Sensor

Looking for information on how to withdraw based on the vision of my character changes in the environment, anyone have any idea?



I need to note changes in the quantity of water in a given environment, but would like to please as I have access to this information…

There are many ways to do this. You can use bounding volumes to get all nodes in-front of the character, inspect them, and then make decisions. You can also use ray collision to check if there's a wall in-front of the character (collision avoidance). Those two tools are very powerful and can be used for quite a lot of AI algorithms.

Thank you for your help, more so I decided that in use, but I think for my case the second type is the most plausible, because the land where I look around me and I decide to go by, for example, if ingrime would be much better is perhaps impossible to give back, because viso simulate intelligent agents interacting with the environment in the midst of a disaster, like a flood. What do you think?

if you are the lazy type as i am, you could just create a camera that sees what the character sees, so check what the character sees with the same method or a adaptation of the method that the render uses to decide wathever to draw or not a object… or something like this

I do not know how this can be good because the character style is a bot, it will be checked out by an environmental framework, and what he should do is move the principle and framework to control where it should move, using very simple rules, for example if the path can get more background, otherwise go up the slopes to think that there should be a review over the ground, I do not know if the camera would identify it as simple, because I never had contact with a code so if you have this type could I go?

the camera have a "inside frustrum(object here)" or something like that, that checks if the object is in the viewrange of the camera, if yes, you can then ask the direction and the distance with math stuff, and create then a map of the already processed objects by the bot, or something like it, i guess

I do, I'm using a code that does something that is "ThirdPersonController" which identifies changes in terrain and adjusts the character based on it.



So far perfect, now I do a sort of a 'polling' to notice the change on the ground as the code already does, so it uses the rules, for example if I find a high point it should go to him, that would be a function of code that come up, because my characters are agents.

I found an interesting topic, which may be the solution, I would like to know opinions?



http://www.jmonkeyengine.com/jmeforum/index.php?topic=5073.0