Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Uriah]

Hey everyone!

I have spent the past few days researching how to implement a mod which I'm calling AeroKinetics. It will introduce a number of procedural aerospace components, including everything to assemble multistage rockets and put payloads into orbit. The concept may sound far fetched or unrealistic, and indeed it may be difficult, but I believe it is achievable in a series of smaller steps. The concept is inspired by Kerbal Space Program, if you have never played I highly recommend it. KSP is great, but I have been disappointed in a number of aspects including how difficult it is to mod and how poor the aerodynamics and flight controls are. I have also modded in Digital Combat Simulator (DCS), but the maps are finite so going into space doesn't really work.

Instead of making hundreds of components, as in KSP, I'll make a procedural component for each standard type (listed below) which allows the player to construct complex assemblies of all shapes and sizes, with a Rocket Stage Calculator to help compute empty/gross weight and delta V per stage. Everything, including rocket design, operation (aerodynamic center, center of mass and center of thrust) and orbital mechanics (semi-major axis, inclination, periapsis, apoapsis, ecentricy, etc...)  will be based on real world data and equations and the gameplay should look and feel as authentic as possible. I plan that all of the components can be damaged or destroyed, via strain, impact or excessive heat, making dynamic pressure of ascent or reentry, critical for successful design and operation.

Download the Jupiter 246 Demo for Outerra: J246.GNU.OTX

Controls:

P                   Information display
. (period)      Hold-down force (on=1, off=0)
pgup            Increase throttle
pgdn            Decrease throttle
enter            Exit/enter rocket
G                  Google Maps

To enter a prograde orbit with minimal inclination, spawn the rocket using Google Maps to face due West. This will orient the autopilot to make an Eastward gravity turn. To make sure the rocket does not have pitching moments or go into a spin directly after launch follow these instructions carefully. After spawning the rocket, press '.' (period) to take off the hold-down force. This will allow the rocket to rotate into a vertical position. Wait until the rocket stops moving completely before turning back on the hold-down. By pressing 'P' you may see the status of the hold-down force, (on=1, off=0). Slowly increase the throttle until above 70%, this will fire the four SSME RS-25 engines, wait a few seconds for the thrust to stabilize, increase to 100% to fire the two SRBs. Release the hold-down force after the SRB thrust has stabilized. After the SRBs burnout, they will be jettisoned in JSBSim, and the mesh hidden (separation still not working). When the main External Tank is empty the first stage will be jettisoned. Enjoy your flight!









Procedurally Generated Components:

Liquid Fuel Tank (insulated or uninsulated)
Liqui Rocket Engine/Nozzle (light, regular, heavy)
Solid Rocket Booster Tank
Solid Rocket Engine/Nozzle (light, regular, heavy)
Reaction Control Thrusters
Fairing/Payload Bay (multiple shapes)
Interstage/Decoupler
Nose Cone (multiple shapes)
Aerodynamic Fin (fixed)
Aerodynamic Wing (multiple airfoils and complex leading/trailing edge options, i.e. smooth or sharp edges)
Aerodynamic Control Surface
Heat Shield (adjustable ablative tile thickness/weight)
Retractable Landing Gear
Multi-stage Parachute (add chute stages and adjust canopy sizes)

I will begin by building a simple rocket, which is put toghether from all of these components, but the components ARE NOT procedurally generated. I will need to develop the vehicle assembly user interface and flight dynamics/orbital mechanics first, and the procedural parts will of course follow with time. If anyone would like to contribute to this mod, please contact me. I would also appreciate you thoughts and suggestions!

Lastly, I would just like to say, the Outtera procedural environment is of course breathtaking, the attention to detail exceptional. I am specifically impressed with the fact that I was able to navigate the area where I grew up in North Central Washington and, without using the Google Maps panel, was able to identify many landmarks and features which I did not expect would be consistent, not to mention mountains such at Mt. Rainier which are unmistakable, very impressive! My only disappointment is the Mig was unrealistically easy to fly, but hey, its only an Alpha! Fly with the Oculus Rift DK1, it is amazing even at low resolution.

Regards,
Uriah

[Uriah]

THIS POST WILL BE CONTINUALLY UPDATED:

JSBSim FGRocket:
http://jsbsim.sourceforge.net/JSBSim/classJSBSim_1_1FGRocket.html
http://wiki.flightgear.org/JSBSim_Engines#FGRocket
http://jsbsim.sf.net/JSBSimReferenceManual.pdf
http://wiki.flightgear.org/Vostok-1

Rocketry, Physics, Orbital Mechanics:
Development of an Open Source model rocket simulation software: http://openrocket.sourceforge.net/thesis.pdf
http://www.braeunig.us/space/
http://www.braeunig.us/space/orbmech.htm

[PytonPago]

Sounds like a great plan mate !

  ... for the procedural part - if ya meant the model, maybe for the starters, a set of components would do - lets say some basic mesh and some additional for all of the possible interconnections for it (connecting bridges and structural parts, or just sizes if bases for the cones have different diameters). Those additional would be than scripted visible, if viable other components would be added at the certain positions around it. ... Slight aerodynamic influences of those would be then adjusted by the set of visible parts.

P.S.: That vostok cockpit rocks ! Id like to ride that baby down the OTs atmosphere !

[Uriah]

1


2


3


4


5


6


7


8


9

[PytonPago]

 ... you probably posted stuff whyle being logged onto the google+ service (you have to post addresses loged-off, otherwise it wont register guests and wont proceed -- the same as if ya post youtube vids from your page on it)

[Uriah]

I apologize about that, I thought using two browsers, Chrome logged into my Google account, and Firefox for posting the links would work, apparently not. It has been corrected!

-Uriah

[necro]

Wow, thats impressive. I appreciate this mod! Are this ingame screenshots of the UI and the AK overlay?

[PytonPago]

Also, do you plan to do some enhanced outer-space interactions ? ( Atmosphere bounce-off etc. )

[Uriah]

necro, no the UI is an overlay I developed, simply conceptual at this point.

I had no problem configuring a new aircraft model and defining its FDM, etc..., yet I am having a lot of difficulty getting the JSBSim FGRocket/FGNozzle/FGThruster to work, and confused how to setup the aircraft script and objdef files to include all of the rocket components from JSBSim.

(HELP ME! )

I would really appreciate some help if there is anyone who is more experienced with the FDM models, I cannot find detailed information, or any for that matter about implementing rockets for Outerra.

Pyton, JSBSim has a great deal covered already if we can figure out how to implement it. Using an atmosphere model of Earth, (elevation/pressure table), reentry by means of aerocapture and a series of parachutes or propulsive landing is the only thing that will physically slow you down enough from orbital velocity. If that isn't the case, it isn't being simulated correctly.

I do plan on using the volumetric effects coming soon to Outerra to create reentry plasma-shock boundaries (angle/Mach number), as well as write a script for thermal and aerodynamic damage and failure modes for each part. Ablative coatings, tiles, and heatshields will be needed for reentry, as well as multi-stage parachutes which deploy a small drogue chute first at high velocity.

In terms of bouncing off the atmosphere, having too shallow of a reentry angle, and too high of a velocity, your point of apoapsis (farthest orbital point) will not have been reached as you begin aerocapture during descent. So the velocity will drop from atmospheric drag, but shortly you will begin to gain elevation again, until you reach the point of apoapsis. Once you begin to descend again, and if you've lost enough velocity from the short aerocapture, your point of periapsis (closest orbital point) will have been reduced enough to reenter unharmed.

Honestly, I hope OT implements these features, but I am willing to get a head start.

I'm doing more research on the JSBSim physics model, to see what is the best method for realistic orbital mechanics, including Lagrangian points. I believe it is all possible, the model seems to be comprehensive enough from what I have read so far.

-Uriah

[PytonPago]

Yes ... im not much good in JSBSim sadly, but i know that aircrafts had some issues beyond a certain altitude. Hope there isnt something engine based in it to stay in your way.

 Still ( and this is probably for Cameny and co. ) once Sun n other solar objects are implemented, wont be there some change in coordinate systems ? Cause yet is all bound to Earth center, if im not wrong. Or will stuff positions on planets (ground vehicles and ships) be handled towards theyr centers and beyond a certain distance on a Sun based center ( or just a separate one ) with some kind of switching between them done for space-capable crafts, escaping them?

[Uriah]

Indeed, I am trying to ascertain what kind of base physics model is used to see if multibody simulation (n-body) to calculate the gravitational forces excerted on an object by multiple planets is possible.

[Uriah]

By the way, aircraft using turbine engines will loose propulsion at high altitudes. The Mig29 in Outerra uses a fuel mix to extend its altitude ceiling, but oxidizer deprivation will reduce efficiency and max thrust. The JSBSim Rocket, Nozzle and Thruster should have no problem in orbit.

I need to be able to determine the orbital elements of the vehicles trajectory; Semi-Major Axis, Eccentricity, Inclination, Argument of Periapsis, Time of Periapsis Passage, and Longitude of Ascending Node.

[necro]

Probably you are the first one who needs that. So these parameters wont be accessable because of no implementation yet. You will have to calculate them by yourself. Its possible to calculate the relative velocity and altitude. This should be enough to calculate the orbit and its apo and per. And with these dudes it should be possible to calc the incl/exccentricity.

I only know vehicle scripts yet, and in these i know you can get the position of the vehicle by get_pos() which is enough for the alt. And you can measure the difference pos to the last frame in update_frame() to get its velocity.

[Uriah]

There are a number of methods I could employ using the velocity and position state vectors which contain three components each, respectively; X position, Y position, Z position and X velocity, Y velocity, Z velocity.

I could use Gauss, Euler or Fourth Order Runge-Kutta method (RK4). RK4 is the most stable and least computationally intensive.

http://jsbsim.sourceforge.net/JSBSim/classJSBSim_1_1FGRungeKutta.html

[Uriah]

The AeroKinetics Mod will also include static objects with which to build your own custom launch complex anywhere in Outerra. I am working on all the buildings and structures of the Delta IV Launch Complex 37 in Cape Canaveral. The mod will  load the real launch complex to scale at the precise Lat/Long coordinates, which can be found by using Google Maps.

I contacted with Jon Berndt of JSBSim, and Cameni to get some answers and I am making good progress.

It sounds like multibody physics will indeed be supported in the future.

I plan on having the launch complex done and a flying rocket within a few days!

Not sure about volumetric smoke and flames. I believe the clouds update may include support for that, as OT specifically mentioned rocket trails in regards to volumetric clouds being released. I already have an animated model with material for liquid rocket shock diamonds.