Flock2: A model for orientation-based social flocking

TL;DR

This paper introduces a novel orientation-based social flocking model that realistically simulates bird flock behaviors, including cohesion, agitation waves, and energy-efficient movements, matching observed flock shapes.

Contribution

It presents a new flocking model focusing on orientation influences, improving realism and energy efficiency over previous force-based models.

Findings

Simulates realistic flock shapes like spheres and ovoids

Demonstrates spontaneous emergence of agitation waves

Shows reduced energy consumption during turning

Abstract

The aerial flocking of birds, or murmurations, has fascinated observers while presenting many challenges to behavioral study and simulation. We examine how the periphery of murmurations remain well bounded and cohesive. We also investigate agitation waves, which occur when a flock is disturbed, developing a plausible model for how they might emerge spontaneously. To understand these behaviors a new model is presented for orientation-based social flocking. Previous methods model inter-bird dynamics by considering the neighborhood around each bird, and introducing forces for avoidance, alignment, and cohesion as three dimensional vectors that alter acceleration. Our method introduces orientation-based social flocking that treats social influences from neighbors more realistically as a desire to turn, indirectly controlling the heading in an aerodynamic model. While our model can be applied to any flocking social bird we simulate flocks of starlings, Sturnus vulgaris, and demonstrate the possibility of orientation waves in the absence of predators. Our model exhibits spherical and ovoidal flock shapes matching observation. Comparisons of our model to Reynolds' on energy consumption and frequency analysis demonstrates more realistic motions, significantly less energy use in turning, and a plausible mechanism for emergent orientation waves.