The "meaning" of an empirical model of a physical system such as a mobile robot is an ill-defined concept, and clearly it would strengthen any hypotheses based on empirical models if some formal model verification was possible. In this paper, we present experiments on empirical modelling of mobile robot operation, in which the interactions of Scitos G5 and Magellan Pro mobile robots with purposefully designed environments are measured and modelled through system identification. The experimental setups chosen were such that we could determine from theoretical considerations what the models should be. The comparison between the actually obtained empirical models and the theoretically correct solutions demonstrates that, in the experiments conducted, the obtained empirical models are "correct". © 2010 Springer-Verlag.
CITATION STYLE
Nehmzow, U., McKerrow, P. J., & Billings, S. A. (2010). Do empirical models of robot-environment interaction have a meaning? In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6226 LNAI, pp. 25–37). https://doi.org/10.1007/978-3-642-15193-4_3
Mendeley helps you to discover research relevant for your work.