Abstract

In this study we present a method for the localization problem of Autonomously Guided Vehicles (AGVs) in the framework of well-known Simultaneously Localization and Map Building (SLAM) problem. Methods that are using odometer and IMU together still open to development because of incapability of obtaining significant noise-free measurements from ordinary IMU sensor. Nowadays localization techniques for industrial AGVs are also include specific Landmarks that brings big amount of precision for particularly indoor since that allow sparse mapping without future extraction efforts. This property is also useful for implementation of estimation tools such as Kalman filtering. Our work is based on EKF based localization of based on landmark detection, odometry and IMU measurements together. The proposed method is tested on our AGVs experimentally equipped by various sensors for some navigation scenarios successfully.