Dead reckoning navigation of a mobile robot using an indirect Kalman filter

Abstract

An encoder and a gyroscope fusion algorithm for an autonomous mobile robot (AMR) navigation using indirect Kalman filter scheme is proposed. An encoder based navigation algorithm is developed and based on the navigation algorithm, a navigation error model is derived by linear perturbation. The indirect Kalman filter for fusion algorithm is realized by applying Kalman filter to the navigation error model. Furthermore the use of indirect feedback Kalman filter, which feeds back the error estimates to the main navigation algorithm is proposed for the AMR navigation. The simulation is performed and the results show that the navigation algorithm with the indirect feedback Kalman filter can successfully combine the encoder and gyroscope measurements and estimate the position and heading angle of the AMR.