Human towers or castells modelling. 158th European Study Groups with Industry (long report)

TL;DR

This paper models the mechanical stability of human towers called castells, using dynamical, control, and static approaches to understand factors influencing their construction and collapse, pioneering a mechanical analysis of this cultural practice.

Contribution

It introduces the first mechanical models of castells, combining dynamical, control, and static analyses to study their stability and safety.

Findings

Preliminary simulations align with castellers' qualitative perceptions.

Three complementary approaches provide insights into castell stability.

Future work suggested for more detailed modeling and safety assessment.

Abstract

Human towers or castells are human structures played in festivals mainly in Catalonia. These unique cultural and traditional displays have become very popular in the last years, but they date from the XVIII century. On 2010 they became part of the Unesco Representative List of the Intangible Cultural Heritage of Humanity. Safety is very important in the performance of castells. To this end, it is crucial to understand the mechanisms that allow a castell to be built and, more importantly, the factors that may cause its collapse. This work is focused on the mechanical aspects that make a pilar (the simplest structure in the castells) to stand. We suggest three different but complementary approaches for the running stage of a pilar (stage where it has been built and has not yet collapsed): the $N$-link pendulum as a first dynamical model, the response of the castellers as a control problem, and a static analysis to capture the feasibility of a given configuration. We include some preliminary simulations to better understand the previous approaches, which seem to match with qualitative perceptions that castellers have of a castell. Possible future developments are also discussed. To our knowledge, this work represents the first one to study the castells from a mechanical point of view.