Automated Discovery of Operable Dynamics from Videos
This approach advances automated scientific discovery by enabling data-driven modeling of dynamics from visual data, though it appears incremental as it builds on existing methods for dynamical systems analysis.
The paper tackles the problem of automatically discovering low-dimensional, operable representations of dynamical systems directly from video, without prior domain-specific knowledge, and demonstrates its effectiveness in identifying stable equilibria, predicting natural frequencies, and detecting chaotic and limit cycle behaviors.
Dynamical systems form the foundation of scientific discovery, traditionally modeled with predefined state variables such as the angle and angular velocity, and differential equations such as the equation of motion for a single pendulum. We introduce a framework that automatically discovers a low-dimensional and operable representation of system dynamics, including a set of compact state variables that preserve the smoothness of the system dynamics and a differentiable vector field, directly from video without requiring prior domain-specific knowledge. The prominence and effectiveness of the proposed approach are demonstrated through both quantitative and qualitative analyses of a range of dynamical systems, including the identification of stable equilibria, the prediction of natural frequencies, and the detection of of chaotic and limit cycle behaviors. The results highlight the potential of our data-driven approach to advance automated scientific discovery.