An Adaptive Estimation Technique for Finite Escape Time Issue in Mobile Robot Navigation

This paper deals with the analysis of H infinity Filter for mobile robot navigation considering the Finite Escape Time problem. H infinity Filter has disadvantages in navigation as the current states of estimation may lead to erroneous results due to the finite escape time issue. This is a case where the estimation may become erroneuous due to huge amount of state covariance produced by the updated state covariance during mobile robot observations. Therefore, the state covariance is modified to avoid the occurrence of finite escape time when mobile robot moves in a surrounding area. The updated state covariance is changed by checking the state covariance matrix through the Cholesky function to provide a reliable positive semidefinite matrix during the observation process. Simulation results have described that the performances are better compared to the normal H infinity Filter which illustrates a number of finite escape time during mobile robot movements.