Combining Kalman Filtering and Markov Localization in Network-Like Environments

Abstract

This paper presents a hybrid localization method designed for environments having the structure of a network (road networks, sewerage networks, underground mines, etc...). The method, which views localization as a problem of state estimation in a switching environment, combines the flexibility and robustness of Markov localization with the accuracy and efficiency of Kalman filtering. This is achieved by letting Markov localization handle the topological aspects of the problem, and Kalman filtering the metric aspects. The two techniques are closely coupled: the Markov model determines the Kalman filters to be initiated, and statistics computed by the Kalman filters are used to define the transition and observation probabilities in the Markov model. This approach has been applied to the problem of localizing a motor vehicle traveling on an urban road network, providing robust and accurate localization at low cost.