State Feedback Control Using Particle Filter

SUMMARY Since particle filters can be used in non‐Gaussian and nonlinear system models, they have a wider range of applications than Kalman filters. In this paper, a construction method for a state feedback control system using a particle filter as an observer for probabilistic state estimation is described. In order to assure robustness to non‐Gaussian noise, a maximum a‐posteriori probability estimation extraction method and an method for evaluation of the effective sample size have been incorporated into the particle filter. The effectiveness of the constructed system was verified experimentally, and the effectiveness of the state observer constructed with the particle filter is demonstrated through a comparison with a Kalman filter.