Skip to content

Performance issues when exiting after prolonged operation #1041

@Kreturn0

Description

@Kreturn0
  1. I will create the BT tree once when the process starts.
  2. Each time the behavior tree is run, a new thread will be created to execute it.
  3. The behavior tree uses parallel nodes. During normal continuous operation, motion in parallel node A is executed, and there are many recursive tree nodes within A. When an exception occurs, detection in parallel node B is executed. After prolonged operation, if an exception is detected and B exits, the interval before other behavior tree nodes wait for execution becomes longer. If no exception occurs, the execution efficiency remains normal.

Why does this phenomenon occur? Should haltNode be called before each execution? Or is it necessary to recreate the BT tree class?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions