A marginalised particle filter with variational inference for non‐linear state‐space models with Gaussian mixture noise

This work proposes a marginalised particle filter with variational inference for non-linear state-space models (SSMs) Gaussian mixture noise. A latent variable indicating the component of considered at each time instant is introduced to specify measurement mode SSM. The resulting joint posterior distribution state vector, and parameters noise respect variables. then approximated by using an appropriate filter. conditionally on system are finally updated Bayesian inference. simulation study conducted compare proposed method state-of-the-art approaches in context positioning urban canyons global navigation satellite systems.