Kalman filter with time-variable gain for a multisensor fusion system

The development of the sensor fusion algorithm for visual control of mobile robots is presented. The visual sensor gives many kinds of valuable information and is especially important for the autonomously controlled mobile robots. The output data from the visual sensor include a time-lag due to the image processing computation. Moreover, in most cases, the sampling rate of the visual sensor is considerably low so that it should be used with other sensors to control fast motion. The method developed enables the sensor fusion system to give the optimal state estimates. A kind of multi-rate Kalman filter which treats the slow sampling rate data from visual sensor is applied for the construction of the sensor fusion system. An important feature of the multi-rate version of Kalman filter is the use of time-varying filter gain matrix. An experimental mobile robot is developed to realize the sensor fusion system combined with a vision sensor system and an inertial sensor system.