Artificial intelligence in RoRo terminals for port evolution and sustainability in terms of efficient traffic management through automated AI training and AI decision-making
Initial situation / motivation
The planned AI-supported image data generation and evaluation, which has not been available in the port to date and is not technically feasible, is fundamentally crucial for resource-saving and sustainable traffic control in the port, as it extracts intelligent information from images – and does not just deliver ‘raw’ images like conventional cameras. The image data currently being captured cannot fully meet the AI-specific requirements for fast response times and solution presentation. Such an AI system architecture does not exist, but it is forward-looking.
The focus is on the integration and subsequent testing of innovative domain-specific AI solutions which, in combination with compatible digital infrastructure and innovative AI system architectures, enable high-resolution capture, processing and communication of image data from and for RoRo traffic. Based on this data, trained AI models generate automated instructions for traffic participants, which are communicated via large-scale LED multimedia displays in the form of dynamic content. This allows direct, situation-dependent interaction with terminal traffic in real time. The automation and increased effectiveness of traffic management in the RoRo terminal is tested and evaluated in real time.
The individual measures (1-3) of the project each go through the following three phases:
Planning, tendering/procurement and construction/installation. All three run in parallel or are interdependent.
- AI-supported RoRo traffic management using AI camera and sensor technology;
- High-performance AI system architecture for the automation of knowledge work;
- Intelligent traffic control based on AI-supported data processing.
