{"references": ["", "Breitenberg, V.: Vehicles: Experiments in Synthetic Psychology. MIT, Cambridge, MA (1984).", "T. Duckett, U. Nehmzow, Learning to predict sonar readings for mobile robot landmark selection, Internal Report, University of Manchester, Manchester, UK, 1999.", "Fleischer, J., Marsland, S., Shapiro, J., Sensory anticipation for autonomous selection of robot landmarks, M. Butz et al. (Eds.): Anticipatory Behavior ..., LNAI 2684, pp. 201-221, Springer-Verlag 2003.", "Khepera 2, User Manual, K-Team, S.A. 2001, http://ftp.k-team.com/khepera/documentation/Kh2UserManual.pdf", "Marsland, S., Nehmzow, U., and Duckett, T. (2001). Learning to select distinctive landmarks for mobile robot navigation. Robotics and Autonomous Systems, 37:241-260.", "Tani, J.: Model-based learning for mobile robot navigation from the dynamical systems perspective. IEEE Trans. Syst. Man Cybern., Part B, Cybern. 26(3), 421-436 (1996)."]}

This paper presents a predictive model of sensor readings for mobile robot. The model predicts sensor readings for given time horizon based on current sensor readings and velocities of wheels assumed for this horizon. Similar models for such anticipation have been proposed in the literature. The novelty of the model presented in the paper comes from the fact that its structure takes into account physical phenomena and is not just a black box, for example a neural network. From this point of view it may be regarded as a semi-phenomenological model. The model is developed for the Khepera robot, but after certain modifications, it may be applied for any robot with distance sensors such as infrared or ultrasonic sensors.