About DARWIN
The DARWIN project, funded by SESAR 3 Joint Undertaking, will bring a solution that will help the aviation market maintain operational efficiency following the increasing emergence of drones and air taxis. The solution will increase safety and efficiency of air travel by introducing human-AI teaming.
The 3 core enabling technologies
The DARWIN project builds on the available technology base in AI and leverages the partners’ excellent position in the aviation supply chain. In doing so, DARWIN addresses the need for scalable, interconnected and highly automated eMCO and SPO operational concepts as one of the inherent fundamental building blocks of the Digital European Sky (SESAR ATM Master Plan Phase D).
Trustworthy Machine Reasoning
The platform will provide capabilities for rule-based, transparent and explainable decision support and/or decision-making.
Human-AI Collaboration Layer
This layer will be implemented on top of the Reasoning Platform. It will provide collaborative capabilities for pilot interaction with adaptive automation and assistants to efficiently keep the human in the loop of the workflow in the eMCO or SPO cockpit with Level 2 automation.
Pilot State and Task Load Monitor
This monitor will provide data on the state of the pilot (e.g. awake, drowsy, incapacitated) and the current and predicted task load to the collaboration layer and to the automation so that it can respond adaptively.
Challenges addressed by DARWIN
Adaptive automation for optimal human-AI collaboration: How to allocate tasks between the pilot and a range of digital assistants in a way that reduces the pilot’s workload, while keeping him in the loop and in charge of the aircraft’s situation.
Certifiable implementation of a Trustworthy Machine Reasoning Platform: How to ensure that artificial intelligence can become a readily available tool for use in aviation.
Integrating AI into regulations and (SESAR ATM Roadmap) operations: Ensuring that we know how to certify, regulate and verify AI (with a focus on machine reasoning) in aviation.
Detection and mitigation of pilot incapacitation: Developing an AI pilot assistant capable of detecting pilot incapacitation and taking control from the pilot.
Partners
