Skip to main content
European Commission logo print header

Article Category


Article available in the following languages:


Better person-machine communication designed to help prevent accidents

The HoliDes project aims to develop human-machine communication and adaptation wherever they work together so that co-operation ensures a fluency in tasks and functionality exists between the two of them.

Human-machine communication is constant across various sectors (automotive sector, tourism, industry, etc.) and it has in fact been in the world of aeronautics that a number of European companies have joined forces to work in this scenario. Let us assume that incorrect encryption or decryption of messages between them and in a specific context could have disastrous consequences, and the fact is that some air accidents could have been avoided if conflicts of this type between the crew and control had been corrected. HoliDes, Holistic Human factors and System Design of Adaptive Cooperative Human-Machine Systems (AdCoS) has emerged on the assumption that pilots and automated cabin systems need to communicate with each other and act together in a co-operative way and in a highly adaptable manner. The achieving of this correct communication, co-operation and mutual adaptability would guarantee the understanding of messages, and therefore, the achieving of tasks in a fluent way while maintaining safety at all times. Globally, we can say that work is being done using a range of sensors and algorithms that measure the contextual state of the human operators so that once the data of the parameters set out have been processed and analysed, they can form part of a system or tool that assists in decision-making and thus has the “co-operative adaptation” capacity between the human and the machine and vice versa. To achieve all this, work is being done in various areas of technological expertise in which the following four are the most salient ones: specifying of patterns of human behaviour and design of the inference tool, unification and storage of information on experiments, the modelling ontology of human factors, and the drawing up of recommendations for correct human-machine communication. The first of these blocks, the design of the tool known as the Pilot Pattern Classifier, consists mainly of sensorising and measuring the relevant parameters in the conduct of a pilot on duty that can help to establish and personalise predictive models in real time for human behaviour (fatigue, distraction, etc.). This calls for a set of devices or sensors fitted to the pilot that monitor brain waves, eye tracking, heart rate, sweating and other biometric measurements; this work is being undertaken by the Universita Degli Studi Suor Orsola Benincasa. This information and other of great documentary value such as the recording of experiments run by Honeywell in its flight simulators, interviews with pilots, questionnaires, etc. must be stored in an accessible and structured way if it is to be truly operative. This information repository referring to the various experiments carried out is being developed by Tecnalia and is called EDA, Experiment Data Archive. We could say that it is a big data tool. At the same time, and hence another of the offshoots of this project, the vocabulary needs to be standardised so that all the systems that use this information can understand each other. In other words, an ontology that models human factors has to be generated so that the systems can be interoperable. Tecnalia is participating in this area as the leading company. The last of the areas in this project is the drawing up of recommendations for communication. Guides or user manuals provide recommendations so that the messages arrive optimally between the transmitter-receiver, in other words, between the human and the machine. So what is taken into account for this purpose is how the information is presented, or what amounts to the same thing, any type of variable that could affect the way a message is understood; so, the visual aspect, the interfaces (information architecture, the colours used, final look of the user interface, etc.), the acoustic aspect (type of sounds used, volume, etc.) or the tactile aspect (vibrations) are taken into consideration.