Project ATHLETE

Quick 1st impression.

Hi John
Have downloaded the Triathlete and started testing it. I'm really impressed on what you have done here. This sets new standards as I didnt think that this kind of vessel with this type of controls was even possible within the framework available. This are my first impressions.
1. When the wheels make contact and the entire baseframe starts to lift of the ground the entire vessel jumps up (at times violently) in the "air". Its probably when your routine of adjusting the contact point starts to work that this happens. This might be an inherent orbiter problem that cannot be helped and that is fine. By a little training I was actually able to minimize the "jumping" quite a bit.
2. There seems to be a mixup in controls at time. I have seen that when pressing the D key the gab in the center starts to expand and then key 1+2 doesnt have any effect. At the moment I'm not able to reproduce this consistantly - sorry.
3. Sometimes the movement (keypad + and -) of the vehicle is very slow. This happend for me more often than not. Again - I'm not able to reproduce constintantly.

At this time I'm not sure what the D key actually does - but I'm sure I will :-).

At the moment I have no ideas or suggestions to improve it - I think its very fine as it is. I can see on the Altair module cfg file that you just need to add attachement specs to your own modules if you want the ahtlete to carry those. And the thing with the "show grapplepoint"...excellent. So..."Just" need the bug fixing IMHO.

Just quick first impressions from me...I will look more into it :-)
And...again...fantastic work.
Best regards

Michael

---------- Post added at 08:05 PM ---------- Previous post was at 06:57 PM ----------

Included the scn.file for verification
Code: 
BEGIN_DESC
Contains the latest simulation state.
END_DESC

BEGIN_ENVIRONMENT
  System Sol
  Date MJD 56823.9425439292
END_ENVIRONMENT

BEGIN_FOCUS
  Ship TRIATHELETE
END_FOCUS

BEGIN_CAMERA
  TARGET TRIATHELETE
  MODE Extern
  POS 3.89 9.41 -36.23
  TRACKMODE TargetRelative
  FOV 50.00
END_CAMERA

BEGIN_HUD
  TYPE Surface
END_HUD

BEGIN_SHIPS
ALTAIRMODULE:ALTAIRTESTMODULE
  STATUS Landed Moon
  POS -33.4375002 41.1184032
  HEADING 108.58
  AFCMODE 7
  NAVFREQ 0 0
END
TRIATHELETE:TRIATHLETE
  STATUS Landed Moon
  BASE Brighton Beach:1
  POS -33.4374580 41.1188458
  HEADING 88.20
  RCSMODE 0
  AFCMODE 7
  PRPLEVEL 0:0.999647
  NAVFREQ 0 0
  XPDR 0
  OPEN 0.0000
  JOINT0 0.5000
  JOINT1 0.2131
  JOINT2 0.3847
  JOINT3 0.6027
  JOINT4 0.6803
  JOINT5 0.4935
  JOINT6 0.5000
  JOINT7 0.3003
  TURN 0.7500
END
END_SHIPS

BEGIN_ExtMFD
END
 
Last edited:
Ok Thanks. D just rotates the wheels to the sides. 1/2 opens/closes the gap. Once you press 1/2 it opens/closes til you press 1/2 again.

---------- Post added 11-28-14 at 06:54 AM ---------- Previous post was 11-27-14 at 05:20 PM ----------

I updated the dll. I adjusted the touchdown points and slowed the arms.
https://dl.dropboxusercontent.com/u/71242599/TRIATHLETE.1.zip
 
Have tested the new DLL. Concerning the "jumping" issue its much much better now. When extending the legs I can get it to jump max 3 mtrs I would say so the vessel is quickly back on solid ground. If the jumping cannot be sorted out I would say that the jumping now is an acceptable level now vs. the leg translation speed.
I have also figured out a mode of failure now that is reproducable. Starting from the enclosed scenario perform the following sequence.
Press "F3">select athlete
This is what I see
Athlete_Start.png

Press "D"
This is what I see
Athlete_D-key.png

After invoking "D" the vessel now is in a mode where it moves extremely slowly and it also "crabs" i.e. the keypad +/- now work in the x axis instead of z axis.
Also..it is as if the 1-2 controls for widening the platform now works through the D key. So pressing D now widens or narrows the platform opening depending on which jey (1/2) has been pressed.
I'm not able to reverse this behavior (i.e. pressing D a 2nd time doesn't do anything)
If D is not pressed from the start then the +/- controls works in the z axis and the speed of the vessel is right and the 1/2 keys works correctly.
So it is as if D sends the vessel in some kind of bad loop.
Hopes it all makes sense
Best regards
Michael
 
Working on it. All D should do is turn the wheels and allow it to open and close the gap. I also noticed that the force direction is based on the wheel pivot being center. I need to adjust that.
 
gattispilot said:
All D should do is turn the wheels and allow it to open and close the gap.
Click to expand...

Ahh...copy all. Makes sense...Now I understand what the purpose of "D" is :thumbup:

I'm ready to test next iteration whenever you are ready :-)

/MC
 
Have tested the new version. I think the heavy bouncing/jumping is back from the first version again. At Leg deployment/Wheel touch down I can get it to jump up with up to 16 m/s.
However...now she drives with a nice speed both in "Crab" (X axis) mode (after D is invoked) as well as before in normal Z axis movement.
Using D still sends the code to a weird place. After invoking D and I have finished translating the bases (using 1 and 2) I cant figure out a way to get it back into "normal" (Z-axis) driving mode. Hence after invoking D I can only get her to translate in X axis and not in Z axis as I would expect.
Also when she is in "idle" she actually moves a little bit. If you time accelerate up to 1000x she actually translate off site considerably.
Maybe you should have some kind of "inert" or "safe" mode that the user could check off if the user want to disable all movements during high time acc.
Like you say I'm not sure about the indepentdent leg movements either. I testes getting off a 5 m high platform by just lowering the the leg for ground contact and then opening the platform (using 1/2) and from there drive off in Z-axis direction (couldnt actually do the last part since I cant get it to drive in Z-axis direction after invoking D - as described above. I have an idea with the pivot mode though that will have to get back to - make an ilusstration or so...
For now I'm off to an X-mas party here today so I wont probably check in here until tomorrow sunday.
Best regards
Michael
 
Yes, the jumping was a little weird at first. Happy that it got fixed. Also, what is the button that picks up the cargo again? In fact, what are all the button functions? One note, I was testing it and it had joint 5 pivoting on the middle leg in the opposite direction of travel. As I turned to the right, the wheel turned outwards, while it should have pivoted inward.
 
Do you still get jumping? I will look at Joint 5. J/k attach/detach cargo from the center.

I am trying it another way but not being successful.
The parent for the animations is at the base. But If I reversed it so the parents is at the wheels I can't get the base to move smoothly

---------- Post added 11-30-14 at 07:41 AM ---------- Previous post was 11-29-14 at 08:05 PM ----------

Here is the latest:
https://dl.dropboxusercontent.com/u/71242599/TRIATHLETE.4.zip

I hope the jumping is fixed. Now joint rotate continuous 360 degrees as real one. I might add an image to the dialogue box to help id the joints
 
Have tested the latest version (ver 4). Happy to report that the jumping is now at a level where I feel it is managable.
Also the issues I have previously mentioned when invoking the "base translation mode" (key D) has been misplaced. I have now figured out how operate the Athlete properly. Made a short revisit to an earlier version and it turned out that it has all along worked OK. So sorry for all my comments about the bad things that happens when invoking the D key.
Speed now works OK. When not in "base translation mode" (i.e. D is not pressed or pressed twice) Athlete now moves along the Z-axis in slow orderly fashion which is nice.
Also I can see you have cleaned up the control dialogue box (I meant to ask what the extra turn arrows were for...but they are gone now I can see so I will refrain from that :cool:))
One thing I have noticed which might improve the experience is changing the "show grabble point" feature. The problem is that if you have to grapple something with the Athlete base platform base then the object that you pick up obscures the grapple point. I was thinking if you to the blue and green arrow grapple point added a "rod" or a thin cylinder with a height of say...3 meters. Then you could align the grapple points of the two objects whilst having the Athlete base near the ground levele which is visually more pleasing. Also if its a vertical cylinder it would be visible from 360 deg.

Finally I was thinking...Based on my experience this vehicle is somewhat complicated to operate. Maybe writing a manual for vehicle operation would help future users (and save some questions for you in this forum:lol:). So...If you want...I can help by drafting a manual of some sort and send it to you for approval/inclusion in the final product?
 
Thanks. A manual sounds great.

The show grapple points. I not sure how to change that.

What if the control box looked like this:
ATHLETEARMCONTROL_zps490f3cae.jpg


I thought about should the tire assembly rotate to match the ground. As it is you need to rotate the tires or you are tilted.

Now I need to figure how to add an image to the dialogue box
 
Last edited:
Great...I will get working on the manual.
About the controlbox. My opinion is that dialogue boxes like that shall be as small as possible and only contain the absolutely nescessary information and input fields for it to work. Additional information should be sought outside. But thats just my opinion :-)
Working with the athlete tonight gave me another idea. It would be nice to have some sort of tell back as to what state the "D" key is in. Either in "base size adjust" mode (or whatever we should call it) or in drive mode. This switch is rather important since it alters the functionality of the vehicle dramaticly. I was thingking like a yellow or red light that came up on the mesh (like a strobe or rotary light) or maybe a sign that came up in the controlbox?
 
Thanks. I added a beacon to tell you when "D" is pressed.
But now a math problem
See in the images the Y position of beacon needs to changed with the y height of the plate. On the ground the beacon needs to be at 1.1 but see how much higher it is at full elevation.
ATHLETEBEACON_zpsa0042f35.jpg


Code: 
oldy = (arm_tip[0].y);
	if (oldy > -.01) oldy = -.01;

	SetTouchdownPoints(_V(0, oldy, 1), _V(-1.6, oldy, -1), _V(1.6, oldy, -1));;


	beaconpos[0].y = abs((oldy)+1.1);//MOVE BEACON ACCORDING TO TOUCHDOWN POINTS
 
Hmmm... I'm not the right to ask....So I cant help you with that. The manual is coming along nicely. I expect that I will be done later tonight (CET) with a first draft. Will incorperate the screenshots above in the first version...but will off course update later once the whole thing is done.
Btw...forgot to say yesterday that during testing she still wanders off when in high timecompression - even if she is not the active vessel. Is it possible to stop all movement calculations when the Athlete is not the vessel in focus??
If the wandering cant be solved then the work-around can be to attach Athlete to another stationary object. UniversalCargoDeck comes in quite handy :-)
 
Last edited:
Michael_Chr said:
Btw...forgot to say yesterday that during testing she still wanders off when in high timecompression - even if she is not the active vessel. Is it possible to stop all movement calculations when the Athlete is not the vessel in focus??
If the wandering cant be solved then the work-around can be to attach Athlete to another stationary object. UniversalCargoDeck comes in quite handy :-)
Click to expand...

I changed not to move the animations unless airspeed is greater than .1. Not sure why there is movement when no thrust is applied.
 
Athlete Manual v0.1

First draftversion of the manual is here. Its not totally finished. I plan to add a small guide on how to stow and emplace the Athlete using some type of Altair Lander. Also I was trying to figure out how to add attachment points to user generated cargo for the Athlete to Grapple. Didn't quite succed so maybe we can get back to that later. When you have the time please proof the manual - I think that I got most of it right but you can never be sure :-)
https://www.asuswebstorage.com/navigate/s/FEEBF74C1C4F44318F463DB2F5692F1F4
Both a docx and a preliminary pdf version is available.
 
You must log in or register to reply here.

Similar threads

francisdrake
Project SpaceReactor
Replies
9
Views
2K
francisdrake
francisdrake
francisdrake
Project Obsidian - Outer Planets Explorer
Replies
6
Views
4K
Buck Rogers
Buck Rogers
G
Console mode
Replies
10
Views
2K
Boxx
Boxx
Matias Saibene
Update Multistage2026 - Development thread
5 6 7
Replies
137
Views
21K
gattispilot
gattispilot
johnnymanly
Problem .obj to .msh conversion acting strange
Replies
2
Views
2K
johnnymanly
johnnymanly
Back
Top