DARWIN’s ambition is to develop AI-powered digital assistants and
human-AI collaboration frameworks to support both eMCO (extended Minimum Crew Operations) and SPO (Single Pilot Operations), ensuring the same or higher level of safety and the same or lower workload as today’s full-crew operations.
About DARWIN
The DARWIN project, funded by the SESAR 3 Joint Undertaking, will provide a solution to help the aviation market maintain operational efficiency as drones and air taxis become more prevalent. The solution will increase the safety and efficiency of air travel by introducing human-AI teaming.
Some of the upcoming
challenges for aerospace
UPCOMING PILOT SHORTAGE
REDUCING PILOT WORKLOAD
REDUCING CREW OPERATION
AUTONOMOUS VEHICLES
SINGLE PILOT OPERATION
The DARWIN project builds upon the available technology base in AI and leverages the partners’ excellent position in the aviation supply chain to address the need for scalable, interconnected, and highly automated eMCO and SPO operation concepts as one of the inherent foundational building blocks of the Digital European Sky (SESAR ATM Master Plan Phase D).
The main operational objective of DARWIN is to realise our ambition of Level 2 automation for eMCO and SPO. This objective is underpinned by three technical objectives and three supporting objectives.
Technical Objective 1:
Develop & Validate Trustworthy Machine Reasoning AI Platform (TMRP)
Provide a reusable, explainable, and safety-critical decision-aiding and decision-making platform that enables the implementation of principles and features for the highly regulated aviation market.
Technical Objective 2:
Develop & Validate a Pilot State and Task Load Monitor
Detect potentially hazardous human states – including incapacitation – and assess the performance and task load of pilots, for use in operational applications in aviation and other fields.
Technical Objective 3:
Develop and Implement Human-AI Collaboration Technology
Facilitate a transition from existing ConOps to an AI-powered Adaptive Automation and Collaboration concept. This will be done in two phases: a phase for agile concept formation, prototyping and early validation and a second phase for agile maturation. This enables rapid iterations to validate the emerging concepts for human-AI collaboration.
Supporting Objective 4:
Safety & Security Aspects
Perform safety and security analysis for level 2 automation for eMCO & SPO using state of the art safety and security risk assessment methodologies.
Supporting Objective 5:
Support Regulatory & Certification Process
Develop a certification and regulatory approach for trustworthy machine reasoning and human-AI collaboration, including the definition of risk mitigation strategies.
Supporting Objective 6:
Human-AI Collaboration Guidelines
Develop guidelines for human-AI collaboration that describe how a human pilot should interact with an AI.
MAIN OPERATIONAL OBJECTIVE
AUTOMATION FOR SINGLE PILOT OPERATIONS
TECHNICAL OBJECTIVE 2
HUMAN-AI COLLABORATION
TECHNICAL OBJECTIVE 3
PILOT STATE & TASK LOAD MONITOR
TECHNICAL OBJECTIVE 1
TRUSTWORTHY MACHINE REASONING AI PLATFOM
SUPPORTING OBJECTIVE 1
SAFETY & SECURITY ASPECTS
SUPPORTING OBJECTIVE 2
SUPPORT REGULATORY & CERTIFICATION PROCESS
SUPPORTING OBJECTIVE 3
HUMAN-AI COLLABORATION GUIDELINES
Impact
Adaptive task sharing between humans and digital assistants and a trusted and certifiable AI platform will transform aviation and ATM operations, supported by the Digital European Sky. It will enable the safe operation in the air transport system of the upcoming new generation of CS-23 (sub)regional zero-emission aircraft and urban air mobility concepts, and promote crew reduction to eventually single-pilot operations for larger and long-haul CS-25 aircraft.
This will provide Europe’s citizens with more clean mobility at an affordable cost and increased safety levels in a more congested and complex airspace environment, thereby mitigating the looming pilot shortage.