Abstract
This research attempts to realize information support such as displaying "Congestion predicted" or "Crossing approaching" for the driver of personal mobility. We propose to utilize annotated map for the support. The driver annotates some information to the map on the display while driving by selecting the place and appropriate information properties. This annotating work may take a lot of time and effort for the driver. Therefore, we constructed a system that detects information candidate using equipped sensors and shows the detected position and classified type to the driver. The constructed system automatically extracts moving objects such as pedestrians, or environmental change such as open/close state of doors from the LRF (Laser Range Finders) scan data on the mobility. It presents the self-position of the mobility and position of the annotation candidates with the presumed type. The experimental results show that the system is able to discover annotation candidate precisely, and type-based information helps the driver to intuitively recognize and determine the essential annotation.
