Showroom Motorcycle 4 DoF Motion sim

Yesterday I picked up the aluminum clamps from the workshop. Feels really cool seeing the design come to life!



Looking forward to mounting and testing, but first, vacation!

Hello Chris, could you please explain the criteria you used to calculate the length of the rods at the back, the angle, and the anchor point where they are mounted? I’m currently planning my rig with Fusion360 and, well, I’m not really sure… I can’t seem to figure it out from your pictures alone. Thank you very much for that.

Honestly I don’t remember putting too much time into that. I think we just wanted the servo arms be 90 degrees to the rods and the rods be 90 degrees to each other at rest. They sit on aluminum extrusion so they are easily adjustable. I think I lowered the rear a bit from the original position to get a good feel on the bike.

Mounted the new clamps to the rig. Great fit!
Haven't really done the full force test yet, but they feel solid enough. The rubber mount for the steering actuator is working itself a little loose though so I have to do something about that. Maybe just solid mount, the motor doesn't seem to be making as much noise as I expected.

Have you investigated further into the torque steering system?

It has been implemented a while now. The current setup is version two of the load sensing steering.

could you point me in the direction of some reading you used or what issues you had to solve to implement it?

I had a theory that a lot of the physics of steering a bike could be solved with how software reacts to the inputs at different stages of speed and the corner. i thought most if the leaning and weight transfer could be managed with the 6dof of a vr headset. Haven't gotten further than that until some bike game opens up those controls to the inputs. Maybe GP bikes already attempts this? I'm brand new to biking in games so am only working with the milestone franchises atm...

GPbikes has a "directsteer" function which I used at first with a normal ffb unit. It is supposed to simulate countersteering by connecting the ffb unit straight to the physics engine of the game pretty much, taking the input as torque input. To me it is as close as any game have come to "the real thing", but does not work very well after version beta19 of the game.
That is what I used for this video:



The issue for me is that there is too much movement of the handlebars required, since it calculates input torque based on the angle input. That's why I wanted to build my own solution.

I do use the VR headset position for estimating rider weight transfer, but I feel I need better measurements. There's a whole lot of communication between the rider and bike that doesn't necessarily mean the rider is moving it's head much. Especially at low speeds.
The idea is to read the roll torque of the upper platform and from there derive the inputs and see how that differs from the head displacement.

Just noticed I didn't really respond to this.
The torque sensing itself is pretty straightforward. The first version had load cells in the handlebars, so I had to have two of everything and then "mix the signals". This is now switched to a single in-line loadcell. Simpler and less noise.
I didn't really read much, just tried to convert my own experience of riding motorcycles into arduino code. Which has been and still is a challenge.

Have you looked at slime body trackers for VR? They're cheap to purchase or its open source so you can make your own. That might help with body transfer where head isn't involved. https://docs.slimevr.dev/ I'd imagine it would make sense attaching them to your hips vs shoulders.

I have thought a lot about this and while body trackers in theory should be possible to use, I have decided to go straight to the source and measure the actual effects of the body movements.
Maybe a bad decision, we will see
My rig is free to move around the rear joint, so I should get good data reading from the front connecting points.

Hallo to all my name is Fabrizio i have a big question , for the simulator you use the mecanic motor , why you don't use direct drice motor like the ones for the handlebar so you fell the ffb on the bike ?
sorry for the stupid question im a chef ...ahahaha

Hello Fabrizio!
Short answer is; Money.

This is the cheapest option available. Direct drive servos would be fantastic, but they would also need to be extremely powerful = expensive.
Linear actuators that use ball screws are the more reasonable option. Since the ball screws have very low losses they can give you much more precise feedback. My motors are quite crude.

Next bike sim I make will most likely use ball screw actuators instead.

Thank for your answee Chris i'm looking for make a new bike project with a ffb system direct drive , but as i told you i'm a chef with a lot of idea ....there is the possibility you 'll help me ? In this project ? I can come allo to you in france im from verona italy

Ideas and passion will take you a long way! That's how I started too.

I'll help if I can. I'm in Sweden, a bit of a drive from Verona

Finally got my ass out of the wagon (is that even an expression?) and mounted the load cells to the upper platform. This will allow me to use the riders physical movements as steering input in a better way.
Hopefully it will make it easier and more intuitive to use for more people.



Well, I do need to connect them and write the code still, but at least steps were taken!

Im happy to hear that...its a lot of time test i imagine it ..i'll let you know waths 'll happen ..thank a lot

Now the weight shift sensors are in place and code has the functionality somewhat where I want it.
Yesterday I realized I’m starting to feel “finished” with this sim. Just one more technical solution going on there, then I don’t have more plans. Feels strange. I have learned heaps that I hopefully will bring with me in the next version platform not too far in the future.



These are parts of the last function. Can you guess?

How can you compress the springs to install them? I have four gas dampers with a stroke power of 300N, but I’m unable to compress them for installation. Unfortunately, I’m running out of ideas.

When my platform in its max position the springs are fully extended so easy to install