• Welcome to PiBoSo Official Forum. Please login or sign up.
 

Motion Simulator/ Controller Whishlist

Started by Gibbon, December 01, 2014, 12:10:57 PM

Previous topic - Next topic

doubledragoncc

Thanks for the great links guys very interesting. Much of the motion we require for a racing rig still missing but close for a good system in many ways

DD
GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls; i7 12700K 5.1GHz Z690 ASUS Strix Z690-A Mobo 32GB 3600MHz DDR4 RAM ASUS Strix RTX3080 OC 10GB DDR6X ASUS Ryujin 360 AOI Cooler ROG Thor 1200w PSU in ROG Helios Tower Case.

HornetMaX

Quote from: h106frp on December 09, 2014, 10:42:51 AM
Interesting link (.pdf) for a bike simulator and estimation of steering forces required for feedback systems, look at the design on the second page using the frame and steering from a 125.
https://hal.inria.fr/file/index/docid/515559/filename/Final.pdf

Their FFB system design is interesting (double pulley-belt). Probably less precise than proper gears, but much cheaper.

Notice that:

  • The bike model used is very very simple (and linear). Much simpler than the GPB one. No biggie for their purpose I guess.
  • The input they use is, guess what, the rider steering torque.
  • The rider steering torque is not measured (which as I said is relatively hard in general) but estimated via the FFB system. Smart. This is a way to implement what I said a while ago : a FFB sistem in which instead of having the bar/wheel position as input of the simulator and the FFB magnitude as output, one has kind of the opposite: the rider/driver applied torque is the input of the system, the bar/wheel angle is the output.
Overall, it's pretty basic stuff from a control point of view. The hardware they use looks interesting however.

Side note: Lol !! Just noticed the link address: I worked at Inria for 5-6 months (a long time ago) :)

MaX.

Gibbon

Quote from: HornetMaX on December 09, 2014, 08:16:01 PMThis is a way to implement what I said a while ago : a FFB sistem in which instead of having the bar/wheel position as input of the simulator and the FFB magnitude as output, one has kind of the opposite: the rider/driver applied torque is the input of the system, the bar/wheel angle is the output.

I would say that's the only way to do it...

HornetMaX

Quote from: wkp on December 11, 2014, 09:07:47 AM
Quote from: HornetMaX on December 09, 2014, 08:16:01 PMThis is a way to implement what I said a while ago : a FFB sistem in which instead of having the bar/wheel position as input of the simulator and the FFB magnitude as output, one has kind of the opposite: the rider/driver applied torque is the input of the system, the bar/wheel angle is the output.

I would say that's the only way to do it...
Yep, but it's definitely not the way it's done in all current steering wheels (for cars the problem is identical: in reality your input is a torque), and nobody in the car-sim world seems to be complaining.


MaX.

doubledragoncc

Very good suggestions coming out here, but you are all forgetting one major factor and that is that you are still sitting in a room and there are not all of the forces of nature being applied to you physically. This lack of natural forces make the riders input over sensitive in feeling. What I am trying to say is you can not just use written down physics, figures from books and calculating this and that wont work properly as you are not fighting mother natures forces. You have to look outside the box and think differently. My SPSS steering is totally wrong if you look at it from the real bike side of it and the theory side from physics books, BUT it works when using it!!! Now take that a step further and a motion system is not made from figures but physical feeling and plain engineering to start with but the motion system is far more involved than any of the papers have shown us yet. The shear velocities  of movement are one of the biggest problems to overcome and then in all the different directions at one time. All these University rigs are precise in moving for a simple plod around town, but put them in a real racing scenario and they fail. They are too slow and unrealistic to the actual feeling you get on a real bike. All rigs I have seen so far stay facing in a straight line as if the bike is riding on the motorway the whole time. Think about a bikes attitude as the back wheel is coming round to kiss the forks, aint no way in hell you can replicate that if the bike just stays straight!!! You have to have pure lateral movement too and also one that pivots the system at the steering head. Just turning the bars is wrong, the whole bike behind the forks swings left and right. I have been working on this problem for years and it is a very costly project. I think it will take a major chunk of money to produce a motion system that works and hope to one day do it.

I love this conversation it has some real deep thoughts in it and really makes this a forum for a true sim like GPB what it is. I have a meeting with the Regional Enterprise Manager from NatWest bank on Monday , my birthday so it has to go good lol and hope to somehow take this conversation to the next level. Wish me luck guys.

Im real busy setting up this new business but as soon as I get a bit more time than just popping in here I will post some designs to further explain what I mean.

Until then think on and keep it sunny side up.

DD
GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls; i7 12700K 5.1GHz Z690 ASUS Strix Z690-A Mobo 32GB 3600MHz DDR4 RAM ASUS Strix RTX3080 OC 10GB DDR6X ASUS Ryujin 360 AOI Cooler ROG Thor 1200w PSU in ROG Helios Tower Case.

h106frp

Good luck with the business, i'm sure you can convince the business manager and your designs appear well received by those who have constructed them.

I think we do have 2 separate camps here though, the full simulation experience and the 'decent' steering controller. I have no arguments about the requirements for full simulation, feeling the back end of the bike step out on the throttle or sliding under power requires full axis simulation and i do hope you can devise a rig for this. However i would be happy for now just to have a steering controller that gave some sense of genuine motorbike steering input beyond the left/right we get at the moment, if decent FF can allow for some feel of counter steering it would improve the experience greatly.

doubledragoncc

I fully agree h106. I have so many requests for a set of bars to fit G25/27 as I built one already and I will make this an option for those who want a handlebar system/FFB and not have to spend too much but get the basics to them. It is easy as my system can do this but will limit the plug n play to throttle, front brake and either clutch or rear brake. you simply unplug the G pedals and plug itn the bars and of you go. I can do switches for shifting too.

I just want to start with the basics so people have a real throttle and hand levers, the throttle makes it sooooooooo much better, its a major factor for a bike.

DD
GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls; i7 12700K 5.1GHz Z690 ASUS Strix Z690-A Mobo 32GB 3600MHz DDR4 RAM ASUS Strix RTX3080 OC 10GB DDR6X ASUS Ryujin 360 AOI Cooler ROG Thor 1200w PSU in ROG Helios Tower Case.

HornetMaX

Quote from: doubledragoncc on December 11, 2014, 01:50:13 PM
All rigs I have seen so far stay facing in a straight line as if the bike is riding on the motorway the whole time. Think about a bikes attitude as the back wheel is coming round to kiss the forks, aint no way in hell you can replicate that if the bike just stays straight!!!
Not sure I get it ... how does your system answer that ?!

Quote from: doubledragoncc on December 11, 2014, 01:50:13 PM
You have to have pure lateral movement too and also one that pivots the system at the steering head. Just turning the bars is wrong, the whole bike behind the forks swings left and right. I have been working on this problem for years and it is a very costly project. I think it will take a major chunk of money to produce a motion system that works and hope to one day do it.
Lateral movements ?! What for ?!

The last system had a yaw degree of freedom, which is good exactly to represent the rear coming round.

MaX.

doubledragoncc

My system does not, I was talking about systems in general not mine.

Maybe lateral is the wrong word, but for the sideways motion of the back sliding and as you say YAW is better word. Sorry for my dumbness lol.


DD

GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls; i7 12700K 5.1GHz Z690 ASUS Strix Z690-A Mobo 32GB 3600MHz DDR4 RAM ASUS Strix RTX3080 OC 10GB DDR6X ASUS Ryujin 360 AOI Cooler ROG Thor 1200w PSU in ROG Helios Tower Case.

HornetMaX

Quote from: doubledragoncc on December 11, 2014, 03:52:27 PM
My system does not, I was talking about systems in general not mine.

Maybe lateral is the wrong word, but for the sideways motion of the back sliding and as you say YAW is better word. Sorry for my dumbness lol.
Ah OK.

I was asking because some systems do actually have the possibility of doing lateral movements (e.g. the 6DOF platforms), but I don't think is something we need.

MaX.

doubledragoncc

Your right on that point. It is only the rear for if it slides out, but the front slides too!!! Makes you think right, I need to put it in pics rather than words. Its a bitch compared to a car rig lol.

DD
GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls; i7 12700K 5.1GHz Z690 ASUS Strix Z690-A Mobo 32GB 3600MHz DDR4 RAM ASUS Strix RTX3080 OC 10GB DDR6X ASUS Ryujin 360 AOI Cooler ROG Thor 1200w PSU in ROG Helios Tower Case.

Napalm Nick

Its Friday and a good day to drag up an old interesting topic that hasn't reached a result yet but is getting there :)

Spurred on by recent talks of controllers I've been on a blast from the past reading up.

Quote from: BOBR6 84 on December 01, 2014, 10:28:14 PM
I would like bars on the desk that tilt left and right as well as turn (steer) l/r for counter steering.
This gets my +1.  Every desktop design I have seen so far translates a controller style handlebar turn to an on-screen lean.  I want a controller style handlebar lean to be an on screen lean. To keep absolute simplicity as the key forget counter steer and the like.
So, on a handlebar style controller replace turn pot with lean pot, simulate gyroscopic counter forces with a big fat spring.
Simples innit?  :D

Cue mechanical engineers to show me the error of my ways.......
"The post you are writing has been written at least ten times already in the last 15ish years. Its already been reported, suggested, discussed, ignored or archived (but mostly ignored). Why are you doing it again?"

HornetMaX

Quote from: Napalm Nick on May 22, 2015, 11:45:15 AM
Quote from: BOBR6 84 on December 01, 2014, 10:28:14 PM
I would like bars on the desk that tilt left and right as well as turn (steer) l/r for counter steering.
This gets my +1.  Every desktop design I have seen so far translates a controller style handlebar turn to an on-screen lean.  I want a controller style handlebar lean to be an on screen lean. To keep absolute simplicity as the key forget counter steer and the like.
So, on a handlebar style controller replace turn pot with lean pot, simulate gyroscopic counter forces with a big fat spring.
Simples innit?  :D

Cue mechanical engineers to show me the error of my ways.......
I'm not a mech eng but I can tell you what bob is asking won't work. In GPB you have 1 input: it's either the bike lean angle or (if using DSA/DST) the steering angle/torque.
With what bob is asking you'd have 2 inputs: how would GPB use them !?

MaX.

h106frp

Just a little tease - still messing with my ideas and it is a lot of work to get right, working in game though :-) Next step is to get the input controller side working.

https://www.youtube.com/v/YfeA1X4z21s

teeds

It's alive!!  8) lol Looks amazing man. If it's working in-game what's the input controller side of things relate to?