KSP2 EA: Where is the Journey / Mission planner?

[Vl3d]

After reading some forum comments I realized most KSP players got so used to using DeltaV maps that they actually forgot how much <trial, error and writing information down> is necessary to play the game without wikis.

Without getting bogged down in details I can give a few examples. Without maps and wikis each player would have to do experiments and find out:

- how much DeltaV you need to go anywhere

- proper launch windows for all destinations (so we can launch directly to target instead of wasting time in orbit)

- the TWR you need to land

- atmospheric entry altitude / angle

- when and where to deploy science experiments

- the correct altitude to deploy parachutes etc.

So clearly we would benefit from having all this information accessible in-game after doing the appropriate experiments and missions to get the data.

Also it would be really useful to be able to plan a mission in detail before launching - to have a place to write down each step and action we plan on doing (which we can adapt later while we are actually on the mission) and get some alerts to remind us of the mission plan.

So basically it would be really useful to have a Journey / Mission Planner integrated in the game, where we can really thing about all the abstract mission requirements and steps. Some kind of graphical way to represent a plan like:

1. Take off

... DeltaV required to exit atmosphere: X

2. Stage

3. Jettison fairings

4. Circularize

... DeltaV required to circularize : Y

5. Deploy solar panels

6. Burn for Eve

... DeltaV required : Z

7. Stage while on direct intercept trajectory

8. Midcourse corrections

9. Deploy deep space experiments

10. Enter Eve SOI

11. Final adjustments to atmospheric angle

12. Retract solar panels

13. Atmospheric entry

14. Jettison heat shield

15. Deploy drogues

16. Deploy parachutes

17. Deploy high altitude experiments

18. Deploy landing legs

19. Land

20. Deploy solar panels and experiments

etc.

We need a dedicated Journey / Mission planner that is linked to the craft designer, the science system and the planetary information repository! It would improve the game a lot.

PS: also this allows us to have some really cool ideas down the line, like editing logistics automation, auto-piloting, basic Visual Scripting Language template code generation directly from the planner, which we would be able to edit and expand later.

Edited November 3, 2022 by Vl3d

[Guest]

They already said KSP2 is going to feature a journey planner.

Are you going to open a new thread for every single feature and piece of the game that isn't explicitly written in the roadmap image?

[MarcAbaddon]

Well, why not discuss those features? It's detailed that it's worth a post of it own.

Just a comment since it's coming from the 'space is hard' person asking for lots of details like life support and other systems

I think having to find out some of those things on your own is actually a more fitting challenge than life support as most of it is directly linked to the core of the game (orbital mechanics). But I still think an assistant is important to have, but I would love if it shows you the math by which to calculate things like the window, ejection angles and delta-v. At the bottom so you can ignore it if you are not interested.

[Superfluous J]

I've wanted "planning nodes" for... well... ever. Since I learned that single-burn maneuvers (as opposed to the "get out of the SOI, then burn to the destination") were preferred from a dV perspective.

I've not defined a "planning node" in at least 6 months so here goes. It works just like a maneuver node, but you do it on the planet's orbit. You can drag it around, jump to next/previous orbit, etc just like maneuver nodes. Then when you get an encounter with your target world you can set an alarm for that time, launch a ship, and then have the game adapt that node to your ship, so your ship leaves the planet's SOI as close to the same way the planning node went.

It won't be perfect, but then you can learn all about mid-course corrections and the difference between having a plan and executing it.

You could even have such nodes automatically placed for upcoming transfer windows, or the game could do it in the background to come up with contracts (or whatever KSP2 has that would apply) for those worlds.

[OHara]

1 hour ago, Vl3d said:

Some kind of graphical way to represent a plan like:

Pencil and paper work nicely. 
Most of it is text, so some way to edit free-form text stored in association with each command pod, would be nice.

For understanding experimentally what the delta-V map and pork-chop plots mean,

[GoldForest]

1 hour ago, Master39 said:

They already said KSP2 is going to feature a journey planner.

Are you going to open a new thread for every single feature and piece of the game that isn't explicitly written in the roadmap image?

We should not settle for anything less than a full list of features and changes from KSP 1. And we should get it now. /This was a joke

Anyway, joking aside, I think a show and tell showing off the feature would be nice. @Nate Simpson Could we see the journey planner in action? 

[Vl3d]

1 hour ago, Master39 said:

Are you going to open a new thread for every single feature and piece of the game that isn't explicitly written in the roadmap image?

Actually it's not only the roadmap.. I've been searching for the journey / mission planner in the gameplay images also.

And yes, I think it's good to have a thread open for discussion for most points on this list:

 

Edited November 3, 2022 by Vl3d

[Vl3d]

Imagine something like these visual mission profiles, but editable and with alarms / notifications. It would also be really cool to have something like this generated at the end of the mission.

Edited November 3, 2022 by Vl3d

[GoldForest]

6 minutes ago, Vl3d said:

Imagine something like these visual mission profiles, but editable and with alarms / notifications. It would also be really cool to have something like this generated at the end of the mission.

I don't think the journey planner will look anything like those images.

It will be in the map view and be either 1 of 2 things in my opinion:

A) It will preplan multiple maneuver nodes in one of three settings: Most efficient, Quickest Arrival Time or Mix of the two. It will then automatically place those nodes where you need to perform the maneuver.

B) I suspect it will be a way to put in multiple maneuver nodes at once and see how all of them affect each other. So, let's say you planned out a journey from Kerbin to Jool, taking a somewhat direct route. You plan it out. Then you see that Duna will be in a close position to slingshot to save some Dv or make the trip a little faster at the expense of a little Dv. You move your Kerbin TJI maneuver node to make it a TDI maneuver node. You can then adjust your other nodes based off the new information.

As far as a 'replay my mission route' kind of time-lapse replay, seems something that a mod should do, as I don't think there's much interest in seeing your journey from destination A to B in time-lapse form. I could be wrong of course, but personally, I don't see much use in a time-lapse replay mod for KSP. If you want to rewatch your journey, record it yourself. 

[Vl3d]

23 minutes ago, GoldForest said:

As far as a 'replay my mission route' kind of time-lapse replay, seems something that a mod should do, as I don't think there's much interest in seeing your journey from destination A to B in time-lapse form.

That's actually your idea, not mine. I just want the planners in 2D abstract graphical form like the images . Sure, they can also be integrated in map view as maneuver / action nodes, alarms, etc. But the point is to see the whole mission in one simple image with text.

Edited November 3, 2022 by Vl3d

[GoldForest]

13 minutes ago, Vl3d said:

That's actually your idea, not mine. I just want the planners in 2D abstract graphical form like the images . Sure, they can also be integrated in map view as maneuver / action nodes, alarms, etc. But the point is to see the whole mission in one simple image with text.

So, you want ksp2 to procedurally plan, render and display an image of your mission? Yeah, I don't think that's possible in game at all.

[Vl3d]

What I'm saying it's that the system would be useful both as an interactive 2D graphical Mission Profile Planner and as a final Mission Report.

But that's just an inspiration for how it would look like. The planner itself can be represented in a lot of ways.

The most important thing is that it's integrated well with other systems like the craft designer and the scientific information repository that it has triggered notifications for when we are on the real mission.

[GoldForest]

 

28 minutes ago, Vl3d said:

What I'm saying it's that the system would be useful both as an interactive 2D graphical Mission Profile Planner and as a final Mission Report.

But that's just an inspiration for how it would look like. The planner itself can be represented in a lot of ways.

The most important thing is that it's integrated well with other systems like the craft designer and the scientific information repository that it has triggered notifications for when we are on the real mission.

I see no value or benefit in a system designed to generate 2D images for your journey. It's a waste of time and resources imo.

Now, I understand that it can be a visual aid to some, but to create a system that will read your craft file, then show it at every stage of your journey is frankly ridiculous to me.

If something like that was to be added to the game, I would like to see it as pre-made 2d images stored in the tutorial or KSPedia section, so it's not taking up processing power and and I don't have to sit around waiting for it to generate the image. Being pre-made means you won't see your exact ship, but it doesn't take up processing power, so is the best thing.

[Vl3d]

5 minutes ago, GoldForest said:

but to create a system that will read your craft file, then show it at every stage of your journey is frankly ridiculous to me.

I think it's already implemented as the orthogonal viewer in the VAB.

[GoldForest]

6 minutes ago, Vl3d said:

I think it's already implemented as the orthogonal viewer in the VAB.

The entire craft in the VAB, yes, but you're asking it to break your craft down, and know what stage is supposed to do what, which I don't see happening. Too many different ways the computer can read your craft and display all the wrong images. It could display a lunar lander as the csm. And how is it supposed to tell a spacesplane docked to an interstellar craft apart from said interstellar craft? What you're asking for is almost impossible without some major work, work better suited in other areas of development. Best you'll get is a pre-made 2D image, and not a procedurally generated one.

[Xelo]

Tbh when it comes to mission planning, I would really just be ok with just the ability to label the stages themselves in the VAB. For a relative beginner like me, Ive found the best way to counter the lack of experience on dV is just to pack 2km/s + more dV then you need to for ventures into the unknown, If you have fuel left over great! more exploration! :D. But otherwise, if you make a mistake in timing or even the direction of the orbit you can just expend more fuel. Over time as you get more familiar you naturally start making more efficient rockets and stop leaving mainsail rockets with 4km/s dv left attached to tiny relay probes in minmus orbit c: 

Comment warranty is void in  KSP1 career mode no refunds.

[Vl3d]

26 minutes ago, GoldForest said:

And how is it supposed to tell a spacesplane docked to an interstellar craft apart from said interstellar craft?

Subassemblies are already confirmed. They are saved in the mission workspace.

It looks like each stage and craft is treated as a separate entity. They can be merged to create bigger assemblies.

[GoldForest]

21 minutes ago, Vl3d said:

Subassemblies are already confirmed. They are saved in the mission workspace.

It looks like each stage and craft is treated as a separate entity. They can be merged to create bigger assemblies.

Yes, but again, the 2d image would not know what is a lander, what is a spaceplane, what is an interstellar craft, etcetera. Not unless there was some way to mark that.

But still, this system is just a waste of time and effort imo. On top of that, I see it being a pain to code, and it would just delay KSP 2 even more from official release. 

We don't really know how the assembly/sub-assembly system will work. We've been given tid bits here and there, but we don't know the whole picture. All we know is that there's the main assembly and sub-assemblies in the same workspace. We don't know if the sub-assemblies go with you when you go to launch, if they get deleted when you go to launch, etcetera. Calling the sub-assemblies 'separate entities' kind of implies the former. Which in truth would be kind of neat, being able to make three different planes or rockets and launch all of them at once, but I doubt that's how it works. 

Unless you mean it will treat two spaceplanes docked to an interstellar craft as 3 separate entities, then no. It will treat them as one giant entity since the crafts are docked.

[mcwaffles2003]

I think it might be possible if the mission planner accounted for each staging event instead of simply the maneuver nodes, but that would also be a fairly imprecise as there are a lot of dV sinks to account for. As for showing what each stage looks like, the game would only have to simulate staging, which is already set up in the VAB. Just show the rocket with the staged parts deleted in sequence. Something like this could be helpful for very complicated set ups and help new players visualize how rocket staging behaves. Not to mention, having a feature like this might help with the whole "check your staging" problems all of us have encountered a million times We could see before a launch that a parachute is staged to fire in the wrong place.

[Gargamel]

5 hours ago, Vl3d said:

And yes, I think it's good to have a thread open for discussion for most points on this list:

It’s really not, that’s why they’re in that thread.  

[Pthigrivi]

I am hoping we'll get a mission planning aid, and there even seems to be a button for it in the VAB though we haven't seen how it would work UI-wise. I actually think this kind of thing is a really critical part of getting players who get stuck in KSOI to take that next leap to interplanetary missions. The images Vl3d has shown are probably more involved than it needs or wants to be. What you want is a split-screen. On the left you could have a kind of simple subway diagram where you add a starting location (Kerbin or wherever) and could then add a string of destinations. If for instance you wanted start at Kerbin and land on Duna you'd put in "KSC" as your starting location, "Duna: Surface" as your Destination and it would display a little graphic listing: KSC > Low Kerbin Orbit > Kerbin to Duna Transfer > Low Duna Orbit > Duna Surface. Between each is a burn, so it could give dV costs for each leg and list a total or even allow you to auto-populate an alarm clock. Things like transfers could default to next window or allow you to fine tune them using a pork chop window. On the right side of the screen you would could see each leg of your journey in map view. So if you clicked on "Kerbin to Duna Transfer" on the left it would show this transfer and your start and arrival dates on the right. If this was all it was--a fancy dV estimator--that would be great all by itself. Just label and save them for future reference. BUT, if you were clever, you might also be able to attach these to individual subassemblies so that you could plan in a more fine-grained fashion or even refer back to them in flight. 

The other, other nice thing about this is it gets you 80% of the way to setting up automated supply runs. As we've talked about previously they don't actually need to be real, physical craft. If you'd completed and saved a demonstration run the game wouldn't need to literally re-fly the mission on schedule. It could simply repeat the resource transfer on plausibly scheduled intervals. 

Edited November 3, 2022 by Pthigrivi

[t_v]

46 minutes ago, mcwaffles2003 said:

I think it might be possible if the mission planner accounted for each staging event instead of simply the maneuver nodes, but that would also be a fairly imprecise as there are a lot of dV sinks to account for. As for showing what each stage looks like, the game would only have to simulate staging, which is already set up in the VAB. Just show the rocket with the staged parts deleted in sequence. Something like this could be helpful for very complicated set ups and help new players visualize how rocket staging behaves. Not to mention, having a feature like this might help with the whole "check your staging" problems all of us have encountered a million times We could see before a launch that a parachute is staged to fire in the wrong place.

It might work for basic missions, but will pose a serious problem for more complex ones. Take a mothership headed to Eve with a deployable rover and plane. The rover is attached to the top of the rocket upside-down in a fairing and the plane is on a docking port on the side of the mothership. Because the rover is the final stage, the visualizer takes that as the control point and the rocket appears upside-down on the launchpad. The booster uses a strange form of onion staging, and the delta-v readouts make the visualizer believe that all of the side boosters detach at the same time in the lower atmosphere and the transfer stage starts burning in the upper atmosphere. Assuming that the messed up dV doesn’t stop the mission from arriving at Eve, the visualizer sees the parachutes on the plane, and the plane isn’t staged to decouple because it is attached with a docking port, so the whole mothership takes a plunge. Clearly this doesn’t work in the actual game, but the visualizer has no problem with letting a space station touch down hanging from a plane by one docking port. Finally, the rover decouples from the ship in the low Eve atmosphere and touches down, heat shield and all, ignoring the “jettison heat shield” feature. All throughout, the entire ship is rendered upside-down because the rover is upside-down. 
 

And this is still a relatively simple mission. Getting these displays to work with pretty much any compound mission is going to be an absolute nightmare to get working very badly, and I think almost impossible to get working well. 

[mcwaffles2003]

1 hour ago, t_v said:

And this is still a relatively simple mission.

C'mon now, no it isn't this is intentionally complex and you can argue your points without falsely embellishing them.

1 hour ago, t_v said:

Because the rover is the final stage, the visualizer takes that as the control point and the rocket appears upside-down on the launchpad

Why would it do that if the rocket when actually sent to the launch pad is right side up? Also, if this is so easily foreseeable you don't think it would be mitigated?

1 hour ago, t_v said:

Take a mothership headed to Eve with a deployable rover and plane. The rover is attached to the top of the rocket upside-down in a fairing and the plane is on a docking port on the side of the mothership. [...] The booster uses a strange form of onion staging, and the delta-v readouts make the visualizer believe that all of the side boosters detach at the same time in the lower atmosphere and the transfer stage starts burning in the upper atmosphere. Assuming that the messed up dV doesn’t stop the mission from arriving at Eve, the visualizer sees the parachutes on the plane, and the plane isn’t staged to decouple because it is attached with a docking port, so the whole mothership takes a plunge. Clearly this doesn’t work in the actual game, but the visualizer has no problem with letting a space station touch down hanging from a plane by one docking port. Finally, the rover decouples from the ship in the low Eve atmosphere and touches down, heat shield and all, ignoring the “jettison heat shield” feature.

I don't think  a full scale simulation of the planned flight is practical, you can simply subtract everything that's not attached to the root part after each stage though and show how the created vessel transforms to assist mission planning. For instance, once the root vessel stages away from the plane, just show the root vessel without a plane. So on and so forth.

1 hour ago, t_v said:

All throughout, the entire ship is rendered upside-down because the rover is upside-down. 

>.> (see 2nd reply)

 

I get the vibe that you are looking for reasons for none of this to work and flat out ignoring any mild effort that would fix very obvious errors your proposing. If you put as much effort into doubting it's possible to make a stage by stage dV adding system you might be able to convince yourself it wouldn't be possible too, despite it currently existing in stock as an overlookable widget.

[Rutabaga22]

A mission planner would be great, but I'm not going to use it. I find my fun in ksp by doing inefficient, unplanned missions and by figuring things out as I go. 

[t_v]

6 minutes ago, mcwaffles2003 said:

C'mon now, no it isn't this is intentionally complex and you can argue your points without falsely embellishing them

Okay, let me restate: this is a relatively simple mission for me. It is just one step up from an Apollo-style mission (2 detachables instead of one) and is honestly much less in-depth than a mission I would normally  use for a single-launch Eve colonization. I dislike that you think I am making false embellishments when I was just trying to point out some technical difficulties with this kind of system to the best of my knowledge. 

9 minutes ago, mcwaffles2003 said:

Why would it do that if the rocket when actually sent to the launch pad is right side up? Also, if this is so easily foreseeable you don't think it would be mitigated?

Okay, there’s one fix. Have the starting image be the same orientation as the rocket in the VAB. When does it flip? Because if you have a mission that makes a burn in the upside-down direction (e.g. you have an ion engine on the front of the ship), you would want to flip that again. So do you calculate thrust direction for each stage and orient the craft based on that. What about maneuvers using RCS engines then? This problem is already pretty complex, and needs to be solved for a visualizer to work. 

12 minutes ago, mcwaffles2003 said:

I don't think  a full scale simulation of the planned flight is practical, you can simply subtract everything that's not attached to the root part after each stage though and show how the created vessel transforms to assist mission planning. For instance, once the root vessel stages away from the plane, just show the root vessel without a plane. So on and so forth

As I said before, the plane is attached via a docking port, so that it can re-dock to the station.  There’s nothing in the staging that suggests if or when the plane detaches from the mothership. So the assumption is that it stays attached and the two land together. The alternative is that another mission with a drone that is meant to stay attached and parachute a base down decouples and suddenly a surface base is left in orbit. 
 

If you just want stages of the rocket separating and you don’t care about where those stages separate, that could solve a lot of these problems. You wouldn’t be showing burn information, location at each of those separations, or any other information that would be useful for a mission planner, but it might be reasonably achievable, still with a lot of issues. 
 

17 minutes ago, mcwaffles2003 said:

I get the vibe that you are looking for reasons for none of this to work and flat out ignoring any mild effort that would fix very obvious errors your proposing. If you put as much effort into doubting it's possible to make a stage by stage dV adding system you might be able to convince yourself it wouldn't be possible too, despite it currently existing in stock as an overlookable widget.

You are getting the wrong vibe. I would really, really like for this to be in the game. It would make for such a special experience to see your own rocket represented in the same professional style as rockets that have millions of real-life bills invested in them. If this could possibly work, I would strongly advocate for it. But just like planetary collisions, there is no way that you can program a tool to take all the possibilities and represent them well in the scope of KSP 2. The sheer amount of craft that people can design means that any system you create, no matter how well developed, will break when someone tries creating something new like a stack separator loop. 
 

I think that Pthigrivi’s idea of mission planning is much more feasible- we already have good delta-v readouts on ships, at least for 99% of cases, and doing things like pork chop plots is a problem that has been solved. Full-trip visualizers are not going to happen, only numerical trip planning and maybe some partial, restricted visualizers.