Set membership localization and mapping for autonomous navigation (Q2732935)

This paper considers the problem of simultaneous localization and mapping (SLAM) of mobile robots in unknown/uncertain environments. It is assumed that the robot is equipped with proper sensors to get relative measurements of its position with respect to the relevant elements of the environment in which it is moving and that no 3-D model is required. No statistical assumption is made on the errors that affect the sensors, but it is assumed that the errors are bounded in norm by some quantity. This leads naturally to a set-theoretic approach to the problem. Estimates of the position of the robot and the selected landmarks are derived in terms of feasible uncertainty sets, which provide a good quantitative description of the estimation errors. The development of the paper relies on the so-called set membership estimation theory. Using this theory, efficient recursive algorithms, suitable for online implementation, are derived. The theoretical results are supported by simulation experiments that show the high-level performance of the algorithms.