Fuzzy Model of Bird Flock Foraging Behavior

We present a fuzzy logic based model for the simulation of bird foraging behavior. The core of the model is based on the fuzzy model for the computer simulation of bird flocking presented by Lebar Bajec et al. The later was extended in such way that allows the simulation of bird foraging. In order to upgrade the original model it was necessary to advance the artificial world and the synthetic bird (synbird). The former was expanded with the introduction of feeding areas circular regions containing food. Similar to the synbirds they were implemented as animats, i.e. by means of a three stage transition function. The first stage of this function is responsible for the selection of information about the synbirds that are currently within the area, the second for computing the change of the available food and the third for computing the new state of the feeding area. In addition to the introduction of feeding areas we upgraded the synbirds with the notion of hunger. In order to do this two drives were added to the model’s basic drives and the synbird’s internal state was changed as well. To support the attraction of the feeding areas a new perception function was added and the action selection was upgraded so as to take into account the influences of the newly introduced drives. The behavior of the synbirds is governed by their behavior type, which can be feeding, take off, flying or landing. While the behavior type drive is responsible for the transition between behavior types, the feeding drive, with respect to the current behavior type, influences hunger, flight altitude, speed and flight direction. With the extensions made we moved the model a step closer to a more naturalistic simulation of bird behavior.