Host: The Japan Society of Mechanical Engineers
Name : [in Japanese]
Date : May 10, 2017 - May 13, 2017
Service robots require map data that describe the structure of the environment for localization and path planning. In a 2D wide environment mapping, there may be a problem of inconsistency in a map since it cannot associate and misaligned the same locations that are temporally apart. This is a so-called loop-closing problem; the robot has to have a loop-closure function to solve this problem. We propose a mapping method that optimizes a gridmap by making correspondence with past and current locations using a WiFi signal strength similarity. The map can be used for estimating a user location with a smartphone using WiFi signals. We developed a robot call system that dispatches a robot to the estimated user location and makes the robot guide the user to a selected location by using the built map.
