Attentive Strategies for Autonomous Navigation

摘要

This paper describes a dynamic" approach to vision-based navigation in unstruc-rntured environments. The navigation system combines the information derived from in-rnternal odometry and from images collected by a single camera mounted on-board. Thernsoftware/hardware architecture presented in this paper is based on the automatic collec-rntion of images and the organization of visual features into visual graphs, covering thernwhole path. The control of the vehicle is based on the visual servoing paradigm and itrnis performed by continuously and incrementally updating the vehicle's trajectory. Thernautomatic selection of generic visual features of the scene is performed by a newly imple-rnmented multi-resolution technique, based on morphological fltering and a winner takernall" strategy. Even though this technique allows to fnd any type of generic feature, inrnour implementation we restrict the analysis to corners. The planning module is basedrnon the recognition of the current image within the current image mosaic of the visualrngraph. Results related to the navigation of a mobile robot along simple indoor paths arernpresented.