SELF-ORIENTING WITH ONLINE LEARNING OF ENVIRONMENTAL FEATURES

Evidence from recently conducted neurophysiological experiments on fre ely moving rats has revealed that the firing of the head-direction cel l ensemble predicts the future head direction in response to the vesti bular input and that visual cues strongly influence the shift of the t uning curve represented by the firing of the head-direction cell ensem ble. in this article, we investigate the possibility of using learned landmark features to self-orient an autonomous agent in a partially kn own environment. A model is suggested that incorporates an artificial head-direction system for emulating the behavior of head-direction cel l ensembles in biological systems, a lattice-based dynamic cell struct ure for categorizing and classifying environmental features, and an ex pectancy-based learning mechanism that learns to associate each head d irection with a certain environmental feature. Our experimental result s show that the suggested model is capable of correcting the drift in the orientation estimated by dead-reckoning.