Machine learning based asynchronous computational framework for generalized Kalman filter

Although the Kalman filter algorithms are well suited to be executed on most digital systems, they become slow when applied large-scale dynamic systems. Therefore, efficient execution of for time-critical and applications is essence. This work aims address this necessity by developing a novel framework improve performance generalized with unknown inputs (GKF-UI) using multithreaded-multicore processors machine learning (ML) classification methods. An asynchronous model based OpenMP message-passing developed integrated supervised ML-based thread classifier GKF-UI algorithm enhance efficiency. The experimental results show that proposed approach can achieve up 35.5× speedup over serial single-threaded implementations no losses in accuracy or changes generality structure. Moreover, play significant role realizations computational advantages systems as prediction applications.