Landing accurately from Kerbin Orbit

[Neves]

Hey Space Cadets,

I am a new pilot in KSP (I LOVE THIS GAME), and I am having some issues figuring out how to land on a specific place from orbit around Kerbin. The big problem I having is figuring out when I should burn off my orbit speed and still hit the mark.

Right now, when I burn retrograde with my nav node set directly on my target, it obviously moves the landing point, so I have tried overshooting the target and then burn retrograde to de-orbit, and I always fall short or overshoot the nav node too far and then burn off all my fuel. Right now, I am just taking guesses on every flight and have had to revert A LOT. I have watched a Scott Manley video about de-orbiting and landing, and there is A LOT of math involved. (I am bad at math ) Plus, his tutorial was for the Mun, which has no atmosphere.

TLDR: How do I make a pinpoint landing from orbit around Kerbin? Maths? MechJeb? If there is a good tutorial or videos for this, I would greatly appreciate it!

Happy space racing!

Edited October 29, 2015 by Neves

[Snark]

Doing pinpoint landings on a planet with an atmosphere (e.g. Kerbin) is tricky, due to the effect of air resistance on your trajectory. It's hard to predict that in a general sense, since how much of an effect it has depends heavily on the ship design (mass, aerodynamics) as well as the way that the ship is piloted.

Basically, you have two main choices:

1. Don't care (too much) about where you land. It's not hard for an orbital ship that has the aerodynamics of a brick to choose approximately where it lands, as long as you're not too choosy, e.g. if it's OK to be off by a dozen kilometers or more. This is mainly useful for choosing land-versus-ocean to land on, or land-somewhere-fairly-near-KSC.

2. Have a steerable ship, i.e. one with plenty of lift and aero control surfaces-- i.e. a spaceplane. Then you can do steering on the way down.

What are your constraints? i.e. why do you care about the pinpoint landing, what are you trying to accomplish? Are you trying to do it with a particular ship, or are you willing to design your ship to make it easier to do? Would be helpful to have some context around the problem that you're trying to solve.

[Neves]

Doing pinpoint landings on a planet with an atmosphere (e.g. Kerbin) is tricky, due to the effect of air resistance on your trajectory. It's hard to predict that in a general sense, since how much of an effect it has depends heavily on the ship design (mass, aerodynamics) as well as the way that the ship is piloted.

Basically, you have two main choices:

1. Don't care (too much) about where you land. It's not hard for an orbital ship that has the aerodynamics of a brick to choose approximately where it lands, as long as you're not too choosy, e.g. if it's OK to be off by a dozen kilometers or more. This is mainly useful for choosing land-versus-ocean to land on, or land-somewhere-fairly-near-KSC.

2. Have a steerable ship, i.e. one with plenty of lift and aero control surfaces-- i.e. a spaceplane. Then you can do steering on the way down.

What are your constraints? i.e. why do you care about the pinpoint landing, what are you trying to accomplish? Are you trying to do it with a particular ship, or are you willing to design your ship to make it easier to do? Would be helpful to have some context around the problem that you're trying to solve.

Hi, thanks so much for your answer. I am trying to do a contract to measure temp on the surface of a specific loc on Kerbin, which I read somewhere that I shouldn't bother with, but I wanted to learn how to do it the right way in case I have to do it again later in the game. I would be willing to change the craft to make it easier to do. I am just going with a basic design using what's available. I am still figuring out all the aspects of rocket-building, so my design is probably really flawed, and I have been flying manually and not very accurately. I am very noobie, essentially know how to get into a halfway decent orbit, but landing on-the-spot escapes me at this time.

[Tex_NL]

This frequently recurring issue does not have a single 'quick fix' answer, it greatly depends on altitude, attitude and craft design. The only way to find a working answer is practice, practice, practice. F5 & F9 are your friends.

Know where on the map KSC is located and remember where you made your burn and how far you dropped your Pe. If you come short burn later, if you overshoot burn earlier and if you burn up you've dropped your Pe too low.

[mabarry3]

If you have to measure it on the surface of Kerbin, it's probably significantly easier and more cost-efficient to take a plane there than trying to land a rocket there and read after that. One thing to note about the contracts that require you to take readings on the surface: the area in which you can read and fulfill the contract parameters is VERY small. Even with a mod like Trajectories that helps adjust your flight trajectory after counting drag and all that, the chances of you landing exactly where you need to be is near on impossible without being an ace pilot, and even then, it would probably take several tries...

A plane you can make relatively early that has enough fuel to fly half way around Kerbin (aka anywhere on Kerbin that requires a surface or low altitude reading) could look something like this:

It was the first plane I made that was aerodynamically sound and got me loads of science and funds for doing contracts like these basically anywhere on Kerbin. You can use whatever science parts you want on it as well, it's a good early plane. The only thing (I think) you don't have access to incredibly early-game is the circular intake on the front, but you can replace the mk1 inline cockpit with a normal mk1 cockpit and add a radial intake to the bottom of the plane instead.

If you want the craft file, let me know. But seriously, don't try taking a rocket for a surface reading. It won't happen

[Neves]

If you have to measure it on the surface of Kerbin, it's probably significantly easier and more cost-efficient to take a plane there than trying to land a rocket there and read after that. One thing to note about the contracts that require you to take readings on the surface: the area in which you can read and fulfill the contract parameters is VERY small. Even with a mod like Trajectories that helps adjust your flight trajectory after counting drag and all that, the chances of you landing exactly where you need to be is near on impossible without being an ace pilot, and even then, it would probably take several tries...

A plane you can make relatively early that has enough fuel to fly half way around Kerbin (aka anywhere on Kerbin that requires a surface or low altitude reading) could look something like this:

http://oi64.tinypic.com/r0upeh.jpg

It was the first plane I made that was aerodynamically sound and got me loads of science and funds for doing contracts like these basically anywhere on Kerbin. You can use whatever science parts you want on it as well, it's a good early plane. The only thing (I think) you don't have access to incredibly early-game is the circular intake on the front, but you can replace the mk1 inline cockpit with a normal mk1 cockpit and add a radial intake to the bottom of the plane instead.

If you want the craft file, let me know. But seriously, don't try taking a rocket for a surface reading. It won't happen

Okay, haha. I was suspicious that I would need a spaceplane vs. a rocket. Thanks again everyone for your answers!

[Guest]

Design of the craft will depend on where you are in the tech tree and what parts are available. I think the easiest way to land on target in early career is going to be through trial and error. The difficulty is also going to depend on how far away the target is from the equator. Assuming a target that is close to the equator and that the tracking station has been upgraded, set up a node that brings your predicted landing as close to target as possible. Hit alt-f5 to save the game. Now execute the node as good as you can and see where you land. If you land short of the target, use alt-f9 to reload the game, and burn say 30 seconds later that you did the first time, more or less according to how far off you were on the first trial. Keep your burn execution as close as possible each trial, only adjust the timing. After a few trials you should get pretty close. And the more times you do this, the better feel you are going to get as to where to aim in the first place.

As you progress through the career game and more parts become available, then more advance techniques will become available. The need for understanding how to manipulate the velocity vectors on the navball in relation to the target becomes necessary. This will allow accurate, pinpoint landings every time.

[DancesWithSquirrels]

I would second the airplane notion. And add that if you land your plane some distance outside your survey area (for example, landing a bit long), you can generally taxi back to it.

Which works on airless worlds if your survey craft is a rocket equipped rover! My simplest one is a Mk1 can under a Science Jr and other instruments, with a pair of (horizontal) FL-T400 tanks on the sides, two Twitch engines, solar panels, and four RoveMax M1s. Just need to get close...

Edited October 29, 2015 by DancesWithSquirrels

[Alshain]

It is possible to land a rocket from Kerbin orbit. It's a little tricky but not THAT hard if you understand how your actions affect your trajectory. The key tips are:

1. Always overshoot your target and as you descend bring it closer. Flipping a rocket to burn prograde is not going to happen and you don't have lift to help expand your trajectory either. The only thing you really have power over is contracting your trajectory arc, no expanding it.

2. Make sure you have plenty of fuel to do this. Unlike a plane, unpowered pinpoint landings with a rocket are not likely to happen even if you have parachutes.

3. If you use parachutes to slow down, make sure you hotkey the Cut action because it is unlikely you will stop exactly over your target and you will need to make some correction burns on the final approach.

4. After that it's like landing on the Mun, but with a lot more gravity and resistance.

[G'th]

Could also just do a high orbit dive. Get up into a very high, almost vertical suborbital trajectory (IE, burn straight up until like, 900km up) and adjust it so that your reentry point is to the north west of your target. After that just ride the trajectory and get ready for one hell of a ride.

You'll still probably want some way to traverse the terrain though as unless you get lucky it still won't be accurate enough to get right on the target. And it will require some extensive heat shielding.

Another thing if you're not opposed to mods is to get the Trajectories mod which displays approximately where you're going to land after reentry. Highly useful for getting it into the ballpark so you can adjust after you're nearby the target.

But the efficient thing to do is to take a plane up to your target.

[Alshain]

Could also just do a high orbit dive. Get up into a very high' date=' almost vertical suborbital trajectory (IE, burn straight up until like, 900km up) and adjust it so that your reentry point is to the north west of your target. After that just ride the trajectory and get ready for one hell of a ride.

You'll still probably want some way to traverse the terrain though as unless you get lucky it still won't be accurate enough to get right on the target. And it will require some extensive heat shielding.

Another thing if you're not opposed to mods is to get the Trajectories mod which displays approximately where you're going to land after reentry. Highly useful for getting it into the ballpark so you can adjust after you're nearby the target.

But the efficient thing to do is to take a plane up to your target.[/quote']

That doesn't work so well in 1.0. You NEED air to slow you down and to do that you have to pass through the air for a long period of time. If you do a straight up-straight down, you will have trouble slowing down in time to deploy parachutes and even with engines you will spend an absolute ton of fuel doing it.

[Vegvisir]

Early career I always go for the first (and generally second as well) plane tech unlocks as soon as I can. With a lot of time on Kerbin, doing the "science at this location" type contracts, aeroplanes are the way to go. Downside here though is that it takes longer IRL to do it since your craft is slower. And physics/time warp is a bit iffy above x3 even with a nice and stable construction. But I quite like that part of the career.

[jiminator]

landing a rocket just is going to eat a ton of dV regardless of circumstances. add to that an atmosphere and it increases the complexity. missions like this can be done on a one shot basis but a plane really works much better, especially if you have multiple locations in the same area.

[G'th]

That doesn't work so well in 1.0. You NEED air to slow you down and to do that you have to pass through the air for a long period of time. If you do a straight up-straight down, you will have trouble slowing down in time to deploy parachutes and even with engines you will spend an absolute ton of fuel doing it.

True, but it is an option and you don't even have to do a high suborbit. A basic low altitude hop can get the same job done. Just a matter of aiming it just right.

[Neves]

I appreciate all of the discussion about my question! Right now, I think I am going to concentrate on getting enough Science from other contracts, etc. so that I can unlock some basic aviation (namely landing gear) and follow up with the missions using space planes I don't think I'm quite skilled enough yet to pinpoint a drop from Kerbin orbit using the parts I have available to me. REALLY appreciate the discussion -- very valuable to a noob!

[Alshain]

True' date=' but it is an option and you don't even have to do a high suborbit. A basic low altitude hop can get the same job done. Just a matter of aiming it just right.[/quote']

I don't think smashing into the ground at high speed is much of an option.

[Brainlord Mesomorph]

I’ve been trying to refine of the following rule of thumb for ballistic reentry directly to the KSC.

There is a peninsula that juts north from an equatorial coastline approximately 90° to the east of the KSC. I call it Reference Peninsula. I do a deorbit burn on the opposite side of the planet from Reference Peninsula, to bring my perigee to about 22 km directly over Reference Peninsula.

From there, given atmospheric friction on the way down, you should land almost directly at the KSC.

I say that I’ve been trying to refine it, because (1) in new aero, the drag of your vehicle effects the equation. And (2) the exact height of your apogee also effects it. 22 km is right from a 70 km apogee, and for every kilometer above 70, I’ve been subtracting a kilometer from 22. So a 73 kilometer apogee gives you a 19 km perigee.

But that isn’t 100 percent either, because 80 km apogee and a 12 km perigee does not work.

But I’m getting close!

[cantab]

Doesn't even need to be a spaceplane. If the survey site is reasonably close to KSC just a regular aeroplane will do fine, and money permitting you can leave it landed at the site so it might be well placed for another contract in future. If it's further away and you don't fancy a few hours flying then try a mini-shuttle like the X-37.

At least unlocking landing gear is a good idea though. Whatever you use to get to the general area, if you have wheels and an engine you can taxi to the exact location, meaning you don't need to be as precise on the original landing.