The motivation for the E-PILOTS project is to identify cognitive tasks that pilots perform during different phases of the flight, the data and information elaborated during the decision-making process of the pilot together with the influence variables that drives their preferences in order to implement a cognitive computing e-pilot supporting tool for skill- based, rule-based and knowledge-based pilot tasks with seamless transitions between the pilot and the e-pilot roles.


The E-PILOTS challenge is very attractive considering the cognitive computing learning from past experiences and data. Results achieved in some areas shown an exceptional ability to learn and improve themselves, accelerating their use for some knowledge-based tasks that not long ago were seen as the exclusive domain of humans. However, a word of caution is necessary considering vexing questions about how cognitive computing can influence on pilot cockpit operations. Some authors argue that “for any given skill one can think of, some computer scientist may already be trying to develop an algorithm to do it”, but the implementation of AI solutions to replace some pilot tasks and responsibilities can be a source of disruptions and therefore it requires a particular methodology to ensure the right acceptability of new functionalities. This approach should consider a holistic analysis of cockpit operations and an innovative cognitive computing architecture to experiment with new solutions. AI technologies should mature enough to avoid frustration with cockpit automation research.


To address the identified challenges and enhance cognitive computing in cockpit operations in unexpected or very complex situations as a driver for the transformation of decision making, the E_PILOTS project will address the following key sub-objectives:

O1: To setup a modelling framework to synthesize the fascination with cognitive computing in some areas with its inertial proclivity towards automation and potential displacement of some pilot tasks considering rigorous analysis of cockpit operations and cognitive gaps in unexpected and complex.

O2: Identify functional and interface requirements of the simulation framework suitable to characterize and detail the scope of cockpit pilot supporting skill-based, rule-based and knowledge-based tasks considering overlapping responsibilities and roles in a dynamic context, which would help to overcome acceptability barriers.

O3: Development of a new pilot ‘natural-feeling’ profile by means of supporting cognitive functionalities transforming the cockpit operational context from a human-in-the-loop to a human-in-the-mesh. It considers a dual control in which pilot supervise e-pilot suggested actions and vice-versa as a measure to mitigate slip and lapses errors. In addition, agent-based models supporting socio-technological dynamics will be analysed to identify the right context dependent overlapping tasks between pilot and e-pilot.

O4: Trade-off context-aware mechanism between pilot workload and e-pilot overlapping tasks to minimize rule-based and knowledge-based pilot mistakes usually caused by an incorrect situation diagnosis. Cognitive computing service will be enhanced with communication interfaces to provide elaborated information to convince pilots about their.

O5: To design a service-oriented architecture to coordinate and enable efficient and effective collaboration between pilot and e-pilot to perform tasks in different operational conditions maintaining the performance variability under a performance envelope. Multicrew procedures will be enhanced with teamwork organization methods to exploit the benefits of pilot and e-pilot collaboration keeping cockpit operations performance under given.

O6: Implementation of a cognitive infrastructure to support the benefits of a symbiosis during the decision-making process when unexpected or very complex situations arise, in which tacit knowledge will be explicit shared between pilot and e-pilot including potential responsibility transferences considering reliable structures. Cognitive infrastructure implementation will be guided by what the pilot needs to know, when the information should be shared and how pilot tasks could be.

O7: Development of a roadmap from cognitive computing technological and cockpit operational points of The certifiability and predictability of the solutions will be at the core of the analysis driving into the selection of technologies considering the lessons learned during validation exercises to accomplish a given cockpit operational development level.