humanoid model detection

manscom

Member
hi

I want to ask how to setup the values on the CanSeeObject Task in order to get the best result for detection, specially for humanoid model.

Currently I setup a scene for testing how to use timeline on the tree, but the result return vary base on the value set on the CanSeeObject Task for example the Offset, Target Offset and the Use Target Bone.

I have record a few video, please check the below links for your reference

Please explain more details on these values for me to make a good combination for them.

Thank You!
MansCom
 
I recommend enabling Draw Debug Ray so you can see why the task is failing:


Targeting the chest is a good start but the look ray will tell you precisely what is going on.
 
I recommend enabling Draw Debug Ray so you can see why the task is failing:


Targeting the chest is a good start but the look ray will tell you precisely what is going on.
thank you for your reply, but I am really don't understand how to get a good result from this, i can't find the reason why the CanSeeObject task will sometimes miss the target and trigger other logics from the tree. I made an example, please take a look the below image and link for your reference.

 

Attachments

  • 螢幕截圖 2021-10-17 上午1.26.40.png
    螢幕截圖 2021-10-17 上午1.26.40.png
    836.2 KB · Views: 1
  • 螢幕截圖 2021-10-17 上午1.30.05.png
    螢幕截圖 2021-10-17 上午1.30.05.png
    110.3 KB · Views: 1
For deepr debugging I recommend placing a log statement within MovementUtility to determine where the task is no longer able to see. You could also try adjusting the offset/target offset.
 
Top