著者
小谷内 範穗 皿田 滋 菅原 一宏
出版者
The Robotics Society of Japan
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.26, no.6, pp.514-521, 2008-09-15 (Released:2010-08-25)
参考文献数
8
被引用文献数
4 4

Authors have been conducting the research on the unmanned pile loading operation by an automated wheel loader, “Yamazumi-4.”The system is a fully self-contained autonomous system. The unmanned wheel loader has various sensors to measure the self-position, the position of pile, and the position of the dump truck to be loaded. After path planning according to the positions of the pile, the truck, and loader itself, the path following control guarantees the total task completion consists of targeting the pile, scooping the pile, carrying a bucket of pile to the truck, loading the pile to the truck, and re-going to the pile. For the accurate positioning control, RTK GPS and Moving Base GPS systems are combined to the dead reckoning based on the virtual dual wheel vehicle model. The dead reckoning calculated by the wheel velocity and steering angles can not avoid accumulating the positioning errors. Although lower rate Moving Base RTK GPS system supplies more accurate global position and heading direction data, they contain time delay. Authors developed the algorithm to detect the time delay of the GPS data by comparing to the recent dead reckoning data and to re-calculate the current position and direction from GPS data. The path following feedback system is developed to follow the line part and the curve part in the planned path. The planned curve part consists of a pair of the symmetric psuedo clothoid curves. For the sure loading task, the path following control is switched to the bucket tip position control helped by the laser range finders. In the scooping task, bucket control is switched by the reaction force measured by the hydraulic pressure sensor. Totally the whole tasks were well controlled and completed.

言及状況

外部データベース (DOI)

Twitter (1 users, 1 posts, 0 favorites)

無人自律ホイールローダ「山祇4号」によるすくい取り・積込作業 ―作業制御と軌道制御― https://t.co/bm8ZbSYThe

収集済み URL リスト