To build teams of humans and robots in intralogistics, a flexible and unobtrusive way for them to collaborate needs to be found. This requires on the one hand enough safety features and on the other hand a higher degree of intelligence at the robot side. Finally, special care has to be paid to the grippers of mobile manipulators since not only the arm itself but also the gripper and the handling of objects must be done in a safety-oriented way.
The superordinate coordination system in our project is the key element for flexible intralogistics. This system must be flexible enough to work even in environments where only single, customized products are produced (i.e., each produced item is unique).
A planning and scheduling engine will be developed which monitors the current order requests, production states and intralogistics situation. Based on this, a mission plan is generated which assigns intralogistics tasks to humans and robots based on task requirements. The generated plan is constantly updated and the proper execution is monitored. As an example, if a robot runs into a complex situation and needs remote support, it will be taken out of the resource pool for the time being. Another unit will be assigned to take over tasks to compensate.
The planning engine will also exploit the data collected during operation to optimize the planning. Data analysis techniques will be applied to extract patterns from task assignments to be able to preemptively send transportation units to areas where they might be needed next. In addition, the arrangement of the shop-floor can be optimized using this data.