Motion Simulation with Lighthouse System

[Novo Reality]

I am developing a custom solution for integrating the Vive Lighthouse system onto a motion simulation platform. The envelope of motion is so large that a stationary setup will not track the headset properly.

We are looking to install the lighthouse system itself onto the motion platform, allowing the whole system or "room" to move. The biggest issue we are facing right now is with the automatic cut off sensors inside the base stations. We are looking for a solution to bypass or disable the automatic cut off feature in the base stations themselves to allow for them to be moved.

Is this a possibility, is there anyone who has tried this or is there a solution in place from HTC?

[bendotcom]

The motion cutoff isn't arbitrary. The base station stops emitting signals when the rotors go out of spec.

The next problem you're going to run into is that the IMUs in the tracked objects are still relative to an absolute reference.

[Novo Reality]

Thanks for the response, fortunately we don't have any tracked objects or controllers to worry about. We are working on a solution for the IMU's for the headset.

When you say "the rotors go out of spec" could you define that. Are you saying when they rotate faster or they lose tracking of each basestation?

[76561198072641873]

The spinning rotors are controlled using feedback from the frame of the base station, if you rotate the base station with respect to the inertial frame the base station is operating in it looks like an excursion in the angular velocity of the rotor that the control system will detect and try to correct. When the detected angular velocity is outside a fairly small tolerance margin the lasers will be switched off to prevent tracking on poor data from a disturbed lighthouse.

[Novo Reality]

That makes a lot of sense in why such an implementation would exist to keep tracking consistent. Is it possible to manipulate the control system to increase the tolerance or bypass the correction? Again, the feedback is really helpful. I understand our implementation is outside the typical use case for the HTC Vive.

[gevarre]

You're dealing with a mechanical issue, not a software problem. In each base station you have, in effect, two gyroscopes. Moving a spinning disk = bad.

The solution you're looking for is probably one where both the head and the platform are tracked (try strapping a controller to the platform for a quick and dirty test). Then you need to extrapolate the difference between the head movement and the platform movement and adjust accordingly.