[theory] Simulating G-forces with motion sim

@Pit I've been considering using AC motors so driving method is out of question - it's VFD. But I couldn't find AC motor sufficiently compact, whereas winch motors do good job given with constrained space.

@all the video below is showing reality and no theorie. Our head is our target, lets fooling him (or it?? - it's human so...)

Sorry for the bad sound, it is an old MS LifeCam...

http://en.wikipedia.org/wiki/Vestibular_system

Great discussions here guys. In our quest for realism, it is important to recognize experience...and experience from members of this community confirm that rapid movements of a "seat shaker" with one pivot and two actuators produces a very good "simulation". Commercial success of Simxperience and Frex also confirms this. Still---no reason not to look for ways to progress to even better results.

Here is one thing I would add--the physiology of simulation rests more on what we do to the human head than on other parts of the body. (Read the link.) The "seat shaker" exploits this by moving the head faster than any other part of the body. Movement of the head tells our brain what is happening when it comes to acceleration and movement.

I would also add that the "most important" motion cues are the ones that tell us that something has changed or is beginning to change--braking, turning, loss of traction. The actual acceleration force is probably less importance than the timing of it's application.

As many have said, we are in the brain fooling business. Physics may be important, but only if the physics of actual movement effect our perception. The brain is easily fooled, as every pilot knows.. Here is a link to an article that tells all the ways a pilot's perception of motion can be very different from the physics.

http://en.wikipedia.org/wiki/Sensory_illusions_in_aviation

Mention here early in the thread of the 401 Motion Simulator by Force Dynamics in New York. Arguably one of the most beautifully constructed units out there. It is almost a work of art. Like SimCraft's Apex GT unit (also a very well constructed unit) they promote a "high center of rotation" theory of movement. Here is a quote from Force Dynamics website promoting their design and being critical of other designs that we know work well.

"Again, I can't emphasize this enough. Machines with centers of rotation below the seat are essentially providing the precise opposite of the intended acceleration during the beginning of every movement." http://force-dynamics.com/

I disagree with this statement because the concept diminishes head motion. As mentioned above, moving the head is the best way to provide a motion cue. Reducing head movement by moving the center of rotation to the shoulders is not a goal I would agree with. I think one of the reasons the seat shaker concept works so well is the fact that the center of rotation is below the seat. Granted, during acceleration, the seat is not pushing on your back, but it is moving your head backwards causing you to resist with your head muscles--the same sensation you feel from acceleration.

This is the same argument I would make against your "hanging pendulum" concept where the center of rotation is nearly even with the head.

All that being said---the mind adapts--so once the "driver" learns the relationship of the simulator to the visual cues, then it will associate acceleration with sensation produced by the simulator--etc.

As bsft says---one important goal should be to keep costs down and get the most results for the least cost. The best designs are often the most simple.

All good stuff you have posted here @stroutmail! But I will add a few thoughts.

As I tried explaining before in another thread, one major reason the brain can be easily fooled is because our brains use shortcuts that we have learned from common experiences as we grew from birth. As amazing as our brains are, there is just no way it can process every bit of information our senses take in. Compare our brains to a modern computer for example, they are very, very, slow. In order to shortcut some of the processing time, we use what we have learned is true from past experience to bypass the need to process the same information over again and just except certain experiences as reality.
In the other post I showed an example of how we use shadows in a major way to shortcut processing time. I showed how you could take a ball that is traveling in a straight line and make it appear to the brain to be bouncing by only adding a shadow underneath the ball that is bouncing. Yet most people perceive that it is the ball that is bouncing even when they know it is actually the shadow that is moving up and down, not the ball.
Since a chair mover moves your body in the same way that it would be moving in reaction to how it would be in a real car, it feels proper and correct. Since actual g forces created by our simulators are small, we don’t really notice that the g forces are initially reversed from reality when the chair moves. However, rotations that occur outside the center of gravity are more noticeable.

Personally, I prefer rotations to be at the center of gravity or mass of my body, not below as in a chair mover. Not only for the reason above but a low pivot point makes for more of a bucking bronco kind of ride imho which is not what I desire. But maybe that is somewhat due to the fact that I have issues with my back already.

I completely agree with this statement! Again, it’s the brain learning and using shortcuts to make associations.

Yup. "Making associations" to motion "cues" is the secret.

Perhaps the bucking bronco feeling results from to much surge and pitch? It is my understanding these can be adjusted to lower volume/amplitude. I notice some guys really being pushed around but others look pretty smooth on the same track.

@stroutmail Great info there! Also thanks for quoting Force Dynamics. They're basically saying what I'm saying, and if seriously advanced guys say that it can't be terribly wrong.
>but it is moving your head backwards causing you to resist with your head muscles--the same sensation you feel from acceleration
Albeit, in opposite direction than it's required. In case of rapid jerking that aims to provide sense of road surface this isn't very important though.

@BlazinH
>Since actual g forces created by our simulators are small, we don’t really notice that the g forces are initially reversed from reality when the chair moves
Yes that's typically the case. In order for force-simulating sim to really work, it needs to simulate acceleraions of 0.5G and above. Otherwise it's just a more expensive motion sim that does all the same thing.

@stroutmail , you said
"I disagree with this statement because the concept diminishes head motion. As mentioned above, moving the head is the best way to provide a motion cue. "
With regards to actually moving the head to recreate or supposedly "train" a driver in game is actually going against what the head is designed for. Its for keeping balance and to force it to move goes against what you head is trying to do, and that is keep a sense of balance what the rest of your body is being moved around
Watch this guys head in the bottom right hand corner, granted its moving from shaking, but he is trying to keep a steady head while his body is moving around

This is a reason a seat mover is a good start. You are moving while the wheel and pedals stay still.
In my seat mover video, again my head moves but that is to try to counteract the body movement

I get the "feeling" of g-forces because I "think" I am under force, even though I am not.

Excellent points to add to the conversation @bsft! Yes, since your body moves in a similar fashion to being in a real car, you use the same muscle groups in countering the forces to right yourself.

bsft..You are disagreeing with me when I am agreeing with you!!!! The seat mover moves the body..and the head..the head is connected to the body..the head resists the movement and trys to remain stationary..providing the sense of acceleration forces. I am trying to learn and share here..not argue.

While discussion has its value, actually constructing something to prove or disprove a "theory" is likely to be more beneficial. There are lots of examples on this forum where people took the "suck it and see" path, which resulted in huge benefits for others following their lead. I personally prefer less talk and more action.

Mike

Agree. But suck it and see could result in try it and error, the 3DOF simulator shown in the video below was an enormous effort for building it but would boring me to death as a race simulator (as a flight sim OK).

The discussion is taking a very negative and surly turn. I for one will be very quiet going foward. Shutting down discussion is not conducive to a successful forum. Action speaks louder than words but action before inquiry and discussion is often wasteful. Open discussion and contributions from newcomers require encouragement, not argument, criticism, ridicule or cold water. I will reserve my posts in future to questions only.

I don't know, it seem very friendly and productive to me. It's a constructive arguing over variety of points of view on a number of problems regarding motion sims. People share their experiences, opinions and expertise, which is the whole point.

Regarding practice over theory, it goes without saying that practical implementaiton is what ultimately matters. However having solid theoretical basis under the project in quesiton will greatly benefit to the design. It's worth nothing that complex things can not be done adequately only using hunches and without building a concrete theory behind them. As a programmer I have exceptionally vibrant understanding of importance of the theory (to much farther extent than many people I beleive). But that's probably because programming stems from mathematics and thus being The kind of discipline that has barely any practice part to it and nearly entirely consist of theoretical knowledge. So in programming if you suck at theory then you just suck in general, and no amount of practice will anywhat measureably improve your skill. With building sims specifically that's not the case obviously - you can build a perfectly workable sim off crude blueprint and by picking components with "should be about right" kind of judgement. Yet it still benefitical to know inside out how it works, because you will be able to make adjustments to the project here and there, which isn't always possible when the rig is already finished, and when it's possible it still may take unreasonable amount of effort.

Sorry @stroutmail my comment was not intended as a rebuke or to stifle discussion, I was only expressing a personal preference. I think you are being a bit sensitive to the tones used in discussion, Again, just my opinion.

Mike

Yeah I'd join in the conversation but I think its all way too serious you should just build a Rig and have fun on it, that's all I have to say

whos is that?

Ok, apologies, I misunderstood part of it then. I do recall at least 2 commercial attempts at putting someone in a static frame, and the player wears a helmet that gets pushed around by actuators. They claim is more true motion, but too many people and health care professionals shot them down in flames. I though you were suggesting that.

As you wish @stroutmail , however, if you are going to ask questions like this , and it has been done before, there will be some more "blunt" replies if you take them like that.
Agreed, asking for build advice and suggestions is the right thing. This topic has gone into deep territory so answers from experienced builders may "knock you over".
If you choose to look at this way, then so be it.
The original thread starter @raidho36 comments that he has taken all information positively and has not taken to much offence. Considering he was offensive at the beginning.
With regards to "So in programming if you suck at theory then you just suck in general, and no amount of practice will anywhat measureably improve your skill." Others will disagree with this including those who have learned to program arduinos virtually from scratch with very basic information.
With sim building, yes I have stuck to a 2 DOF design to make it easier, but from joyrider to seat mover to full frame, its still been a challenge with each design. I have not found the perfect design and I dont expect to . I just found what works in different situations and styles of builds.