[Request]Tutorial AI ShooterGame bots

just trying to sort out the Bots but I am stumped on the blackboard/behavior tree workflow. if possible for the nest release or just a documentation page on how things work.

Any guidance on this topic or classes would be great.

thanks again for all of the help

Hi Emile,

Thank you for your feedback. I have brought your request to our documentation team. As Victor said, it is still a work in progress and subject to change. Additionally, you may want to look into this question that was asked before, as it may have information you are looking for:




The system is still WIP, that why it doenst have any documentation and has changed a bit each release of the beta. But ive been able to reverse engineer it.
The behaviour tree is made from different logic that selects one or several tasks to run, if you look at the shooter game example it should be more or less obvious what the nodes do. The TASK, is a special class, you can create more, that encapsulates a small behaviour, and they can run un parallel, If the task returns as finished, the behaviour tree looks for the next task to do, it can also return as failed, or “in progress”, there are selection nodes that can run several tasks in parallel, calling the “tick” function, in each of them, so you could have a “find optimal position” task for the bot, and another for “shoot enemy in the face”. The Blackboard, is a memory helper, it can be used to store values for the behaviour tree itself, like the enemy, as the shootergame example does. Each of the tasks can access the memory in the Blackboard, and also the selection for the task to do is done driven by the Blackboard data(if it has enemy or not, for example).
This is all ive been able to reverse engineer looking at the headers and the shootergame example code. Of course, official documentation is what would be best, and of course, and ETA of when it will be usefull for games, as behaviour tree is BROKEN, and you cant add new nodes properly, even with it showing