Components in the material handling segment are increasingly turning into modular units. International competitive pressure continues to mean that new requirements, such as better integration of the operator while simultaneously increasing functional safety, are decisive factors for the acceptance of solutions on the market.

The first part of the article in issue 10/22 on p. 34 looked at how control systems and IIoT platforms should be designed in the future in order to meet the aforementioned requirements. The next important point is indoor localization.

Indoor localization

The localization of machines and people is a basic prerequisite for being able to provide location-based data via an application programming interface (API). It is therefore important to integrate a robust and reliable localization system to support functions that require a location reference. One pillar of the localization implemented in the Comnovo project is positioning via ultra-wideband (UWB), a radio-based technology that uses a broad frequency spectrum of at least 500 MHz. Compared to other radio concepts, this makes UWB robust against interference. Interference with other narrowband solutions is also very low, allowing different systems to coexist. In combination with a very precise clock on the UWB chips, it is possible to determine the distance between two devices
To determine a position from the distances measured by the UWB system, a Kalman filter based on the principle of trilateration is used. This requires a distance measurement to three fixed anchors with a known position in order to determine a unique position in a two-dimensional space or to four anchors in a three-dimensional space (Fig. 1).

In addition to UWB, an Inertial Measurement Unit (IMU) is also used to localize people. An IMU measures the accelerations and rotational speeds along each of the three spatial axes. Both measurements are used to first align the local IMU reference frame to the real world on the vertical axis via gravity and track its rotation. Then the measured (and rotated) acceleration is integrated twice to calculate the person's change in position. The permanent integration of noisy sensor values results in a steadily increasing error that affects the accuracy of the localization over time. This effect is mitigated by zero velocity updates (ZUPT) with foot-mounted sensors. This method uses the time periods in which a foot touches the ground to analyze and correct the accumulated error, as the sensor is at rest and thus the position and orientation remain unchanged for a moment.

UWB localization is primarily used by industrial machines, cranes, floor transport systems and forklifts responsible for material flow. As the cranes not only move in the XY plane, but also vertically in the Z axis, each crane has two UWB tags, one on each trolley to determine the X and Y coordinates, and another tag on the load hook so that the corresponding tag on the trolley can determine the Z coordinate. The tags on the trolleys are connected directly to the DCP controller via a CAN interface. The particular challenge with people is that it is not possible to attach a UWB tag to every position on the body. This results in shadowing of the antenna, which leads to a drop in the accuracy of UWB localization. For safety reasons, a combination of UWB localization with an IMU system was therefore used from the outset for people such as a crane operator. As a result, the University of Rostock, Thorsis and Comnovo developed a "wearable" that allows various sensors to be attached to the body (Fig. 2). The location data is published via OPC UA, using a "lightweight" version of the IIoT platform.

3D planning, simulation and visualization

The aim of the 3D tools is to plan the geometric layout of the plant or production or assembly processes, visualize the assistance functions developed in Optimum and then simulate them in a virtual environment. The factory layouts created in the planning tool can be transferred directly to the simulation application. The 3D geometry models created, such as those from Tarakos, are reused in simulation modules. Figure 3 shows the definition of areas that cranes and other machines are not allowed to pass over.

For communication with control systems, the simulation tool offers a plug-in interface that can be used to implement a bidirectional connection to the DCP distributed control system. All other simulation components are also connected to the distributed control system via this plug-in interface. The operator can move the virtual crane using a smartphone - just like the crane in the real demo assembly process. There are modules in the 3D simulation to simulate the behavior of the cranes, operators and vehicles. The position data of the individual objects resulting from the simulated behavior is transferred to the DCP control system.

If no control applications are required for either the operator or the manually operated forklift truck, the position data generated in the simulation model is sent to the IIoT platform. In the real demonstrator, the IIoT component obtains the position data directly from the position tag via UWB and makes it available on the OPC UA information model of the IIoT platform. The IIoT platform behaves in the simulation in the same way as in the real demonstrator, as it has no direct interface to the physical components. In the simulation, the distributed control system is adapted as closely as possible to that of the real demonstrator. The data flow from the HMI to the IIoT platform and to the DCP is exactly the same, as it only requires standard network connections with corresponding communication protocols.

The question of cybersecurity