Parallel Kalman Smoothing Augmented Lagrangian Method for Constrained State Estimation

This paper presents a novel parallel computing framework for solving constrained state estimation problems. The proposed methodology integrates an augmented Lagrangian formulation with a parallel Kalman filter and smoother implementation, specifically designed to optimize the primal update step through efficient parallel computation. Experimental results demonstrate that our approach provides a significant computational time improvement while maintaining the estimation accuracy.