A Friction Based Social Force Model for Group Behaviors

Group behaviors of organisms can be described as the response to personally and socially acquired information. Many previous works, for example traditional social force model and alignment rule, concern the influence of neighbors' behaviors, but overlook the individual direct response to the information of environment. Here, we propose a novel friction based social force model and focus on the individual initiative, which is a direct response to environmental stimuli. By employing multi-agent methods, our model simulates a grazing case: a group of sheep graze in a meadow. We demonstrate that, guided by friction force (namely the individual initiative of unwilling to move), initiatives of individual and partners have variable influences on the benefit and motion of individuals in group behaviors. Furthermore, simulation results are consistent with two recent biological observations, that cannot be mimicked by traditional social force model and alignment rule.