Interference-resistant harmonic ISM radar for real-time logistics and distribution

Typical intralogistics operations are characterized by dynamic environments and complex material flows. The processes in logistics are subject to constant change. Factors such as the warehouse situation, the positions and dimensions of the shelves and the movement of the logistics units influence these operations. The strong growth in the field of digitalization is making it increasingly easy for companies to use autonomous systems. The efficiency of modern material flow systems depends not only on the orderly structure of the warehouse racks, but also on the speed of the mobile robots used for sorting, picking or transporting. It is also essential that employees have the necessary knowledge to determine the position of the goods at all times.

In the logistics space, robots move together with other robots, human workers and other logistical entities. In warehouse environments where space is limited, this form of joint working leads to a considerable coordination effort. In addition, disorganized arrangements of rack contents and blind spots lead to longer picking times and higher storage costs. To ensure high productivity and efficiency in the warehouse, a highly accurate and precise localization system with a monitoring service for the logistics units in the form of a cyber-physical system (CPS) is essential.

The current status of localization technologies in warehouses varies depending on the framework conditions. In the case of automated guided vehicles (AGVs), guidelines using magnets or optical lines were previously used. Today, however, laser-based Simultaneous Localization And Mapping (SLAM), camera-based SLAM and RF-based techniques are used to provide greater flexibility and freedom in navigation. Laser-based SLAM typically uses high-cost LiDAR (Light Detection and Ranging) sensors to create detailed maps of an environment and simultaneously track the position of the object within that map. In dense and dynamic warehouse environments, this can lead to several challenges, such as highly varying reflective properties of objects. Camera-based or visual SLAM (VSLAM) utilizes visual input from one or more cameras and uses algorithms to detect and match features. In warehouse logistics, VSLAM can be used for the navigation and perception of automated guided vehicles (AGVs) as well as for the identification of specific items. However, its effectiveness is susceptible to fluctuations in lighting conditions and is dependent on environmental influences.

RF-based localization is currently carried out in industrial environments using ultra-wideband signals in the 3.6-10.6 GHz frequency band as well as Bluetooth or WLAN signals. In addition to propagation time measurements, signal levels, direction of reception and, depending on the technology, a time difference method, a mathematically closed solution or trained fingerprinting are used. The quality of the methods is linked to the accuracy of the inputs. A high accuracy leads to a high accuracy of the calculated position. The sub-meter accuracy achieved is still too imprecise in many cases. To achieve significantly higher accuracy, signal bandwidths are required that are not sufficiently available in the lower frequency bands.

The harmonic ISM radar at 61 and 122 GHz enables precise and interference-free measurement of echo signals with a high bandwidth. The recorded data cannot be used directly for the described application of localization without further preparation and processing. In this context, it is essential to define and fulfill not only the necessary software but also the hardware requirements and interfaces as well as the input and output parameters. A particular challenge in the new type of 3D position determination using distributed and independent radar systems lies in the physical implementation and signal processing. Added to this are integration into the industrial environment and robust implementation, even under complex conditions such as multi-path propagation or multi-transmitter scenarios. Interfaces such as web sockets and REST interfaces as well as MQTT are used to connect to external systems.

Project partners

• Heuel & Löher GmbH & Co. KG, Lennestadt (consortium lead)
• Ruhr-Universität Bochum, Lehrstuhl für Integrierte Systeme, Bochum
• Fraunhofer-Institut für Hochfrequenzphysik und Radartechnik FHR, Wachtberg
• Fraunhofer-Institut für Materialfluss und Logistik IML, Dortmund
• Technische Universität Dortmund, Lehrstuhl für Förder- und Lagerwesen, Dortmund
• SDFS Smarte Demonstrationsfabrik Siegen GmbH, Siegen
• Würth Industrie Service GmbH & Co. KG, Bad Mergentheim

Funding

This project is funded with a total 2.728.014 € by the European Union and the state of North Rhine-Westphalia as part of the EFRE/JTF program NRW 2021-2027.