I have a child sphere in a parent box. The box is made up of 6 wall colliders. It is a rock throwing game.
I need to use the parent rotation vectors to rotate the force applied to child object.
I believe this new line of code will sit in FixedUpdated();. The child’s position moves with the rotation of the parent but the force is always in the direction of the instantiation or deflection. Once the physics engine gets the force direction it is a static direction even while the parent is rotating. The child could be rotating on it’s own axis through the parent space. The child’s rotation has no effect on the force direction. Like an asteroid tumbling through space. The child trajectory is unaffected by child rotational angles but should be affected by parent rotational angles.
Lets say I throw a rock. I then face left. Rock does not change course in the real world. But in what I have the force direction controls trajectory. Since force direction doesn’t change, the trajectory follows the force direction. Now the rock goes left. Yep, it takes a turn in mid air and goes in the direction of the viewer! Well that is what the viewer sees because they rotate with the environment, as in swing around the circumference.
Funny? It is a feature I could use like throwing rocks in a room of rubber mirrors. Ouch! somebody is going to get hurt or loose points. Haven’t decided yet. I call the game ‘Rubber room’. Its more like ‘Dodge Death!’
I believe that the behaviour you are after would be achieved by using rigidbody.AddRelativeForce() instead of rigidbody.AddForce(). AddRelativeForce() applys a force in the objects local coordinates.
Or if you are using constantForce.force, change this to constantForce.relativeForce which will do the same thing