Papers › NDKF: A Neural-Enhanced Distributed Kalman Filter for Nonlinear Multi-Sensor Estimation
NDKF: A Neural-Enhanced Distributed Kalman Filter for Nonlinear Multi-Sensor Estimation
Siavash Farzan
We propose a Neural-Enhanced Distributed Kalman Filter (NDKF) for multi-sensor state estimation in nonlinear systems. Unlike traditional Kalman filters that rely on explicit, linear models and centralized data fusion, the NDKF leverages neural networks to learn both the system dynamics and measurement functions directly from data. Each sensor node performs local prediction and update steps using these learned models and exchanges only compact summary information with its neighbors via a consensus-based fusion process, which reduces communication overhead and eliminates a single point of failure. Our theoretical convergence analysis establishes sufficient conditions under which the local linearizations of the neural models guarantee overall filter stability and provides a solid foundation for the proposed approach. Simulation studies on a 2D system with four partially observing nodes demonstrate that the NDKF significantly outperforms a distributed Extended Kalman Filter. These outcomes, as yielded by the proposed NDKF method, highlight its potential to improve the scalability, robustness, and accuracy of distributed state estimation in complex nonlinear environments.
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.
Tasks
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