simple question about goap.
is goap just fsm without any hard-coded change-state conditions? in every condition just a planner code chooses serial of state changes instead of conditions?
Not many can claim 25 years on the Internet! Join us in celebrating this milestone. Learn more about our history, and thank you for being a part of our community!
simple question about goap.
is goap just fsm without any hard-coded change-state conditions? in every condition just a planner code chooses serial of state changes instead of conditions?
A series of nested state machines is a very direct way to implement it, but it isn't required. If you're feeling creative you could come up with other ways to do it.
As nested state machines, If you have a goal, your character looks at the tasks it needs to do to accomplish the goal as a series of states and steps. Those can be applied recursively, looking at the steps to achieve a sub-goal, then the steps to achieve that sub-goal. Eventually all the sub-steps will be done and it will have accomplished the goal.
http://alumni.media.mit.edu/~jorkin/goap.html
http://alumni.media.mit.edu/~jorkin/gdc2006_orkin_jeff_fear.pdf
On 11/10/2018 at 1:30 AM, moeen k said:is goap just fsm without any hard-coded change-state conditions? in every condition just a planner code chooses serial of state changes instead of conditions?
Bit late to the party, but I find goap quite similar to path finding, except the path is not in a terrain, but across available actions to a desired end state.
Of course once you found a "path" (a sequence of actions to perform, also known as "plan"), you need something like a state machine to execute it, to convert it to behavior.