Abstract:To solve the problems of low positioning accuracy and dependence on highprecision IMU of the existing UWBIMU positioning system for wheeled mobile robots, a localization algorithm using error state Kalman filter to integrate UWBIMUOdometer is proposed to improve the position and attitude estimation accuracy of mobile robots using linear velocity measurement of odometry and pseudomeasurement implied by the nonholonomic constraints. Meanwhile, for the nonlinear system composed of the multisensor measurement models, a detailed theoretical analysis and mathematical proof of the observability of the system is carried out by an observability rank condition analysis method based on the Lie derivative, and the conditions under which the system is locally weakly observable are concluded, which determines the required measurement outputs and control inputs for unbiased estimation of the system states. The simulation results show that when the observability conditions are satisfied, the state estimation approach proposed in this paper can effectively obtain the accurate 6DOF poses of the mobile robot and significantly improve the positioning accuracy compared with the conventional methods.