3301 自律搬送車の複数台運用時における位置認識のロバスト化(OS8 無人化・遠隔監視,次世代物流システム・建設機械,オーガナイズド・セッション(OS))

概要

論文の詳細を見る
We proposed a robust method for self-localization of infrastructure-free automated guided vehicles (AGVs) using laser range finders (LRFs) in environments that include multiple vehicles. In those environments, occlusion of a LRF is caused by other vehicles, and false localization occurs with a conventional method. To solve this problem, our proposed method was divided into two stages. In first stage, to distinguish other vehicles' sides from walls, an AGV detects other vehicles with retroreflective materials from environments' shapes and reflection intensities measured by only a LRF. In second stage, to complement information for self-localization, an AGV integrates measurement results of a LRF and information of wheel rotation using Kalman Filter. Our method has an advantage that it doesn't need any additional infrastructures or electronic devices.
一般社団法人日本機械学会の論文
2010-12-01