Imagine replacing a physical airport control tower with a screen of cameras. To make sure controllers trust the AI that spots drones or runway mistakes, this system explains its reasoning in a way that matches the person's stress level. It's like having a digital co-pilot that doesn't just shout 'danger' but quickly shows why it's a problem so the human can act fast.
Remote digital towers lack the direct visual intuition of physical towers, making AI alerts potentially untrustworthy or overwhelming for controllers. This creates a safety gap when monitoring runways and taxiways for hazards.
A prototypical AI system for RDTs featuring self-explainable ML models and an adaptive HMI dashboard that changes explanation levels based on the user's cognitive state.
If you are an airport operator dealing with the transition to remote digital towers — this project developed a trustworthy AI system that monitors taxiways for bird hazards and drones. It provides explanations for alerts to ensure controllers act with confidence.
If you are an ATM provider dealing with runway misalignment risks — this project developed a monitoring tool that gives warnings and clear explanations. This reduces the risk of human error during approach and landing phases.
If you are an AI developer dealing with the 'black box' problem in critical safety systems — this project developed transparent ML models and adaptive interfaces. This allows for better human-AI teaming and higher user acceptability.
Based on available project data, no specific pricing or cost figures are provided.
The project includes technological objectives for delivery and deployment in close to real environments and specifically evaluates scalability potential for the ATM ecosystem.
Based on available project data, there is no specific information regarding IP or licensing terms.
The project focuses on 'trustworthiness', 'fairness', and 'accountability' to meet the high safety requirements of remote digital towers.
The project period runs from 2023-09-01 to 2026-02-28, with a prototypical system planned for deployment in close to real environments.
The consortium is small and academic-heavy, consisting of 4 partners from 3 countries (France, Italy, Sweden). With 3 universities and only 1 SME, the industry ratio is 25%, suggesting the project is primarily driven by research and validation rather than immediate commercial mass-production.
Contact Malardalen Universitet in Sweden
Contact us to explore licensing opportunities for trustworthy AI in aviation.
