Understanding Collective Motion by Computational Modelling and Data Science

Collective motion refers to the spontaneous, coordinated movement of self-propelled agents, often leading to emergent patterns such as alignment or clustering. It appears across scales—from flocks of birds and schools of fish to cellular migration and protein self-assembly. Understanding collective motion is important not only for interpreting group behavior in biological systems but also for designing materials and robotic systems with programmable interactions. However, quantifying these dynamics can be challenging due to the complexity of the patterns involved and the high computational cost of detailed molecular simulations. In this project, we will use simplified agent-based models to simulate collective motion under controlled conditions. The goal is to characterize how interactions between individuals lead to large-scale coordination. After running simulations, 3 out of the 4 patterns predicted in 2D space: High Parallel, Dynamic Parallel, and Swarm. We used simulation date to to train machine learning model to classify different collective motion patterns based on the tri-zone parameters.