Variations in Angular Tracking in a Known Rotation Based of Pivot Point Placement

Post Reply
archimedes
Posts: 1
Joined: Tue Apr 30, 2019 8:26 pm

Variations in Angular Tracking in a Known Rotation Based of Pivot Point Placement

Post by archimedes » Tue Apr 30, 2019 8:57 pm

Hello,

The specs for the application I am discussing are a Trio bar and Motive 2.1

In testing my end application using the Trio bar, I saw results I was not expecting.
So I started to test my underlying assumptions.
As part of this I noticed an unexpected result when I moved a rigid body defined by 3 points through a 90 degree rotation in a flat plane.
When I used the default pivot point position for the rigid body, the resulting Euler angles tracked the rotation correctly to <.5 degrees.
When I moved the pivot point to be on one of my 3 defining points, so I could align the local initial axes with the other rigid body points
as my end application requires, and went through the same known rotation I got a rotation of 93 degrees.

I have high confidence in the way I am getting and processing the euler angles.
I also have high confidence in the 90 rotation I am moving the rigid body through.

So is there something in altering the pivot point from the default position and rotation at all that can increase angular tracking error?
If so, is this caused specifically by moving the pivot point to be one of the points defining a generic rigid body?
If so, is this caused specifically by moving the pivot point to be one of the three points defining a rigid body defined by only three points?

Thanks

steven.andrews
NaturalPoint Employee
NaturalPoint Employee
Posts: 521
Joined: Mon Jan 19, 2015 11:52 am

Re: Variations in Angular Tracking in a Known Rotation Based of Pivot Point Placement

Post by steven.andrews » Wed May 01, 2019 10:10 am

Hello archimedes,

Thank you for posting your questions. Moving the pivot point can have an effect on the rotation values you receive, since you are moving the location from which the orientation is reported. This may all depend on how close the pivot point is to the location the physical object is actually rotating from.

I see you have opened a ticket with Support. We will request some additional data through your ticket, so you can demonstrate what you are seeing.

Best regards,
Steven
--
Steven Andrews
OptiTrack | Senior Customer Support Engineer
help.naturalpoint.com
support@optitrack.com

Post Reply