Multi-sensor based state estimation is still challenging because sensors deliver correct measures only for nominal conditions (for example the observation of a camera can be identified for a bright and non smoggy day and illumination conditions may change during the tracking process). It results that the fusion process must handle with different probability density functions (pdf) provided by several sensors. This fusion step is a key operation into the estimation process and several operators (addition, multiplication, mean, median,...) can be used, which advantages and drawbacks. This chapter presents the probabilistic framework of state estimation from several sensors and more specifically, stochastic approaches that approximate the state distribution as a set of samples. Finally, several simple fusion operators are presented and compared with an original algorithm called M2SIR, on both synthetic and real data.