Project Overview

In my current project, I am building a machine learning pipeline to classify expressive dance movements based on Laban Movement Analysis (LMA), focusing initially on effort qualities such as “float” and “punch.” Using pose estimation and motion-derived features (velocity, acceleration, jerk), I am training classifiers to detect dominant movement qualities from real-time or recorded dance footage.

The first phase of the project focuses on motion analysis and classification. After completing the ML system, I will integrate it with a group of TurtleBot3 robots to create an immersive, responsive robotic swarm. Each robot behavior will reflect the dancer’s expressive state — using decentralized swarm algorithms to dynamically embody the movement qualities identified by the model.

If you would like to see the current state of the project, you can view the repository here: GitHub: Laban Dance Classification and Swarm Response