Information Model for A-UGV Performance Measurement Standard

Automatic through autonomous - Unmanned Ground Vehicles (A-UGVs), which can make decisions and perform tasks by itself, has the potential to be applied over a wide range of manufacturing systems. Logistic assistance can be performed more autonomously and quickly, and further, it can directly perform manufacturing operations in combination with a manipulator. In order to use A-UGV efficiently in a manufacturing system, it is necessary to select A-UGV suitable for each manufacturing site. It is not just a matter of specifications and requirements, but it is necessary to check whether the A-UGV can operate. Rough floor conditions can cause numerous driving errors, and an unusual type of obstacle, such as a folk lift arm, may be difficult to detect for certain A-UGVs. To do this, a framework for evaluating the performance of A-UGV under a specific manufacturing environment is needed. ASTM International Committee F45 has been developing standards providing a basis for A-UGV manufacturers and users to compare tasks and performances of the A-UGV capabilities. This paper proposes information models for A-UGV performance measurement as part of the standard development. To this end, this paper first analyzes why standards and information models for A-UGV performance evaluation are needed from the manufacturer and user side. Manufacturers and users have different views on the performance of A-UGV, which is an obstacle to introducing A-UGV into the manufacturing site. The A-UGV performance standard can bridge this gap by the following activities: 1) to describe A-UGV capability in the same maneer, 2) to describe factory environments for A-UGV, 3) to form a common understanding of A-UGV and factory environment, and 4) to guide to use A-UGVs efficiently and safely. Such activities are meaningful only when the practical issues related to driving of A-UGV are specifically defined. This paper introduces the issues identified so far, and analyzes what elements should be reflected in the standard and information model through each issue. Then, this paper proposes an information model for A-UGV performance measurement. The information model consists of A-UGV and manufacturing environments that affect A-UGV’s driving performance. Using this model, it is possible to express the performance of A-UGV and the components of the factory, and to evaluate the possibility of introducing A-UGV. The information model is developed in UML and can be used for defining A-UGV requirements models or for duplication of manufacturing environments. As a case study, an information model object of an actual A-UGV performance testbed was developed. The developed information model object was used to design duplicates that mimic the original testbed through third parties not related to this study. Duplicated testbed was analyzed in terms of whether it fully included the capability evaluation elements of A-UGV, and through this, the validity and necessity of the standard and information model were verified.