In connection with the development towards cyber-physical production systems, the importance of logistics processes in industrial production for reducing lead times is increasingly being recognized. This goal results from current challenges such as increasing variant diversity and customer requirements and also affects the organization of non-value-adding processes. For this reason, the use of mobile robot platforms as part of an autonomous transport system offers a wide range of potential for optimizing production. The joint research project FORobotics, funded by the Bavarian Research Foundation, is looking at mobile, ad-hoc cooperating robot teams. This means that autonomous transport vehicles are supplemented by an HRC-capable robot. The use of these robot platforms not only requires appropriate hardware integration, but also integration into the higher-level planning and control system. In addition, the interaction of mobile robots with humans is being investigated and a user evaluation is being carried out.

Production planning and control, taking into account the capabilities of all resources used, forms the basis for order processing in a digitally networked production environment. Sub-project 3 "Task-oriented planning" of FORobotics addresses the planning and control of intelligently networked and mobile production resources. The prerequisite for flexibility at the resource level is responsiveness at the planning and control level. The overarching goal is still to plan and control the production process in terms of quantity, deadlines and capacity.

First, the foundations were laid for the development of a planning system that optimizes order processing with the dynamics resulting from resource mobility. It became clear that in the area of production planning involving mobile resources, some fundamental adjustments to the concept were initially necessary. The planning and control process goes beyond current PPS systems. The considerations begin with the incoming order and the preparation of the order for automatic scheduling, which must be adapted to the "new" flexibility.

From order receipt to the creation of the production program, implementation takes place centrally in the planning software of industry partner Software4Production. The production program is then transferred to the control system, which is implemented in the IoT platform Thingworx. Industry partner Software Factory provides Thingworx and supports the implementation of a flexible production control process. Order release and monitoring for the resources is therefore carried out via a flexible IoT solution that can process feedback in the event of unexpected incidents or ad-hoc situations initiated by a plant employee. On the other hand, the current order status is visualized.

The handling of short-term disruptions, such as obstacles, is controlled decentrally at resource level. Depending on the type of decision to be made, however, the central control unit is also consulted, for example to take into account data from other resources and order information when avoiding an obstacle. Information and knowledge that can be derived from the situation-specific behavior of the autonomous resources is fed back into planning via order monitoring. The knowledge is taken into account for the next planning run in order to transmit a production program to the control system that is optimized compared to one without such a feedback loop. The regularity of communication between planning and control, i.e. updates in the event of changes and disruptions at store floor level, must be determined depending on throughput times and the planning horizon.

Another research focus is on the integration of global path planning for the mobile platform into production planning and control. To determine transportation times, it is necessary to carry out global path planning during the planning stage when checking the availability of resources for a job to be planned and when comparing requirements and capabilities. One way of dealing with the uncertainties that arise in a dynamic production environment is to plan static buffers. This approach is not preferable in the context of flexible production systems with autonomous, intelligent resources, as a defined variable is used here without knowledge of the production situation at the time of transportation. Based on the capabilities of the mobile resources used in the project and the supplemented hardware to detect the environment and identify objects, the transportation times are determined with possible deviation based on the environmental situation prevailing at the time of the transportation task. This allows the routes of other resources and dynamic obstacles to be taken into account at the planning level.

The modules developed at planning and control level will be evaluated on partial demonstrators and in simulations and will then be integrated into the overall demonstrator to enable an overall evaluation.
J. Schilp, M. Krä/as

Scientific Society for Assembly, Handling and Industrial Robotics (MHI e.V.) © MHI

Briefly explained: The MHI e.V.
The Wissenschaftliche Gesellschaft für Montage, Handhabung und Industrierobotik e.V. (MHI e.V.) is a network of renowned university professors - institute directors and chair holders - from German-speaking countries. The members conduct both fundamental and application-oriented research on a wide range of current topics in the fields of assembly, handling and industrial robotics. Further information on the society, its members and activities: www.wgmhi.de.

Chair of Production Informatics at the University of Augsburg © University of Augsburg

Briefly explained: Chair of Production Informatics
The Chair of Production Informatics at the University of Augsburg was founded in 2015 by Prof. Dr.-Ing. It deals with cyber-physical production systems for the optimization of industrial value and process chains, including in assembly and additive manufacturing. The research focus is on IT-based methods, concepts and solutions for end-to-end information and data processing and the networking of production resources. In addition to research and teaching, the Chair of Production Informatics also offers consulting and services as part of industrial contract research. www.informatik.uni-augsburg.de/lehrstuehle/pi/