Application of Unscented Kalman Filter in the SOC Estimation of Li-ion Battery for Autonomous Mobile Robot

Abstract

When the autonomous mobile robot (AMR) is popular in unknown environment, accurate estimation of SOC (state of charge) is becoming one of the primary challenges in autonomous mobile robots research. However, as defects of the extended Kalman filter (EKF) in nonlinear estimation, there exists estimated error, which affects the estimation accuracy, when it is adopted in nonlinear estimation of a battery system. In order to yield the higher accuracy of SOC estimation, a novel method - unscented Kalman filter (UKF) was employed in SOC estimation for a battery system. The EKF and UKF are compared through experiments. Experimental results show that the UKF is superior to the EKF in battery SOC estimation for AMR