Semi-automated map creation for fast deployment of AGV fleets in modern logistics

详细信息    查看全文

• 作者：Patric Beinschob^a ; ^{patric.beinschob@sick.de}Author Vitae ; Mark Meyer^b ; ^{mark.meyer@tuhh.de}Author Vitae ; Christoph Reinke^a ; ^{christoph.reinke@sick.de}Author Vitae ; Valerio Digani^c ; ^{digani.v@elettric80.it}Author Vitae ; Cristian Secchi^d ; ^{cristian.secchi@unimore.it}Author Vitae ; Lorenzo Sabattini^d ; ^{lorenzo.sabattini@unimore.it}Author Vitae
• 关键词：AGVs ; 3D mapping ; Roadmap creation
• 刊名：Robotics and Autonomous Systems
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：87
• 期：Complete
• 页码：281-295
• 全文大小：6236 K
• 卷排序：87

文摘

Today, Automated Guided Vehicles (AGVs) still have a low market share in logistics, compared to manual forklifts. We identified one of the main bottlenecks in the rather long deployment time which involves precise 2D mapping of the plant, 3D geo-referencing of pick-up/ drop positions and the manual design of the roadmap. The long deployment time has various reasons: in state-of-the-art plant installations, designated infrastructure is still necessary for the localization; the mapping process requires highly skilled personnel; in many cases unavailable or inappropriate position information of drop points for goods must be corrected on site. Finally, the design of the roadmap, performed by expert technicians is manually optimised in a tedious process to achieve maximum flow of goods for the plant operator.In total the setup of a plant to be ready for AGV operation is taking several months, binding highly skilled technicians and involves very time-consuming and costly on-site procedures.Therefore, we present an approach to AGV deployment which aims to drastically reduce the time, cost and involved personnel. First, we propose the employment of a novel, industrial-ready SICK 3D laser scanning technology in order to build an accurate and consistent virtual representation of the whole warehouse environment. By utilizing suitable segmentation and processing a semantic map is generated that contains 3D geo-referenced positions as well as a 2D localization map eliminating the need for dedicated solution to 2D mapping. Second, the semantic map provides a free space map which is used as a basis for the automatic roadmap creation in order to achieve optimal flow.So, this paper proposes an innovative methodology for obtaining, in a semi-automated manner, the map of an industrial environment where a system of multiple AGVs will be installed with less time and cost.