Title :
Application of joint parameter identification and state estimation to a fault-tolerant robot system
Author :
Sun, Zhen ; Yang, Zhenyu
Author_Institution :
Dept. of Energy Technol., Aalborg Univ., Esbjerg, Denmark
Abstract :
The joint parameter identification and state estimation technique is applied to develop a fault-tolerant space robot system. The potential faults in the considered system are abrupt parametric faults, which indicate that some system parameters will immediately deviate from their nominal values if a fault happens. The concerned system parameters consist of deterministic parts as well as those describing the stochastic features in the system. Due to the purpose for design of reconfigurable control, these deviated system parameters need to be identified as precisely and quickly as possible. Meanwhile, it would further simplify the reconfigurable design task and possibly speed up the system recovery, if the system state information under the new operating circumstance can be available along with faulty parameter information. The joint parameter identification and state estimation using the combined Kalman Filter and Maximum Likelihood (KF-ML) techniques is discussed and applied in this study. The simulation results on a space robot system showed that the proposed method is quite promising in providing both faulty parameter information and state estimation in a quick, accurate and robust manner.
Keywords :
Kalman filters; fault tolerance; maximum likelihood estimation; parameter estimation; robots; state estimation; Kalman Filter; abrupt parametric faults; fault tolerant robot system; joint parameter identification; maximum likelihood techniques; state estimation; stochastic features; system recovery; Integrated circuits; RNA; Robots;
Conference_Titel :
IECON 2011 - 37th Annual Conference on IEEE Industrial Electronics Society
Conference_Location :
Melbourne, VIC
Print_ISBN :
978-1-61284-969-0
DOI :
10.1109/IECON.2011.6119361