This paper introduces a robust recognition for mobile robot by using Kalman filter. A novel disturbance rejection method by using Kalman filter is introduced. The covariant matrix of proposed Kalman gain varies to environment. The covariant matrix of Kalman gain has to change depending on the conditions of disturbance or brightness variation. Because these signals are usually unknown, this covariant value is assumed constant, however. Therefore, enough accuracy of rejection is not obtained by this method. We propose a time-varying Kalman gain based on template and show the validity of our method for path recognition of mobile robot under changes in illumination.