Recursive parameter estimation algorithm of the Dirichlet hidden Markov model

Vaičiulytė, Jūratė; Sakalauskas, Leonidas

doi:10.1080/00949655.2019.1679144

Cited by 3 publications

(1 citation statement)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The estimated emission probability matrix affects the accuracy of the HMM model to a certain extent [ 26 ], and the reliability of determined through experience remains to be tested. In this study, is determined using the analytic hierarchy process (AHP) method.…”

Section: Determining Hmm Parameters and Recognizing Driver Intentimentioning

confidence: 99%

Clamping force control of electro–mechanical brakes based on driver intentions

Fan

et al. 2020

PLoS ONE

View full text Add to dashboard Cite

Electro-mechanical brakes (EMBs) are the future of braking systems, particularly in commercial vehicles. Therefore, it is important to design a simple EMB scheme and establish its clamping force control strategy to satisfy the demands of commercial vehicle braking systems. This study proposes a pneumatic disc-brake-based EMB for an electric bus. Its working principle was established, and the system model was analyzed. Subsequently, the hidden Markov models (HMMs) of driver decelerate and brake intentions were built and recognized based on the analytic hierarchy process (AHP). Given the time-consuming behavior of the proposed EMB to eliminate brake clearance due to the leverage effect of the arm and motor performance limitation, a clamping force control strategy factoring in the driver intentions was developed to improve the response performance without changing the structure or size of the EMB. Furthermore, simulation analyses were performed using MATLAB/ Simulink. The results confirmed that under the action of a step and 5 Hz triangular sawtooth signals, the clamping force output from the EMB corresponds well with the target signal. The clamping force gradually increases when approaching the target without overshoot and jitter during the process. The overall clamping force response time is decreased by approximately 0.25 s under the driver emergency brake than the conventional control method. Hence, the response performance of the EMB is improved.

show abstract