I’ve been looking around lately to implement a ledge & ladder climbing mechanic.
I seem to have got it working with the player character, except for a few bugs which i might solve sooner or later.
The interesting part that i can’t seem to figure out is, how do i make an AI character do the same.
Like i want him to climb atop a wall… then run on top of it & at the edge of the wall i want hime to jump on to a ledge, climbs on top of it and then continue moving to the next WP.
Can someone help me on how to got about doing this??
You will want to do trigger boxes and animations to climb, it wouldn’t be actually climbing through navmesh. At the top of the ledge it could be navmesh, and to jump to a ledge you will want to use a navlink proxy between the points they can jump
Thanks for the reply.
I was of the same opinion since it’ll require less computation compared to running an actor tick to detect for ladder volumes and ledges.
Still I’d like to experiment a bit and will post my findings.
If anyone knows of a more robust solution, plz do share.
Hey there. Old thread, but posting this in case anyone’s curious.
This Reddit thread suggests using Nav Links, in combination with a “Smart Link” option and force-teleport/lerping position of the Pawn once it reaches for a ladder. Basically, you’re breaking up pathing into two distinct parts: the moment before it hits the ladder, and the moment it exits the ladder. The ladder itself is not part of the NavMesh interaction technically, because the Pawn is momentarily disconnected from AI pathfinding as it’s “climbing” the ladder. After it exits the ladder, NavMesh pathfinding is resumed (by plotting a new path, so yes, a slight extra path-finding overhead is introduced).