Behavior-Based Herding Algorithm for Social Force Model Based Sheep Herd

Inspired by real-world sheepdog herding behavior, in this paper, four behavior-based algorithms have been proposed for the social force model-based sheep herd. First, a basic algorithm is designed where types of critical are rigorously defined. The decision made constantly checking positions these sheep. Then, on top algorithm, two extra mechanisms considered to improve performance namely dynamic far-end mechanism and pausing mechanism, thus, forming other three algorithms. helps avoid undesired circling behavior around destination area, while can greatly reduce control cost sheepdog. To validate effectiveness algorithms, comprehensive tests conducted. evaluated compared from aspects, namely, success rate, completion step, cost. Moreover, parameter analysis provided examine how different design parameters will affect algorithm. Finally, it shown that when size herd increases, as expected, takes more time effort complete herding.