Project: Intelligent AGVs for autonomous pallet Identification/Localization/Transportation in non-structured industrial Environments

Nowadays Automated Guided Vehicles (AGVs) are used in many production environments and warehouses such as the automotive industry, logistics or container harbours. In general, the AGVs are highly expensive and often not flexible. The path of an AGV is in most cases strictly defined and cannot be changed on demand. The environment, in which the AGVs are working must therefore be well structured and paths must show a high repeatability rate. Many reconstructions in the infrastructure are needed to install and make the system work. Due to high costs and their limited robustness and flexibility AGVs are not often used in medium sized enterprises and especially not in non-structured environments._x000D__x000D_Therefore the idea of the project is to develop a small, robust and flexible AGV for semi- or non-structured warehouses targeted to the needs of medium sized enterprises. The project Ps are the SME Ps Fusion Systems and Digipack and the academic P the Department of Mechanical and Structural Engineering at the University of Trento._x000D__x000D_The AGV will be able to identify and estimate the location of a pallet autonomously, to automatically reach the pallet for loading, to automatically carry the pallet and unload it at a defined location. Thereby it needs to cope with indoor and outdoor industrial environments, where artificial landmarks for AGV navigation can be missing, or achieve a low reliability level, while pallets are stored in not well defined positions. _x000D_In the past the power of digital signal processors were not sufficient to solve optimisation problems in real-time within reasonable time constraints. Today these capabilities increased to an acceptable size. Also accuracy and resolution of laser rangefinders were not sufficient to guarantee an accurate estimation of the object location. Today these have at least four times more angular resolution than few years ago. Similarly, in the past CCD cameras had a number of pixels that could be inadequate to cover the whole field of view in front of the vehicle with the resolution required. Today the number of pixels of industrial cameras are continuously increasing, thus making them an enabling technology for this application._x000D__x000D_In the field of methodology several methods/instruments were defined, but none of them could cope with the 99.9% of reliability required for industrial applications. Each method/instruments has one or several weak points that makes it interesting for a valuable algorithm/system for research, but not to develop an effective industrial application. Low reliability requires a high level of COtenance and increases the probability of danger in case of failure. However, comparing the different methods, we discovered that both the methods and the instruments are in many cases complementary to each other. This means that, if different approaches are simultaneously taken into account, and there is a capability to estimate the status/accuracy of each method and instrument output, it is also possible to combine them in order to develop an entire system with a higher level of reliability. This concept is on the basis of fusion methods that we believe will be an enabling methodology. _x000D__x000D_The academic world produces many publications in the field of AGVs in every aspect of its design and control. Manifold scientific prototypes exist. The consortium is interested in the analysis of systems regarding path planning, localisation with natural landmarks, objects/pallet identification and localisation in order to combine best technologies to develop a marketable product._x000D__x000D_Thanks to recent advances in embedded systems, sensor fusion methods, laser and CCD technology, we plan to develop an industrial system able to estimate the pallet location and to make the vehicle reach it for loading/unloading purposes in applications, where its location is not strictly defined. This will open the opportunity to automate the entire production line till the shipping via unmanned truck, wagon or even ferry loading._x000D_

Acronym	AGILE (Reference Number: 4587)
Duration	01/06/2009 - 31/05/2011
Project Topic	The project goal is to develop an intelligent Automated Guided Vehicle that offers higher flexibility paired with better reliability and safety for the use in non-structured industrial environments. The AGV will be cost-efficient and easy-to-implement to meet the needs of medium-sized enterprises.
Project Results (after finalisation)	The CO result of the project is the development of demonstrator of an Automated Guided Vehicle (AGV) that offers high flexibility paired with reliability and safety for the use in non-structured industrial environments.
Network	Eurostars
Call	Eurostars Cut-Off 2

Project partner

Number	Name	Role	Country
3	DIGIPACK S.R.L.	Partner	Italy
3	Fusion Systems GmbH	Coordinator	Germany
3	Università degli Studi di Trento (University of Trento)	Partner	Italy