Article Development and Experimental Validation of an Adaptive Extended Kalman Filter for the Localization of Mobile Robots IEEE英文电子书.pdf

IEEE TRANSACTIONS ON ROBOTICS AND AUTOMATION, VOL. 15, NO. 2, APRIL 1999 219 Development and Experimental Validation of an Adaptive Extended Kalman Filter for the Localization of Mobile Robots Leopoldo Jetto, Sauro Longhi, Member, IEEE, and Giuseppe Venturini Abstract—A basic requirement for an autonomous mobile robot The main drawback of absolute measures is their depen- is its capability to elaborate the sensor measures to localize itself dence on the characteristics of the environment. Possible with respect to a coordinate system. To this purpose, the data changes to environmental parameters may give rise to erro- provided by odometric and sonar sensors are here fused together by means of an extended Kalman filter. The performance of the neous interpretation of the measures provided by the localiza- filter is improved by an on line adjustment of the input and tion algorithm. measurement noise covariances obtained by a suitably defined The actual trend is to use sensors of different nature and estimation algorithm. to properly weigh the relative data according to their reliabil- Index Terms— Adaptive filtering, localization systems, sensor ity. For this purpose Kalman filtering techniques represent a fusion, wheeled mobile robots. powerful tool [8], [10], [18]–[20]. In this paper, the data provided by odometric and sonar sensors are c