About mixing FSM with behavior tree #19
Replies: 4 comments 4 replies
-
Technically yes, there should be no problem. But you need to sync the start/abortion of trees with the state switch. I do something similar by using a state machine to update a "state" variable in the Blackboard, and then only have one tree with parallel nodes which correnspond to the different states. In this way you don't worry about syncing problem and the tree becomes responsive to the state change. |
Beta Was this translation helpful? Give feedback.
-
Yep I see what you mean. By the way, I'm in similar discussions with the author of XSM. |
Beta Was this translation helpful? Give feedback.
-
I try it and the state switching works well, just like FSM does. (Please point out If I misunderstand ) In addition to these, I still don't have a proven way to end some state immediately. Maybe I should make these actions as short as possible. LOL |
Beta Was this translation helpful? Give feedback.
-
Thanks. This gives me a deeper understanding about how behavior tree should be designed. |
Beta Was this translation helpful? Give feedback.
-
Interrupting flow #6
![image](https://user-images.githubusercontent.com/36298330/116177875-2d551780-a747-11eb-8da9-fcda277e9402.png)
About the Interrupting flow issue you guys talked about before, I wonder if there is any way to combine the advantage of FSM(clear state switch) and behavior tree(low coupling, extensibility)
:)
Beta Was this translation helpful? Give feedback.
All reactions