The award-winning thesis is entitled "Development, implementation and comparison of force sensor systems for the gripping unit of an order picking robot". In the master's thesis, a force sensor system is selected for implementation in a gripping unit in order to support and monitor the gripping process of an order picking robot in a meaningful way in order to be able to make statements about the gripping forces, successful (or unsuccessful) gripping, slipping, loss of the article and/or successful depositing of the article. The gripper system used, to which the sensors must be adapted, is a modular system called "Soft-Gripper" developed at the Helmut Schmidt University.

The gripper has a modular design and can be equipped with a different number and size of elastic silicone fingers and a vacuum suction cup. The major challenge of the work was to integrate the sensors into the gripper unit in a sensible way without restricting its function. The advantage of the "soft gripper" is the considerably lower cost, which compensates for the use of expensive sensor technology.

In a previous project, two sensor systems were selected for implementation on the basis of a methodology for variant evaluation and comparison, consisting of a self-built velostat sensor and a modified commercial foil sensor. These sensor systems were adapted to the geometry of the gripping unit. After checking the functionality of the developed or modified sensors and their

calibration, the two sensor systems were tested in gripping tests of a two-finger gripper with ten different gripping objects. The measurement results were recorded using an Arduino controller and a self-programmed visualization and evaluation tool (in Matlab). Depending on the settings, the tool displays the force or weight values of the respective sensor system over time. An activity control of the individual sensors allows the detection of gripping, sliding and depositing of an object.

The evaluation of the tests showed that the gripping force could be measured for all objects. For certain objects, it was even possible to identify the individual picking phases based on the measured gripping forces. Furthermore, it was determined that the self-built Velostat sensor is preferable to the commercial sensor in this area of application due to the more accurate measurement results, greater flexibility and lower costs.

The work is characterized by a very structured approach to the task. The components required for the sensor solutions are presented and placed in a scientific context before the preliminary tests and the subsequent gripping tests are explained, which led to the aforementioned result.

After a thorough review, it was determined that no commercially available sensor could meet these requirements without modifications. With the development of the Velostat sensor, we have succeeded in finding a sensor solution that meets these criteria. Thanks to the flexibility and low cost of this solution, scalability and adaptation to different geometries is very quick and easy. The data obtained from the calibrations can easily be entered into the programmed tool in order to use sensors of new dimensions. This has resulted in a sensor solution that can be used universally for gripping systems.

Mr. Nasser studied at the Bundeswehr University of Applied Sciences in Hamburg and currently holds the rank of first lieutenant in the army with a focus on maintenance. The thesis was supervised by Prof. Dr.-Ing. Rainer Bruns, Chair of Machine Elements and Technical Logistics, Faculty of Mechanical Engineering, Helmut Schmidt University, Hamburg.