Square root spherical simplex radial information filter

Abstract

In order to further increase the precision of information filter, square root spherical simplex radial information filter (SRSSRIF) is put forward in the article. SRSSRIF is combined with square root filter and fused the spherical simplex radial rule (SSRR) into information filter. We utilize the rule of spherical simplex radial to choose the integral points and the computation of the corresponding weights. The efficiency of the new algorithm is demonstrated with a 3D Lorenz system and a two permanent magnet synchronous motor. Simulation results indicated that SRSSRIF can gain preferable precision compared with extended information filter (EIF) and square root unscented information filter (SRUIF) for single or multi-sensors system.