Timeline for Creating orientation Quaternion from forward vector
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 16, 2018 at 7:18 | comment | added | sam hocevar | @off99555 then there is one piece of missing information: the roll value, otherwise many values for the right vector can be good, especially the one parallel to the ground. | |
| Oct 15, 2018 at 19:44 | comment | added | offchan | I want right and up vector that is relative to the object entirely. So the right vector needs to not be parallel to the ground like this. I think your code gives right vector that is parallel to the ground. | |
| Oct 15, 2018 at 12:39 | comment | added | DMGregory♦ | Similar to gimbal lock, the method you propose has a singularity when looking straight up or straight down, but it's not caused by gimbal lock in this case because we're not rotating on gimbals (dependent rotation axes). The symptoms are also different: in gimbal lock, we'll still have a valid orientation but we'll lose a degree of freedom, left with only two axes. The singularity in this construction won't cause a loss of a degree of freedom, it will cause either an invalid/underspecified orientation, or once dealt with, a discontinuous change in orientation for a continuous change in inputs. | |
| Oct 15, 2018 at 10:53 | history | answered | sam hocevar | CC BY-SA 4.0 |