Semantic Swarm Exploration for Factory Environments

Automated mapping of industrial interiors to create a better foundation for factory planning and inventory management: That is the goal of the SemaFact research project.

Many manufacturing companies – especially small and medium-sized enterprises (SMEs) – face the challenge of increasing their productivity within limited space. To accomplish this, they need up-to-date information about spaces, equipment, and inventory.

The research project is developing a server-based software platform. It processes video data captured by lightweight indoor drones and uses it to create a three-dimensional, semantically enriched map of the environment. At the same time, the software determines the drones’ positions and controls their exploration. The drones themselves simply carry out the instructions generated by the server. This allows the hardware to remain simple, safe, and cost-effective.

The software is designed to have a modular structure and will initially support factory planning and inventory tracking. Later, applications for inspections, condition monitoring, and analysis can be built upon this foundation. In addition, new embedded systems modules for drone deployment are being developed and tested with indoor drones. This creates a scalable platform for intelligent, cooperative, and semantically supported spatial perception in industrial environments.