The use of drones or Unmanned Aerial Vehicles (UAVs) is changing the way firefighters can perform their duties, especially at larger industrial sites. Mediaan collaborated with PXL Smart ICT in developing a proof of concept called “Fire Drones”. This pilot case was in the context of the Rapid Drone Software Prototyping“- research conducted by PXL Digital and PXL Smart ICT.

Together with a team of six IT students, led by Nick Andries from Mediaan and Tim Dupont from PXL Smart ICT, we created a proof of concept which allows firefighters to quickly inspect and monitor dangerous fire areas at the industrial site. In testing the prototype, we collaborated with Brandweer Zuid-West-Limburg (fire department) to see how this works in a real-life situation.

The Proof of Concept

The idea behind the Fire Drones concept was to allow firefighters to quickly assess and monitor dangerous areas. Upon arriving at the industrial fire, firefighters can send out multiple drones to scan the whole area thus getting an immediate bird’s-eye view of the situation. Once send out, the drones are flying autonomously and will look for any hazard symbols, which then will be displayed on the dashboard with the appropriate actions to take. This solution results in more safety and limits the time it takes to scan the whole area.

YouTube

By loading the video, you agree to YouTube’s privacy policy.
Learn more

Load video

The Team

The six IT students involved in this project had backgrounds in software management, artificial intelligence, and application development. This whole concept was developed at the Mediaan office at the Corda Campus in Hasselt, and was supported by Project Manager Nick Andries and Computer Vision expert Raf Thewis from Mediaan, together with AI & Robotics researcher Tim Dupont from PXL Smart ICT:

“Providing students and Mediaan with the opportunity to collaborate with our research project is a pure Win-Win construction. It’s the ideal ground to test our preliminary results from prior research, to round the rough edges of our implementations and to get feedback from the industry. Students get hands-on experience with state-of-the-art solutions and may conduct, albeit with some guidance, research on their own. The win for Mediaan is twofold: they are introduced to new technologies and get in touch with very competent students. The goal was to create a Rapid Drone Software Prototyping showcase using a modular smart UAV architecture. The final result is outstanding. A big step has been taken. I look forward to implementing the valuable feedback and working together with the students and Mediaan to finalize this showcase.” – Tim Dupont (AI & Robotics researcher at PXL Smart ICT and lecturer at PXL Digital)

“The project that PXL Smart-ICT came up with under the supervision of Mediaan was one that stood out for me personally. It was challenging because some of the technologies we had to use were very new to us. Fortunately, we were working from the Mediaan office in Hasselt, which allowed us to ask for help and feedback on topics like computer vision, the agile way of working and our presentation skills. Vic Segers (PXL AI-student) and I are very happy with the progress we made and are going to continue exploring the possibilities further in our internship at PXL Smart-ICT.”- Borcherd Van Brakell (AI-Student at PXL Digital)

Technology Stack

The use of artificial intelligence is the core of this project, especially applying Computer Vision in practice and showing its capabilities. Furthermore, SLAM (Simultaneous Localization& Mapping) is used, which is a technique to map an unknown area. On the base of this map,  the UAVs (Unmanned Aerial Vehicles), which in this case are the drones, know how to fly and evade objects as they are scanning the area for safety information.