Papers › Lagrangian Gradient Regression for the Detection of Coherent Structures from Sparse...
Lagrangian Gradient Regression for the Detection of Coherent Structures from Sparse Trajectory Data
Tanner D. Harms, Steven L. Brunton, Beverley J. McKeon
The archive published only this paper's code-link row. Authors, date and abstract are from arXiv's metadata (CC0), read from the Kaggle arXiv metadata snapshot of 2026-09-12 where its title matched the archive's; the title is the archive's.
Lagrangian Coherent Structures (LCS) are flow features which are defined to objectively characterize complex fluid behavior over a finite time regardless of the orientation of the observer. Fluidic applications of LCS include geophysical, aerodynamic, biological, and bio-inspired flows -- among others -- and can be generalized to broader classes of dynamical systems. One of the prevailing paradigms for identifying LCS involves examining continuum-mechanical properties of the underlying flow. Such methods, including finite-time Lyapunov exponent (FTLE) and Lagrangian-averaged vorticity deviation (LAVD) analyses, provide consistent and physically-meaningful results but require expensive computations on a dense array of numerically integrated trajectories. Faster, more robust, sparse methods, on the other hand, are typically non-deterministic and require a-priori intuition of flow field behavior for interpretation. If LCS are to be used in the decision-making protocol of future autonomous technologies, a bridge between dense and sparse approaches must be made. This work begins to address this goal through the development of Lagrangian gradient regression (LGR), which enables the computation of deformation gradients and velocity gradients from sparse data via regression. Using LGR, velocity gradients, flow map Jacobians (thereby FTLE), and rotational Lagrangian metrics like the LAVD are accurately computed from sparse trajectory data which is orders of magnitude less dense than traditional approaches require. Moreover, there is no need to compute velocity or numerically differentiate at any point when using LGR. Therefore, it is a purely Lagrangian technique.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Results from the paper archive 2025-07-28
No leaderboard rows for this paper in the archive.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections