Abstract: An improved interacting multiple models particle filter (IMM-PF) algorithm is proposed for multi-measurement nonlinear system with parameter perturbation. It divides the perturbation region into sub-regions and assigns each of them a particle filter. Hence the perturbation problem is converted into a multi-model filters problem. It combines the multiple measurements into a fusion value according to their likelihood function. In the simulation study, we compared it with the IMM-KF and the H-infinite filter; the results testify to its advantage over the other two methods.

Key words: IMM-PF, Parameter perturbation, Multi-measurement

Please provide your name, email address and a comment

Your Name: Affili: E-Mail Address:

Comment (Please comment in English):


Verification Code(click to change a pic):