摘要

In this paper an alternative approach using adaptive lattice filters (ALF) for the on-line identification of the transfer function of a single flexible link with shifting payloads is presented. The link's dynamics are altered through the introduction of various payloads, which result in a shift of its natural frequencies. The ability of the ALF algorithm for the identification of the time-shifting system's dynamics and the tracking of its corresponding natural frequencies during their shiftings is the motivating factor behind the ALF-utilization. The provided experimental results and their comparison with the theoretical ones prove the efficacy of the proposed approach. Finally the capabilities of the proposed identification scheme are compared with the experimental results gained from the classical least mean squares algorithm (LMS) which has similar computational complexity with the ALF algorithm.

  • 出版日期2010-7

全文