Physics constrained learning of stochastic characteristics
This work addresses uncertainty in noise modeling for state estimation, which is crucial for applications like autonomous vehicles, but appears incremental as it builds on existing learning approaches.
The paper tackles the problem of accurately estimating vehicle states by learning stochastic noise characteristics, presenting a learning-based methodology with different loss functions that improves real-time estimation performance.
Accurate state estimation requires careful consideration of uncertainty surrounding the process and measurement models; these characteristics are usually not well-known and need an experienced designer to select the covariance matrices. An error in the selection of covariance matrices could impact the accuracy of the estimation algorithm and may sometimes cause the filter to diverge. Identifying noise characteristics has long been a challenging problem due to uncertainty surrounding noise sources and difficulties in systematic noise modeling. Most existing approaches try identifying unknown covariance matrices through an optimization algorithm involving innovation sequences. In recent years, learning approaches have been utilized to determine the stochastic characteristics of process and measurement models. We present a learning-based methodology with different loss functions to identify noise characteristics and test these approaches' performance for real-time vehicle state estimation