im going off of the midpoint formula and i want my ai to stay inbetween two points that change dynamically(which is why i cant haave a set path)
I currently have this setup
It works but its not going smoothly, would there be a way of doing this smoother?
The ai has to stop before going to the next destination
the rotate task is suopposed to have “player_location” and not “location”
Last screenshot is a bit blurry. You don’t need to break vector. You can just add them and divide by 2 directly.
As for why it’s not smoother, you’re not showing us the code that you’re using to move the actor. Are you using AIMoveTo? Are you setting the position directly?
One method might be to create an invisible actor and set the position directly to that and set the AI to follow that actor. The engine should take care of the rest. Once it reaches the invisible actor, you’ll have to call AIMoveTo again. So you could set some kind of overlap event or something.
Set IsFollowing to false when there’s an overlap. If you get an end overlap event and IsFollowing is false, set it to true and call AIMoveTo again.
i used the aimoveto in the behaviour tree? i showed that? the location gets set correctly through location blackboard value?
Ah sorry. I’m not familiar enough with behavior trees, especially with how it uses AiMoveTo. Hopefully someone else can help you.
You may want to clarify if you want to continuously go to the midpoint. Or if you want to go to a calculated midpoint A… once it reaches A, calculated midpoint B and then goes to B and you want the transition to be smoother. If it’s option 1, you need to follow an invisible actor and continuously update the invisible actor’s position. If it’s option 2, you need to have point B ready BEFORE you reach point A. Otherwise it has no idea where to go and must stop. There’s a way to make an actor follow a custom path and update it dynamically, but that’s more complicated.
Good luck!
