News Forums RAIN General Discussion and Troubleshooting Attack harness multitargeting?

This topic contains 18 replies, has 3 voices, and was last updated by  Sigil 6 months, 4 weeks ago.

Viewing 4 posts - 16 through 19 (of 19 total)
  • Author
    Posts
  • #39757

    CrispyArrow
    Participant

    Sorry for answering so late, I have been redirected to a different project at the moment, so this one is on hold. However, I still want to fix this before I resume the project.

    I have done as you said and made the behaviour tree as instructed. However, this behaviour tree has no stopping criterion for the waypoint patrol. Because of this, they do choose an attack position, but they do not move to it because they are still patrolling. Here’s an image with the tree and the objects on the scene:

    View post on imgur.com

    • This reply was modified 7 months, 3 weeks ago by  CrispyArrow.
    • This reply was modified 7 months, 3 weeks ago by  CrispyArrow.
    • This reply was modified 7 months, 3 weeks ago by  CrispyArrow.
    • This reply was modified 7 months, 3 weeks ago by  CrispyArrow.
    • This reply was modified 7 months, 3 weeks ago by  CrispyArrow.
    #39808

    CrispyArrow
    Participant

    bump

    Sorry for the bumps, but I really need to fix this.

    #39851

    Mad_Mark
    Participant

    The attack harness is SWEET. That took care of my “come back and fix it later” stance on zombie-AI bunching up and looking like a hydra when they all go in for a melee kill! Thank you VERY much guys.

    Mark

    #39856

    Sigil
    Keymaster

    I’m glad you bumped this, there was an issue in the last behavior tree I posted for @CrispyArrow and I wasn’t able to get back here.

    Just in case anyone else was looking here, given the way the attack harness is setup my behavior tree won’t work, but we can rework it so that it does:

    root
       parallel (succeed: any, fail: any, tie breaker: fail)
          sequencer (repeat: forever)
             detect (form variable: attacktarget)
             custom action (class: AIChooseAttackPosition)
          sequencer
             parallel (succeed: any, fail: any, tie breaker: fail)
                custom action (repeat: until success, class: AIUpdateAttackPosition)
                waypoint patrol
                   move
             sequencer (repeat: until failure)
                custom action (class: AIUpdateAttackPosition)
                move (move target: attackposition)
                move (face target: attacktarget)

    I believe this would work. Honestly I’m not very happy with the way it turned out though. I feel like AIUpdateAttackPosition should return a success/failure, but I understand why it doesn’t (we never wanted a case where an AI took a position and never gave it back).

    I have some plans on taking our attack harness and doing further things with it. Maybe add it to the starter kit, it is pretty useful.

Viewing 4 posts - 16 through 19 (of 19 total)

You must be logged in to reply to this topic.