Dynamic Coordinate Locally Aligned Warping (DCLAW) [Grethlein and Ontañón, 2020] is an extension to the multivariate time series distance measure called Dynamic Time Warping (DTW) [Sakoe and Chiba, 1978]. Both Multivariate Dependent DTW [Shokoohi-Yekta et al., 2017] and DCLAW are used to numerically compare how alike multiple multivariate time series samples are to one another. 

The motivation behind DCLAW was to numerically compare driving simulator recordings stored in time series samples that may present in different lengths (one individual drives a planned route in the driving simulator faster, another individual may drive the same route slower) to one another using the spatial trajectories of the simulated vehicles traversing a planned route in the driving simulator as alignment criteria. That is, comparing how multiple individuals drove the same stretches of road, aligning the frames of time series samples by where spatially they were recorded.

DTW, while a long-standing favorite tool of time series analysts to compare time series samples, is prone to extreme warping of both time series samples being compared in order to "brute force" pattern matching. For example, a behavior at the begginning of one long time series sample may be "brute force" matched to a similar behavior at the end of another long time series sample. This phenomenon inherent to unrestricted DTW is called "spurious warping".