Full motion simulator

[HornetMaX]

About the roll movement.
The thing is that you cant really feel that the bike is changing its roll value, cause force vector is changing it's direction along with the bike lean.

So I'll keep Fw vector along with Gravity vector.

Hmm ... I'm still not sure I understand correctly what you want to do. Let's take 2 examples:

1. Bike going straight (and staying vertical, no lean), constant speed. Track is initially flat but then there's a slope (going up): what happens to the rig ? Will it pitch backward ?

2. Flat track, bike going straight and is vertical, your rig is vertical. The virtual rider feels a vertical force of 1g, the player on the rig feels the same. Fine.
Now the virtual bike leans right: the virtual rider feels the sum of the vertical 1g + the lateral (centripetal) acceleration. What happens for the player on the rig ?

MaX.

[doubledragoncc]

I gotta agree with BOBR, it aint happening for a long time.

One thing that gets me is all this talk and so many points are forgoten.

1. For a real simulator should the rider be in full kit, leathers, helmet, boots and gloves?
2. How many serious young sim racers have the above or for a fact how many on here?
3. How much space should the system take up?
4. Who is it going to be for? (Just the rich from what most are saying)
5. Does it really have to be designed based on mathamatical equations (why not pure physical feeling? try and calculate that!!!)
6. Overall visual looks of the system (initial visual impact sets the mood of the rider. Having to put on Oculus Rift and a load of sensors or mind numbing gimics is a BIG put off and you dont have them on a real bike.

7. It should be fun!!! Not a slow moving riding in  town in traffic DOT trainer...jeeeez god help me....

All this tech talk is good on paper, but in the real world it just dont work. It took me years to figure out the right angles and way it felt as natural as possible sitting in a room. On paper my system is wrong, but in the real world when using it, it FEELS right and thats the whole point. Not one person has said about how you would simulate when the back end slides, look at a bike in 3rd person when the back end is trying to kiss your cheek, how many types of movement does that involve? A shit load!!! To simulate it you would need so many quick action rams pushing a shoving in so many directions at one time (in the right place, not just UNDER a bike) it will take a lot of space and you better have deep pockets. So why not set a realistic point and be satisfied with simulating as much as possible in a given sized area. There needs to be some kind of goal to reach without dreaming up ideas that are just too costly for any of us. If you think of the one basic difference between real and simulation is that its for indoors and most will be in normal clothes at home, sorry to the Oculus Believers but I dont want to stick something on my head, I want to feel free from that sort of thing, so what do you do for those like me? You design something where part of its sensory system depends on forcing someone to wear something in their face!!!  You have to think of the majority not the minority of people. I guess I am too old-school in many ways. I have been professionally custom building bikes for over 25 years, spent 7 years as a motorcycle medical courier and been riding for 36 years, I jsut want to get on and ride and think that is how most people want a system to be. No having to put this and that on, set up this and that (except initial control and bike settings) and RIDE the dang thing.

One thing I thought about is the best senario to test the bike and riders movement and the best I can think of is "The Corkscrew" at Laguna Seca. The uphill approach, the cork screw and then the exit curve downhill. Simulate that and you have it made lol.

Keep up all the great ideas, this is a great thread to read.

Keep it sunny side up!!!

DD

[HornetMaX]

On paper my system is wrong, but in the real world when using it, it FEELS right and thats the whole point.

@DD: your systems (the one I've seen, at least) are only input devices to GPB. The only output they use is (when they use it) FFB.
What Furious is trying to do is much more complex. The system he has designed could be used (on top of as an input device, of course) either to reproduce the bike attitude or to (try to) provide some overall feeling of the forces. The fact he plans to use a Rift also makes a big difference.

Not one person has said about how you would simulate when the back end slides, look at a bike in 3rd person when the back end is trying to kiss your cheek, how many types of movement does that involve? A shit load!!!

Depending on the setup, some things are doable: screen fixed to the ground (i.e. no occulus), 6dof platform, rotation around the vetical Z axis (yaw) could be used (it would represent something like the "bike sideslip angle").

But it's clear to all that you can't simulate what you feel on a moving bike on a rig that, in the end, does not move as much as the real bike.
The goal is, for given a "device", to use it at best as an input device (what you do with your models) and as an input+output device (what Furious plans to do).

MaX.

[doubledragoncc]

Hi Max

I think Furious is really building a great system. I have not started on the motion system as far as a fully built one. I want to start by providing a static system first that is cheaper and more can afford but will eventually have a full motion system too.

I was just pointing out that it is so hard to please everyone and that compromises may have to be made for space and cost.

In the long run if the money is no question and the right people work together a really good system is possible.

There are many things I do not say as I intend to naturally keep a few tricks up my sleeve lol.

I would love to see some photos of your system Furious it is really interesting the concept you have.  Fantastic work.

The one thing you dont really see looking at my systems is the fact that you do not sit still on it. Because of the steering system you are constantly leaning and moving on the system much as you have to on a real bike. Through moving about the mind sences it as an input too and in combination with the bike on the screen(I only use onboard view) it is surprising how much feeling you get from it. After half an hour you feel like you have riden a bike as you ar using the muscles in you arms, back, neck and upper legs quite a lot.
It might be a static system but you cant sit still on it lol.

DD

[HornetMaX]

The one thing you dont really see looking at my systems is the fact that you do not sit still on it. Because of the steering system you are constantly leaning and moving on the system much as you have to on a real bike. Through moving about the mind sences it as an input too and in combination with the bike on the screen(I only use onboard view) it is surprising how much feeling you get from it. After half an hour you feel like you have riden a bike as you ar using the muscles in you arms, back, neck and upper legs quite a lot.
It might be a static system but you cant sit still on it lol.

I do see it DD, I even think your "southern pendulum" (this is how you call it if I recall correctly) is a very clever trick for systems with no FFB.

But between your systems and his system the difference is really in the overall goal.

MaX.

[BOBR6 84]

iv just seen a horse simulator on tv for jockeys to train on..... 

well.. if you can simulate a horse, a bike should be easy lol

[doubledragoncc]

lol BOBR think I will change to horse racing lol

[BOBR6 84]

https://www.youtube.com/v/bn3F1kBHlqU

[BOBR6 84]

lol cheesy advert but.. something along those lines would suit a bike i think..

best bet would be taking a trip to japan though lol

[doubledragoncc]

Thanks for the video BOB that got me thinkin, love the chain drive idea!!! Gotta hit the drawing board lol

DD

[h106frp]

Just wondering if anyone has tried this arrangement for a simple bar controller;

Bars on normal horizontal rotational plane but with the stem on a raked vertical rotational plane with force feedback on both planes. So at low speed you push/pull on the bars and the vertical axis stays vertical and the bike just maneuvers around with low feedback forces. At speed, you initiate the turn by pushing forward on one end of the bars, the game drives the vertical and horizontal axis force feedback servos with the required force making the front want to fold and the bars tilt correctly, you push the other end of the bar to counter the force and control the turn rate before pushing back to retain upright position with the servos reducing the force accordingly.

Having owned a freestyle bar and deciding it felt weird and uncomfortable in either the turn or tilt modes this combined axis system would seem a more natural feeling solution.

Just my two penneth as i'm not sure i would ever have space for a sit on sim controller.

[doubledragoncc]

Hi h106frp, thats a name? lol sorry

the problem is that no softwar, ie bike sim exists where there are ALL the real world inputs needed that it wont matter how real you build a system, until the programmers allow for more inputs, a controller can only do so much. I could build a system that is real if the game/sim had all the different inputs controllable by a controller. Unfortunately because there are only certain controllers on the market, the programmers limit input requirement to gamepad and wheels which will never be real. Before a real control can be made the programmers will have to think outside the gamepad limit as a controller!!!

Think guys before you shoot me down

DD 

[h106frp]

Perhaps PBBikes could; the output DLL supplies a lot of data describing the simulated forces acting on the bike, i'm sure it would not take much to have the DLL as a 2 way communications with external values overriding the computed ones if required. You would just need to agree which parameters needed to flow in which directions.

h106frp was the reg of my old NC30 many years ago wish i had kept it really, i sold it and bought a road rocket 600 thats really too fast to enjoy on UK roads without a custodial sentence waiting for you on the exit of every bend

[Nitrox]

I think Furious' idea is very good. It achieves it's power from using oculus rift, thus the actual position of the rig won't matter in terms of whether it fits with what is displayed on the screen. It can focus on positioning the rig so the rider can feel the forces, he should feel. What he sees is actually another part - and oculus rift is moving with the rider.

I tried oculus rift once (ego shooter) and you absolutely loose feeling for any directions. I'm quite sure you would not notice that the bike "wheelies" but instead assume it actually is accelerating.

Also concerning corner forces, the rig does not have to apply the same lean angle as the virtual bike necessarily. I bet you can simulate the centripetal forces by reducing lean angle during cornering, while the rider still sees full lean angle and assumes he's not upright.


Can't wait to see a video of this, if you really manage to do it Maybe this could also be done with the already mentioned 2dof motion simulator. As long as it can use both axis at the same time, it should be perfect to go. Simulating bumps and stuff would require one more axis though.

[h106frp]

I think the big 'cheat' used by a lot of commercial motion platforms is that you only sense changes in acceleration and the eye can fool you into sensing the constant velocity movements. This limits the total travel required in the platform itself.