 'Orbital'), the Normal/Anti-Normal SAS modes switch, and now lock on to North/South!  I really hope that 'feature' stays in the game as they make changes to KSP2, it is a very useful feature.

I should mention that the prohibition of autopilot use makes circumnavigating RSS Earth wildly impractical.  However, I do feel that using an autopilot is beyond the intention of this challenge.  

Which leads me to my final point- Circumnavigating using autopilots is a different challenge than Elcano, in my opinion.  And I think it would be a great challenge!

35 minutes ago, damerell said:

Target-following would, to me, be included in the usual prohibition on having anything steer for you.

So:  Target-following IS allowed, or using other SAS modes.  Auto-throttle or any sort of automatic speed control is not allowed.  If you use propellers, they are extremely effective at maintaining a specific speed- the more propeller blades you have the more stable your speed will be, even when going up or down slight inclines.  Using jet engines will cause your speed to vary a little as you burn off fuel and become lighter, but I suspect that with the size of vessel you'd need for RSS Earth your speed would not vary much over the course of say an hour.

Link to comment
Share on other sites

10 hours ago, 18Watt said:

It's on there now!  And thank you sincerely for reminding me!  I sometimes get busy with work, and forget to do things I need to.

 

there's an errata: thresomin is not in OPM, it's in whirligig world

Edited by king of nowhere
Link to comment
Share on other sites

One more question about RSS Earth. RSS Earth has a selection of space centre locations which can be switched between. Is it acceptable to do this in order to launch resupply missions from the closest one, or must the KSC location remain fixed?

Link to comment
Share on other sites

10 hours ago, damerell said:

One more question about RSS Earth. RSS Earth has a selection of space centre locations which can be switched between. Is it acceptable to do this in order to launch resupply missions from the closest one, or must the KSC location remain fixed?

Supply missions may be flown from any useable launch location on Kerbin (or RSS Earth...)

I believe you are referring to this rule:

  • The main mission and all support missions must be launched from the launchpad or runway and delivered to the destination body.  Part of the journey is getting the crew/ship there.  (i.e. No HyperEditing or save file editing the vehicle into position.  Any combination of launching, refueling, docking, or ISRU is permitted.

The intention of that rule is to specify that all components of the mission need to be flown from Kerbin (Usually KSC, but does not have to be KSC.  In other words, you aren't allowed to Hyper-Edit or F12 any part of your mission into position.  

Say you have a rover on Eeloo, and discover that you need to deliver some replacement parts or extra fuel.  That support mission needs to be flown from launch all the way to Eeloo, just like your main mission vessel.  

I agree that the rule is somewhat vague or ambiguous.  Anyway, you are allowed to launch the main mission or any support missions from any valid launch site on Kerbin or RSS Earth.

Link to comment
Share on other sites

On 5/21/2023 at 7:47 PM, damerell said:

While I'm reasonably sure I was over that on Eve (at 40 m/s it would take over 30 hours, and I was not running at top speed any time I was going uphill), I think at that point I would consider a bit more time in the VAB might pay off.

I've a rules question, or rather, a request for an exemption. RSS Earth is really big. If I could sail directly around the Equator at the 155 m/s of the water speed record IRL, it would take over 70 hours. In view of that, I'd like to ask if it is permissible to use an autopilot that can only hold a steady course ("steer 275 degrees, limit speed to 150 m/s"), with no ability for the autopilot to adjust that figure or stop the boat - any change to planned heading, maximum speed, or maximum throttle setting would have to be done by human intervention. I'd write the autopilot myself in kOS, no use of MechJeb (not even the stabilisation at sea used on Laythe). It's fine if the answer is "no".

Well, solution sounds easy enough: just place a couple of flags on various points across the Earth and use them as targets for SAS

Now, whether you are allowed to put the flags down with the HyperEdit/F12 menu before the mission or you must manually place them is a different question entirely. I'd suggest you clear this up with @18Watt

Link to comment
Share on other sites

On 5/25/2023 at 7:19 PM, OJT said:

Well, solution sounds easy enough: just place a couple of flags on various points across the Earth and use them as targets for SAS

I did consider just targetting each refuelling mission, but I really don't want either a boat with no pitch control or a boat that's trying to nose into the ocean. I found a different option, ie, a much faster boat.

Link to comment
Share on other sites

On 5/25/2023 at 1:19 PM, OJT said:

Now, whether you are allowed to put the flags down with the HyperEdit/F12 menu before the mission or you must manually place them is a different question entirely. I'd suggest you clear this up with @18Watt

No, all support missions need to be flown, including placing flags or other targets.

Due to the size of RSS Earth, it makes more sense to fly the support missions (flag planting) from the launch site nearest the target location.  You aren’t allowed to F12 the flags into position, but you can fly those support missions from any valid launch site.

Hypothetically, I suspect there are mods which allow the construction and launch of vessels from other celestial bodies, or off-surface bases.  For the Elcano challenge, that’s not allowed, everything needs to be launched from Kerbin (or Earth..)

 

Link to comment
Share on other sites

@damerell, I ran across this while browsing the forum today.

Immediately thought of you, possibly planting targets around RSS Earth.  It sounds like you have a different plan now anyway, but thought I'd pass it on just in case..

Link to comment
Share on other sites

I went on an Elcano Yesterday- the start of one. I'm currently condensing the footage into something more manageable since I've recently been recording and setting up for HD. I really need to install that new SSD.

Here's the progress so far:

i5Cz5mx.png

The craft is called "GarageMonster" It is the "OneOscar" Program's latest Publicity stunt to show it's not a "one-trick pony", just a One-Oscar B fuel tank pony.

k5PYvp3.png

The Starting Point- The goal is to follow lowlands and flats as much as possible.

KgCWOqR.png

The vehicle has some pilot assist countermeasures for Kraken attacks like this one. The ground likes to take the vessel sharply off-course (see my first bug report vid), but I've developed some SOP's for this vessel in preparation for this mission. The research began at the garage...

bQpAESe.png

Following the flats is easy enough. Next up is some highlands? There's definitely challenges ahead.

e3AQTAU.png

This is the point I'll start from within an hour or two. I think I can get across these hills. Yesterday was promising.

Follow me here to watch when I broadcast

Link to comment
Share on other sites

It was much quicker, and inline with Last weeks Challenge, to Circumnavigate Gilly. 

Here's a clip, and a link to the livestream archive:

Video should be linked to relevant timestamp for circumnavigation, or near...  I've been condensing footage as I go. I'll likely submit a full circumnavigation video.

I'm going back to Minimus now.

Link to comment
Share on other sites

I thought it would be Minmus first for me, but here's the full condensed trip around Gilly. It's about 45 minutes as is. Most of it is a 8x edited speed, with much of the run being performed at 2x physwarp.

I hope more people attempt this one. It's totally manageable! Although, maybe a clarification should be made about how much "ground transportation" is actually required for a Gilly Run, because even with wheels alone this craft can make some pretty awesome suborbital jumps. For full acceleration I need to fire up the Dawn engine to full thrust to keep the vessel from performing a wheelie. Super satisfying.

Here's a gallery of detes and comparative screenshots:

Spoiler

Where I started:

SdvPwpy.png

Where I Called it Complete:

tZSs8bF.png

Actual complete was a bit further:

Hso0pgn.png

A Breakdown of some times:

iORNuYy.png

and a cool screenshot:

YtkEilW.png

Pretty neat

How did I do?

I thought it would be Minmus first for me, but here's the full condensed trip around Gilly. It's about 45 minutes as is. Most of it is a 8x edited speed, with much of the run being performed at 2x physwarp.

I hope more people attempt this one. It's totally manageable! Although, maybe a clarification should be made about how much "ground transportation" is actually required for a Gilly Run, because even with wheels alone this craft can make some pretty awesome suborbital jumps. For full acceleration I need to fire up the Dawn engine to full thrust to keep the vessel from performing a wheelie. Super satisfying.

Here's a gallery of detes and comparative screenshots:

Spoiler

Where I started:

SdvPwpy.png

Where I Called it Complete:

tZSs8bF.png

Actual complete was a bit further:

Hso0pgn.png

A Breakdown of some times:

iORNuYy.png

and a cool screenshot:

YtkEilW.png

Pretty neat

How did I do?

Link to comment
Share on other sites

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