TY - JOUR
T1 - Event-Triggered Reduced-Order Filtering for Continuous Semi-Markov Jump Systems with Imperfect Measurements
AU - Zhang, Huiyan
AU - Sun, Hao
AU - Qiu, Xuan
AU - Yang, Rongni
AU - Wang, Shuoyu
AU - Agarwal, Ramesh K.
N1 - Publisher Copyright:
© 2024 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
PY - 2024
Y1 - 2024
N2 - This article conducts the issue of event-triggered reduced-order filtering for continuous-time semi-Markov jump systems with imperfect measurements as well as randomly occurring uncertainties (ROUs). Specifically, the sojourn-time-dependent transition probability matrix (TPM) is presumed to be polytopic and a quantizer is introduced to quantize output signals aiming to reflect the reality. Both ROUs and sensor failures are generated by individual random variables belonging to be mutually independent Bernoulli-distributed white sequences. First, sufficient conditions for the existence of the event-triggered reduced-order filter are obtained by utilizing the dissipativity-based technique to ensure the asymptotical stability with a strictly dissipative performance of the filtering error system. The time-varying TPM is then fractionalized, which enhances the results as stated. Furthermore, the required reduced-order filter parameters are obtained by introducing slack symmetric matrix as well as cone complementarity linearization algorithm. The effectiveness of the suggested event-triggered reduced-order filter design method is shown through simulation results.
AB - This article conducts the issue of event-triggered reduced-order filtering for continuous-time semi-Markov jump systems with imperfect measurements as well as randomly occurring uncertainties (ROUs). Specifically, the sojourn-time-dependent transition probability matrix (TPM) is presumed to be polytopic and a quantizer is introduced to quantize output signals aiming to reflect the reality. Both ROUs and sensor failures are generated by individual random variables belonging to be mutually independent Bernoulli-distributed white sequences. First, sufficient conditions for the existence of the event-triggered reduced-order filter are obtained by utilizing the dissipativity-based technique to ensure the asymptotical stability with a strictly dissipative performance of the filtering error system. The time-varying TPM is then fractionalized, which enhances the results as stated. Furthermore, the required reduced-order filter parameters are obtained by introducing slack symmetric matrix as well as cone complementarity linearization algorithm. The effectiveness of the suggested event-triggered reduced-order filter design method is shown through simulation results.
KW - Event-triggered control
KW - quantization
KW - reduced-order filtering
KW - semi-Markov jump systems (SMJSs)
KW - sensor failures
UR - http://www.scopus.com/inward/record.url?scp=85206017712&partnerID=8YFLogxK
U2 - 10.1109/TCYB.2024.3432764
DO - 10.1109/TCYB.2024.3432764
M3 - Article
C2 - 39088500
AN - SCOPUS:85206017712
SN - 2168-2267
VL - 54
SP - 6145
EP - 6157
JO - IEEE Transactions on Cybernetics
JF - IEEE Transactions on Cybernetics
IS - 10
ER -