Comparison of Estimation Techniques Using Kalman Filter and Grid-Based Filter for Linear and Non-linear System

The objective is to investigate the relative advantages and performance of grid based method. To achieve this, the grid-based method was compared with one of the known optimal filter i.e. Kalman filter. First two linear system models were tested by using these two methods. One was 1st order system and another was 2nd order system. The results obtained by these two methods were then compared. Again one nonlinear system model was tested by using these two methods. For non-linear system extended Kalman filter was used in which linearization takes place about a trajectory that was continually updated with the state estimates resulting from the measurement. For non-linear system grid-based approximation method was used. The results obtained by these two methods were then compared