作者: Zhi Liu , Caiyun Mao , Jing Luo , Yun Zhang , CL Philip Chen
DOI: 10.1016/J.KNOSYS.2014.03.025
关键词:
摘要: Abstract A three-domain fuzzy wavelet network filter (3DFWNF) is proposed to the physiological tremor in robotic assisted microsurgical procedures, which bases on neural (3DFWN) for estimating modulated signals with multiple frequency components. The domain added 3DFWN handle uncertainties of signals. adaptive parameters are adjusted by using a novel particle swarm optimization (PSO) algorithm training process, namely PSO (FPSO). FPSO adopts sets described Gaussian membership function define position and velocity particles, thus all arithmetic operators updating rules used original replaced procedures defined sets. Without necessity gradients, coordinates exploration exploitation capabilities ensures quick convergence preferable global search. compared existing RBF networks. Experiments carried different situations, experimental results show superiority suppression newly filter. effectiveness accuracy also verified.