System Architecture for Digital Twin Based Collision Avoidance Through Private 5G Networks

In logistics environments, collision avoidance is critical to prevent damage to people, assets and infrastructure. Whether it is a manual or semi-automated fleet with human drivers, the risk of collision is always present. One approach to reduce this risk is to implement collision avoidance systems that intervene in the driving process, for example through warnings or speed reduction. To handle multiple vehicles, a wireless system must be used to collect location information. Collisions can then be handled based on the position data. For moving targets, wireless communication is required with high reliability, low latency and high data rates. Failure to meet these requirements limits the exchange of real-time information between the physical system and its virtualized twin, hindering decision-making. This paper proposes a system environment for virtual collision avoidance that uses a digital twin that is continuously updated through a local 5G network to collect device information. The digital twin is then used to estimate collision events with other obstacles. Both simulated and real-world scenarios in an industrial logistics context are used to evaluate the system's capabilities. The evaluation shows that the use of a local 5G network is advantageous for implementing virtual collision avoidance in intralogistics.