The accurate estimation of robot position and orientation in real-time is one of the fundamental challenges in mobile robotics. The Extended Kalman Filter is a nonlinear real-time recursive time domain ﬁlter that combines available sensor data to produce an accurate estimate of state, and has been successfully applied to the localization problem in mobile robotics and aircraft navigation. This report describes an Extended Kalman Filter implementa- tion for the Khepera II mobile robotics platform that seeks to produce accurate localization estimates in real-time using wheel odometry data, IR sensor range data, and compass heading data.
Mobile robots, localization
Otahal, Thomas J. and Herbert G. Tanner. "Extended Kalman Filter Implementation for the Khepera II Mobile Robot." (2009). https://digitalrepository.unm.edu/me_fsp/5