Projects
RENDERLINE (2022-2023)
Renderline is a user-friendly web-based data visualization tool, known as DataVis, designed to create videos using aeronautical data while preserving data confidentiality. It serves the purpose of crafting communication materials suitable for various audiences, from experts to the general public. Originally initiated to address the challenge of visualizing extensive air traffic data at DSNA, the project gained momentum in 2021 thanks to Stéphane Chatty and Christophe Hurter's leadership. Christophe Hurter leveraged ENAC's expertise in DataVis, honed over 15 years of research and patent development, which had remained underutilized at DSNA. The tool primarily focuses on visualizing trajectory data for effective communication and decision-making.
NExt Vision (2023-2024)
The "Next Vision" project aims to validate the contribution of virtual reality technologies in improving baggage control. The objective is to inspect luggage using a virtual reality headset without opening them. This project is based on three pillars: enhancing airport security with tools for detecting known or unknown threats, equipping security agents with flexible and efficient tools, and developing new artificial intelligence algorithms to explain the tool's findings when luggage is deemed dangerous. In summary, the machine must be able to justify its diagnosis.
Artimation (2020-2022)
In Air Transportation Management the Decision Making Process is already associated with AI. The algorithms are meant to help ATCOs in daily tasks, but they still face acceptability issues. Today’s automation systems with AI/Machine Learning do not provide additional information on top of the Data Processing result to support its explanation, making them not transparent enough. The Decision Making Process is expected to become a “White Box”, giving understandable outcome through an understandable process.
CODA (2023-2025)
The CODA project aims to create a collaborative human-machine system by integrating prediction models for task allocation, neurophysiological assessment to monitor operator mental states, and adaptable AI systems for safety. This system dynamically adjusts task allocation based on real-time cognitive assessments and traffic data, ensuring optimal performance. For example, if an ATCO's workload increases, the system can boost automation or employ AI tools to prevent performance issues and enhance safety.
TRUSTY (2023-2025)
TRUSTY leverages artificial intelligence (AI) to enhance efficiency and safety in the global deployment of Remote Digital Towers (RDT). Its primary aim is to increase transparency and trustworthiness of AI-driven decisions in the RDT context. TRUSTY focuses on tasks like taxiway monitoring (e.g., detecting bird hazards, drones, human intrusion) and runway monitoring, providing explanations for misalignments during approach and landing. The project incorporates self-explanation, transparent machine learning models, interactive data visualization, adaptive explanations, human-centric AI, and human-AI collaboration to ensure trustworthiness. It relies on state-of-the-art developments in data visualization and user-centric explanations, aiming to bridge trustworthy AI, multi-model machine learning, active learning, and user experience in human-AI interactions.