HAL (Le Centre pour la Communication Scientifique Directe), Apr 20, 2022
In transportation and aerospace, more automation and autonomy are continuously added to systems. ... more In transportation and aerospace, more automation and autonomy are continuously added to systems. The ability of human operators to effectively monitor and interact with these systems, poses significant challenges. This research focuses on critical decisions that largely rely on the system capabilities but need to be validated and made under the responsibility of the operator. In the context of unmanned combat air vehicles (UCAV), the experiment focuses on how the communication strategy of semi-autonomous systems modifies the operators' understanding of the situation and the final decision. The study has been conducted in an immersive simulator with a 30 minutes ecological military scenario where the operator had to manage a full mission, including an unplanned missile firing decision. The experiment included the use of physiological measures related to electrodermal and cardiac activities. The paper reports the results of the decision-making performances and the analyses of the physiological parameters. It appears that the communication strategy has an impact on the situation awareness of the operator, the decision taken, and the evolution of the physiological parameters.
HAL (Le Centre pour la Communication Scientifique Directe), Jun 27, 2022
Prendre en compte le processus décisionnel des utilisateurs humains pourrait permettre à des syst... more Prendre en compte le processus décisionnel des utilisateurs humains pourrait permettre à des systèmes d'assistance décisionnelle d'être plus adaptés. Cet article présente une étude qualitative où huit participants ont incarné un opérateur aérien chargé d'effectuer deux missions nécessitant une replanification de plan de vol de drone assisté par une IA. Des entretiens d'autoconfrontation ont permis d'établir un lien entre l'évolution des critères du compromis de plan et le processus décisionnel en cours, puis de développer un suivi automatique de ce processus. Nous en tirons des exemples de situations dans lesquelles le système d'IA peut adapter différemment ses recommandations, selon que l'opérateur est identifié comme impliqué dans une phase d'exploration ou bien une phase d'exploitation des plans de vol.
HAL (Le Centre pour la Communication Scientifique Directe), Jun 29, 2020
L'accroissement du niveau d'autonomie décisionnelle des systèmes de drones nécessite l'introducti... more L'accroissement du niveau d'autonomie décisionnelle des systèmes de drones nécessite l'introduction d'algorithmes complexes qui permettent d'envisager un champ opérationnel plus large, mais dont le fonctionnement devient plus opaque pour l'opérateur en charge de la supervision. Cette étude s'appuie sur les mécanismes cognitifs sous-jacents à la prise décision humaine pour définir des outils d'amélioration de la transparence des propositions du système. Une expérimentation en micro-monde est menée pour étudier l'influence de ces différents types d'assistance sur le jugement de l'opérateur dans le cadre d'une replanification de mission de drone.
HAL (Le Centre pour la Communication Scientifique Directe), Oct 6, 2021
Building human confidence in the decisions proposed by the machine and the inter-comprehension be... more Building human confidence in the decisions proposed by the machine and the inter-comprehension between operators and increasingly intelligent aeronautical systems is a performance and ethical issue. One of the ways to improve the user experience is to implement a more flexible distribution of roles through a collaborative organization. In this paper, we describe interaction principles that support such collaborative decision making, and we present an implementation through an interactive microworld used in experimental studies. This environment allows us to study the behavior of a drone operator in performing flight replanning tasks in collaboration with a simulated intelligent system using dialogue tools at the operational and tactical levels.
HAL (Le Centre pour la Communication Scientifique Directe), Mar 15, 2021
Even if Autonomous Intelligent Cyber-defense Agents (AICA) are dedicated to react to cyber events... more Even if Autonomous Intelligent Cyber-defense Agents (AICA) are dedicated to react to cyber events without human inputs, they will still have to interact with human operators in charge of the overall system security at least during training, planning or reporting phases. Such operators will need to construct a mental representation of the agent from whom they will be responsible whether to understand how they are going to protect the system, or to evaluate the state of this system and remaining risks after their reaction to a cyber event. In this paper, we investigate how individual characteristics of the operator may shape the acceptability and trust they have in the protection strategy deployed by such autonomous agents. Through a micro-world experimental study, we simulate an operator-intelligent agent cooperative decision-making task. We make the hypothesis that the operator's psychological profile could have an impact on their relationship. We found a link between the extraversion of the participants and their feeling of responsibility for the agent to supervise which demonstrates the need to consider characteristics of the operator when developing more sophisticated intelligent agents to have an efficient human-AICA teaming.
The ability to maintain an appropriate level of vigilance over long periods of time underlies suc... more The ability to maintain an appropriate level of vigilance over long periods of time underlies success on a range of tasks. Particularly, staying alert allows to detect infrequent signals and to allocate the right level of cognitive resources to respond to expected or unexpected events. A review of literature shows that some physiological markers can assess this variation of attention in a lot of lab studies. These findings are interesting for human factors in aeronautics as it appears as a way to characterize and quantify the observation during assessment with pilots in a cockpit simulator. The objective of this present study is to integrate a set of physiological metrics in a representative cockpit and to test their robustness in a more ecological environment with the associated constraints. This paper presents the first step of this project with a series of two lab experiments where we tested physiological markers of vigilance proposed in the literature.We first explored these mar...
CNIA 2021 : Conférence Nationale en Intelligence Artificielle, Jun 28, 2021
Avec l'introduction de l IA dans le pilotage des véhicules terrestres ou aériens, la répartition ... more Avec l'introduction de l IA dans le pilotage des véhicules terrestres ou aériens, la répartition des rôles entre opérateur et système devra évoluer de manière dynamique. A travers une expérimentation en micro-monde s r la s per ision d n drone intelligent, nous étudions comment une telle transition entre modalités d in erac ion coopératives et collaboratives peut affecter l'expérience et le choix de l opérateur. Nous observons des variables comme le sentiment de responsabilité ou la confiance et constatons que les opérateurs ont faiblemen conscience de l infl ence de l IA s r le r propre prise de décision.
W e live in an increasingly technological world. Automated systems certainly can make life easier... more W e live in an increasingly technological world. Automated systems certainly can make life easier, but they can also create complexity and uncertainty. Moreover, it is clear that automation does not merely supplant human activity, but also transforms the nature of human work. This review examines an original account of this transformation-a link between automation technology and the sense that our actions cause effects on the outside world (so-called 'agency'). Accordingly, we first discuss the human factor issues related to automation technology. Particularly, we introduce the out-of-the-loop performance problem. Then, we introduce recent findings about agency. We propose that several recently developed psychological approaches to the self-promise to enhance our comprehension of the transformation induced by increased automation. Next, we address the controversial issue of agency measuring, particularly the necessary dissociation between explicit and implicit agency measurement. In particular, we introduce the intentional binding effect as an implicit agency measurement, and we discuss the problems and issues related to the generalization of this effect to more complex situations. Finally, we suggest that the investigation of this authorship processing in the field of human-machine interaction may be fruitful, both to elaborate concrete design recommendations and to evaluate the potentiality for an HMI to satisfy the agency mechanism. Automation and human control in complex systems "The burning question of the near future will not be how much work a man can do safely, but how little." [85] There is perhaps no facet of modern society in which the influence of automation technology has not been felt. Whether at work or at home, while travelling or while engaged in leisurely pursuits, human beings are becoming increasingly accustomed to using and interacting with sophisticated computer systems designed to assist them in their activities. Even more radical changes are anticipated in the future, as computers increase in power, speed and "intelligence". We have usually focused on the perceived benefits of new automated or computerized devices. This is perhaps not surprising, given the sophistication and ingenuity of design of many such systems (e.g., the automatic landing of a jumbo jet, or the docking of two spacecraft). The economic benefits that automation can provide, or is perceived to offer, also tend to focus public attention on the technical capabilities of automation. However, our fascination with the possibilities afforded by technology often obscures the fact that new computerized and automated devices also create new burdens and complexities for the individuals and teams of practitioners responsible for operating, troubleshooting and managing high-consequence systems. Whatever the merits of any particular automation technology, it is clear that automation does not merely supplant human activity but also transforms the nature of human work. As a matter of fact, the role of the human actors may possibly evolve from direct control to supervision. Understanding the characteristics of this transformation is vital for successful design of new automated systems. Automation and OOL performance problem When new automation is introduced into a system, or when there is an increase in the autonomy of automated systems, developers often assume that adding "automation" is a simple substitution of a machine activity for human activity (substitution myth, see [92]). Empirical data on the relationship of people and technology suggest that this is not the case and that traditional automation has many negative performance and safety consequences associated with it stemming from the human out-of-the-loop (OOL) performance problem (see [22], [50]). Classically, the out-of-the-loop performance problem leaves operators of automated systems handicapped in their ability to take over manual operations in the event of automation failure [22]. The OOL performance problem has been attributed to a number of underlying factors, including human vigilance decrements (see [7], [86]), Issue 4-May 2012-Sense of Control in Supervision Tasks of Automated Systems
Adaptive automation appears to be one of the next most challenging milestones in the field of avi... more Adaptive automation appears to be one of the next most challenging milestones in the field of aviation and air traffic management. This approach is aimed to be human centered (Billings, 1996) and has to be user-friendly in order to increase safety (Flightpath 2050). The detection of human operators’ performance inevitably has to be the first step in this activity (Maiwald & Schulte, 2011). Here the edges of acceptable behaviour have to be identified well in time, which means that points where human performance deteriorates have to be found. This has led to the notion of a Human Performance Envelope (Graziani et al., 2016). Within this context a reliable on-line recording and analysis of operational data is needed. In a second step, these data will then set the trigger and recommend when a redistributions of tasks has to take place and a higher level of automation has to take over. While new identification means like behavioral markers (Edwards, 2013, 2014) are under investigation, p...
L’augmentation constante du nombre d’attaques par virus informatique a entraine la necessite pour... more L’augmentation constante du nombre d’attaques par virus informatique a entraine la necessite pour l’Armee de l’Air Francaise, de developper un outil nomme « Recognized Cyber Picture » (RCP) permettant une visualisation globale de l’etat cyber de ses systemes interconnectes. Dans un tel systeme sociotechnique, bien que les enjeux techniques soient colossaux, les problematiques humaines, notamment en termes de transfert d’information entre les differents operateurs, le sont tout autant. Cet article propose d’etudier une partie du volet social de la RCP et plus particulierement le role de la confiance interpersonnelle, entre deux operateurs humains qui ne se connaissent pas, dans la realisation d’une tâche collaborative mediee par ordinateur.
Dans cette these, nous nous interessons aux formalismes de representation et de manipulation de l... more Dans cette these, nous nous interessons aux formalismes de representation et de manipulation de la connaissance dans des sytemes necessitant une representation symbolique d'une realite observee. Nous examinons plus particulierement deux types de traitement de l'information : la fusion et la classification. Nous montrons le besoin d'un otuil algebrique pour structurer la connaissance. Puis nous nous penchons sur les structures de treillis et mettons en evidence leur adequation pour caracteriser un resultat de fusion. En nous fondant sur l'algebre des treillis et la logique des predicats nous construisons le modele des cubes. Celui-ci definit, dans une optique mathematique, une theorie homogene pour la manipulation des conjonctions de proprietes. Le processus d'enrichissement du modele est relance par l'introduction des contraintes. Apres un developpement theorique consequent la structure de treillis est retrouvee pour le modele des cubes contraints. Celui-ci p...
In the Horizon 2020 funded Future Sky Safety programme, the Human Performance Envelope project pu... more In the Horizon 2020 funded Future Sky Safety programme, the Human Performance Envelope project pushed airline pilots to the edges of their performance in real-time cockpit simulations, by increasing stress and workload, and decreasing situation awareness. The aim was to find out how such factors interact, and to detect the edges of human performance where some form of automation support should be employed to ensure safe continued flight. A battery of measures was used, from behavioural to physiological (e.g. heart rate, eye tracking and pupil dilation), to monitoring pilot performance in real time. Several measures – e.g. heart rate, heart rate variability, eye tracking, cognitive walkthrough, and Human Machine Interface (HMI) usability analysis – proved to be useful and relatively robust in detecting performance degradation, and determining where changes in information presentation are required to better support pilot performance in challenging situations. These results led to prop...
This paper reviews processes used by airlines to analyze inflight recorded parameters in their sa... more This paper reviews processes used by airlines to analyze inflight recorded parameters in their safety management system. Then it investigates how the expected crew activity can be modeled, based on airlines' standard operating procedures, and used to get a better understanding of operational risks. The language allowing the hierarchical description of the crew activity is described as well as the process dedicated to the identification of these elements from the recorded flight parameters. As an application case study, a go-around procedure is modeled and used to analyze a set of real aviation data. Results emphasize how this new flight data analysis process can complement existing tools.
In this paper we present a multidimensional approach to the concept of Trust in complex technical... more In this paper we present a multidimensional approach to the concept of Trust in complex technical environments. A survey allows to focus on the two main types: Interpersonal Trust (IT) and Trust in Automation (TA), an extended generic conceptual modeling based on a grounded theory methodology is proposed; finally the first phase of an experimental campaign is presented. The whole project aims at a better understanding of the role of the various components of Trust during the decision process of a human operator in cooperation with sophisticated systems and human partners.
HAL (Le Centre pour la Communication Scientifique Directe), Oct 25, 2021
Les premières attaques informatiques de grande envergure ont entrainé une prise de conscience sur... more Les premières attaques informatiques de grande envergure ont entrainé une prise de conscience sur la vulnérabilité des systèmes interconnectés. Afin de prévenir la présence d'un dysfonctionnement causé par une cyber attaque un système nommé « Recognized Cyber Picture » est envisagé par l'armée Française. Cependant, cette RCP sera potentiellement affectée par la perte de confiance des opérateurs dans la fiabilité des nouvelles technologies de l'information et de la communication. Il devient donc nécessaire de s'intéresser aux différents facteurs permettant la gestion d'une crise cyber, dont la communication directe entre les différents opérateurs. Cet article propose, en se basant sur des études de cas survenus dans l'aéronautique militaire et les résultats d'expérimentations, de regarder le champ relatif à la confiance interpersonnelle.
The aim of this paper is to describe an ongoing research summarized as follows: to investigate th... more The aim of this paper is to describe an ongoing research summarized as follows: to investigate the conjecture “Speaking implies Reasoning.” Indeed, in both professional situations and everyday life, any conversation requires hypothetico-deductive capabilities that are more or less ignored or relegated to a so-called context space. The purpose of the current study is then to formally describe these cognitive functionalities (ie, to design a theoretical model) as reasoning is a permanent invisible process involved in any interaction between human agents.
In the field of cyber-security, software performance optimization is a major focus of research to... more In the field of cyber-security, software performance optimization is a major focus of research to better prevent cyber threats. However, once threats are detected, they have to be managed by a human operator or more often by human operators' joint actions. The purpose of this study is to show that in these collaborative situations, the interpersonal trust level between these actors shapes their handling of the threat. Forty-five participants performed, with twenty-eight different fictive teammates, a collaborative counting task that included aleatory phases of jamming. Each fictive teammate was described through two adjectives selected to induce a predefined level of interpersonal trust (low or high). The subject and his collaborator worked on different systems with different objects to count and different jamming phases. Nevertheless, each participant had the possibility of supervising his teammate's work by checking out his task and modifying his answers (number of targets...
This paper presents the outlines of the STARE project that aims at developing the basis of a Fati... more This paper presents the outlines of the STARE project that aims at developing the basis of a Fatigue Risk Management System (FRMS) that will enable the French regional airlines to comply with the new European regulation on flight and duty time limitations. More specifically, the new regulation requires for an operator to implement a FRMS in the case of reduced rests. The FRMS is a data-driven system integrated in the future Safety Management System (SMS) of the airline. The whole process of the FRMS is described, including the use of predictive models of fatigue, a systematic and focused monitoring. The principles of the FRMS are discussed with regards to resilience engineering.
The aim of this article is to describe a Contextual Rule Induction Model based on Formal Concept ... more The aim of this article is to describe a Contextual Rule Induction Model based on Formal Concept Analysis. This model is the third layer of two algebraic structures previously designed. First layer: a lattice structure - called the Cube model - has been defined so as to capture the extension of the natural set operators to a lattice on conjunctions of first order literals. Second layer: an extension (Cubical-Formal Concept Analysis) to first order contexts (i.e. a mapping from a set of objects onto first order properties) has been built on the Cube model In the paper, the rule status is first defined and algorithms are proposed so as to produce checked symbolic rules. Second, we show how to gradate the rule space using again FCA in a bootstrap step. This rule clustering allows to evaluate the scope of each rule among the set of objects. Indeed, one can have a global view of the syntactic causal relations, the stratification of this set and a graphical representation (via the rule co...
ONERA and IMASSA are currently involved in a research program which aims at defining a methodolog... more ONERA and IMASSA are currently involved in a research program which aims at defining a methodology for the analysis of military aircrews' activity during everyday operations. The proposed methodology is different from the flight analysis processes currently used by airlines, for instance, as it is clearly focused on the human activity : it is based on a model of generic activity, built from aircrews' interviews and flight simulations. This model is used as a reference to identify the possible differences between the generic activity and the actions actually performed during the flight. Those differences are then interpreted using a typology of typical safety relevant events, established through an extensive review of incidents cases, together with the advanced functionalities of constrained logic programming. This paper presents the rationale and the objectives for this research program. It describes the various models of aircrew activity and the methodology used to build th...
HAL (Le Centre pour la Communication Scientifique Directe), Apr 20, 2022
In transportation and aerospace, more automation and autonomy are continuously added to systems. ... more In transportation and aerospace, more automation and autonomy are continuously added to systems. The ability of human operators to effectively monitor and interact with these systems, poses significant challenges. This research focuses on critical decisions that largely rely on the system capabilities but need to be validated and made under the responsibility of the operator. In the context of unmanned combat air vehicles (UCAV), the experiment focuses on how the communication strategy of semi-autonomous systems modifies the operators' understanding of the situation and the final decision. The study has been conducted in an immersive simulator with a 30 minutes ecological military scenario where the operator had to manage a full mission, including an unplanned missile firing decision. The experiment included the use of physiological measures related to electrodermal and cardiac activities. The paper reports the results of the decision-making performances and the analyses of the physiological parameters. It appears that the communication strategy has an impact on the situation awareness of the operator, the decision taken, and the evolution of the physiological parameters.
HAL (Le Centre pour la Communication Scientifique Directe), Jun 27, 2022
Prendre en compte le processus décisionnel des utilisateurs humains pourrait permettre à des syst... more Prendre en compte le processus décisionnel des utilisateurs humains pourrait permettre à des systèmes d'assistance décisionnelle d'être plus adaptés. Cet article présente une étude qualitative où huit participants ont incarné un opérateur aérien chargé d'effectuer deux missions nécessitant une replanification de plan de vol de drone assisté par une IA. Des entretiens d'autoconfrontation ont permis d'établir un lien entre l'évolution des critères du compromis de plan et le processus décisionnel en cours, puis de développer un suivi automatique de ce processus. Nous en tirons des exemples de situations dans lesquelles le système d'IA peut adapter différemment ses recommandations, selon que l'opérateur est identifié comme impliqué dans une phase d'exploration ou bien une phase d'exploitation des plans de vol.
HAL (Le Centre pour la Communication Scientifique Directe), Jun 29, 2020
L'accroissement du niveau d'autonomie décisionnelle des systèmes de drones nécessite l'introducti... more L'accroissement du niveau d'autonomie décisionnelle des systèmes de drones nécessite l'introduction d'algorithmes complexes qui permettent d'envisager un champ opérationnel plus large, mais dont le fonctionnement devient plus opaque pour l'opérateur en charge de la supervision. Cette étude s'appuie sur les mécanismes cognitifs sous-jacents à la prise décision humaine pour définir des outils d'amélioration de la transparence des propositions du système. Une expérimentation en micro-monde est menée pour étudier l'influence de ces différents types d'assistance sur le jugement de l'opérateur dans le cadre d'une replanification de mission de drone.
HAL (Le Centre pour la Communication Scientifique Directe), Oct 6, 2021
Building human confidence in the decisions proposed by the machine and the inter-comprehension be... more Building human confidence in the decisions proposed by the machine and the inter-comprehension between operators and increasingly intelligent aeronautical systems is a performance and ethical issue. One of the ways to improve the user experience is to implement a more flexible distribution of roles through a collaborative organization. In this paper, we describe interaction principles that support such collaborative decision making, and we present an implementation through an interactive microworld used in experimental studies. This environment allows us to study the behavior of a drone operator in performing flight replanning tasks in collaboration with a simulated intelligent system using dialogue tools at the operational and tactical levels.
HAL (Le Centre pour la Communication Scientifique Directe), Mar 15, 2021
Even if Autonomous Intelligent Cyber-defense Agents (AICA) are dedicated to react to cyber events... more Even if Autonomous Intelligent Cyber-defense Agents (AICA) are dedicated to react to cyber events without human inputs, they will still have to interact with human operators in charge of the overall system security at least during training, planning or reporting phases. Such operators will need to construct a mental representation of the agent from whom they will be responsible whether to understand how they are going to protect the system, or to evaluate the state of this system and remaining risks after their reaction to a cyber event. In this paper, we investigate how individual characteristics of the operator may shape the acceptability and trust they have in the protection strategy deployed by such autonomous agents. Through a micro-world experimental study, we simulate an operator-intelligent agent cooperative decision-making task. We make the hypothesis that the operator's psychological profile could have an impact on their relationship. We found a link between the extraversion of the participants and their feeling of responsibility for the agent to supervise which demonstrates the need to consider characteristics of the operator when developing more sophisticated intelligent agents to have an efficient human-AICA teaming.
The ability to maintain an appropriate level of vigilance over long periods of time underlies suc... more The ability to maintain an appropriate level of vigilance over long periods of time underlies success on a range of tasks. Particularly, staying alert allows to detect infrequent signals and to allocate the right level of cognitive resources to respond to expected or unexpected events. A review of literature shows that some physiological markers can assess this variation of attention in a lot of lab studies. These findings are interesting for human factors in aeronautics as it appears as a way to characterize and quantify the observation during assessment with pilots in a cockpit simulator. The objective of this present study is to integrate a set of physiological metrics in a representative cockpit and to test their robustness in a more ecological environment with the associated constraints. This paper presents the first step of this project with a series of two lab experiments where we tested physiological markers of vigilance proposed in the literature.We first explored these mar...
CNIA 2021 : Conférence Nationale en Intelligence Artificielle, Jun 28, 2021
Avec l'introduction de l IA dans le pilotage des véhicules terrestres ou aériens, la répartition ... more Avec l'introduction de l IA dans le pilotage des véhicules terrestres ou aériens, la répartition des rôles entre opérateur et système devra évoluer de manière dynamique. A travers une expérimentation en micro-monde s r la s per ision d n drone intelligent, nous étudions comment une telle transition entre modalités d in erac ion coopératives et collaboratives peut affecter l'expérience et le choix de l opérateur. Nous observons des variables comme le sentiment de responsabilité ou la confiance et constatons que les opérateurs ont faiblemen conscience de l infl ence de l IA s r le r propre prise de décision.
W e live in an increasingly technological world. Automated systems certainly can make life easier... more W e live in an increasingly technological world. Automated systems certainly can make life easier, but they can also create complexity and uncertainty. Moreover, it is clear that automation does not merely supplant human activity, but also transforms the nature of human work. This review examines an original account of this transformation-a link between automation technology and the sense that our actions cause effects on the outside world (so-called 'agency'). Accordingly, we first discuss the human factor issues related to automation technology. Particularly, we introduce the out-of-the-loop performance problem. Then, we introduce recent findings about agency. We propose that several recently developed psychological approaches to the self-promise to enhance our comprehension of the transformation induced by increased automation. Next, we address the controversial issue of agency measuring, particularly the necessary dissociation between explicit and implicit agency measurement. In particular, we introduce the intentional binding effect as an implicit agency measurement, and we discuss the problems and issues related to the generalization of this effect to more complex situations. Finally, we suggest that the investigation of this authorship processing in the field of human-machine interaction may be fruitful, both to elaborate concrete design recommendations and to evaluate the potentiality for an HMI to satisfy the agency mechanism. Automation and human control in complex systems "The burning question of the near future will not be how much work a man can do safely, but how little." [85] There is perhaps no facet of modern society in which the influence of automation technology has not been felt. Whether at work or at home, while travelling or while engaged in leisurely pursuits, human beings are becoming increasingly accustomed to using and interacting with sophisticated computer systems designed to assist them in their activities. Even more radical changes are anticipated in the future, as computers increase in power, speed and "intelligence". We have usually focused on the perceived benefits of new automated or computerized devices. This is perhaps not surprising, given the sophistication and ingenuity of design of many such systems (e.g., the automatic landing of a jumbo jet, or the docking of two spacecraft). The economic benefits that automation can provide, or is perceived to offer, also tend to focus public attention on the technical capabilities of automation. However, our fascination with the possibilities afforded by technology often obscures the fact that new computerized and automated devices also create new burdens and complexities for the individuals and teams of practitioners responsible for operating, troubleshooting and managing high-consequence systems. Whatever the merits of any particular automation technology, it is clear that automation does not merely supplant human activity but also transforms the nature of human work. As a matter of fact, the role of the human actors may possibly evolve from direct control to supervision. Understanding the characteristics of this transformation is vital for successful design of new automated systems. Automation and OOL performance problem When new automation is introduced into a system, or when there is an increase in the autonomy of automated systems, developers often assume that adding "automation" is a simple substitution of a machine activity for human activity (substitution myth, see [92]). Empirical data on the relationship of people and technology suggest that this is not the case and that traditional automation has many negative performance and safety consequences associated with it stemming from the human out-of-the-loop (OOL) performance problem (see [22], [50]). Classically, the out-of-the-loop performance problem leaves operators of automated systems handicapped in their ability to take over manual operations in the event of automation failure [22]. The OOL performance problem has been attributed to a number of underlying factors, including human vigilance decrements (see [7], [86]), Issue 4-May 2012-Sense of Control in Supervision Tasks of Automated Systems
Adaptive automation appears to be one of the next most challenging milestones in the field of avi... more Adaptive automation appears to be one of the next most challenging milestones in the field of aviation and air traffic management. This approach is aimed to be human centered (Billings, 1996) and has to be user-friendly in order to increase safety (Flightpath 2050). The detection of human operators’ performance inevitably has to be the first step in this activity (Maiwald & Schulte, 2011). Here the edges of acceptable behaviour have to be identified well in time, which means that points where human performance deteriorates have to be found. This has led to the notion of a Human Performance Envelope (Graziani et al., 2016). Within this context a reliable on-line recording and analysis of operational data is needed. In a second step, these data will then set the trigger and recommend when a redistributions of tasks has to take place and a higher level of automation has to take over. While new identification means like behavioral markers (Edwards, 2013, 2014) are under investigation, p...
L’augmentation constante du nombre d’attaques par virus informatique a entraine la necessite pour... more L’augmentation constante du nombre d’attaques par virus informatique a entraine la necessite pour l’Armee de l’Air Francaise, de developper un outil nomme « Recognized Cyber Picture » (RCP) permettant une visualisation globale de l’etat cyber de ses systemes interconnectes. Dans un tel systeme sociotechnique, bien que les enjeux techniques soient colossaux, les problematiques humaines, notamment en termes de transfert d’information entre les differents operateurs, le sont tout autant. Cet article propose d’etudier une partie du volet social de la RCP et plus particulierement le role de la confiance interpersonnelle, entre deux operateurs humains qui ne se connaissent pas, dans la realisation d’une tâche collaborative mediee par ordinateur.
Dans cette these, nous nous interessons aux formalismes de representation et de manipulation de l... more Dans cette these, nous nous interessons aux formalismes de representation et de manipulation de la connaissance dans des sytemes necessitant une representation symbolique d'une realite observee. Nous examinons plus particulierement deux types de traitement de l'information : la fusion et la classification. Nous montrons le besoin d'un otuil algebrique pour structurer la connaissance. Puis nous nous penchons sur les structures de treillis et mettons en evidence leur adequation pour caracteriser un resultat de fusion. En nous fondant sur l'algebre des treillis et la logique des predicats nous construisons le modele des cubes. Celui-ci definit, dans une optique mathematique, une theorie homogene pour la manipulation des conjonctions de proprietes. Le processus d'enrichissement du modele est relance par l'introduction des contraintes. Apres un developpement theorique consequent la structure de treillis est retrouvee pour le modele des cubes contraints. Celui-ci p...
In the Horizon 2020 funded Future Sky Safety programme, the Human Performance Envelope project pu... more In the Horizon 2020 funded Future Sky Safety programme, the Human Performance Envelope project pushed airline pilots to the edges of their performance in real-time cockpit simulations, by increasing stress and workload, and decreasing situation awareness. The aim was to find out how such factors interact, and to detect the edges of human performance where some form of automation support should be employed to ensure safe continued flight. A battery of measures was used, from behavioural to physiological (e.g. heart rate, eye tracking and pupil dilation), to monitoring pilot performance in real time. Several measures – e.g. heart rate, heart rate variability, eye tracking, cognitive walkthrough, and Human Machine Interface (HMI) usability analysis – proved to be useful and relatively robust in detecting performance degradation, and determining where changes in information presentation are required to better support pilot performance in challenging situations. These results led to prop...
This paper reviews processes used by airlines to analyze inflight recorded parameters in their sa... more This paper reviews processes used by airlines to analyze inflight recorded parameters in their safety management system. Then it investigates how the expected crew activity can be modeled, based on airlines' standard operating procedures, and used to get a better understanding of operational risks. The language allowing the hierarchical description of the crew activity is described as well as the process dedicated to the identification of these elements from the recorded flight parameters. As an application case study, a go-around procedure is modeled and used to analyze a set of real aviation data. Results emphasize how this new flight data analysis process can complement existing tools.
In this paper we present a multidimensional approach to the concept of Trust in complex technical... more In this paper we present a multidimensional approach to the concept of Trust in complex technical environments. A survey allows to focus on the two main types: Interpersonal Trust (IT) and Trust in Automation (TA), an extended generic conceptual modeling based on a grounded theory methodology is proposed; finally the first phase of an experimental campaign is presented. The whole project aims at a better understanding of the role of the various components of Trust during the decision process of a human operator in cooperation with sophisticated systems and human partners.
HAL (Le Centre pour la Communication Scientifique Directe), Oct 25, 2021
Les premières attaques informatiques de grande envergure ont entrainé une prise de conscience sur... more Les premières attaques informatiques de grande envergure ont entrainé une prise de conscience sur la vulnérabilité des systèmes interconnectés. Afin de prévenir la présence d'un dysfonctionnement causé par une cyber attaque un système nommé « Recognized Cyber Picture » est envisagé par l'armée Française. Cependant, cette RCP sera potentiellement affectée par la perte de confiance des opérateurs dans la fiabilité des nouvelles technologies de l'information et de la communication. Il devient donc nécessaire de s'intéresser aux différents facteurs permettant la gestion d'une crise cyber, dont la communication directe entre les différents opérateurs. Cet article propose, en se basant sur des études de cas survenus dans l'aéronautique militaire et les résultats d'expérimentations, de regarder le champ relatif à la confiance interpersonnelle.
The aim of this paper is to describe an ongoing research summarized as follows: to investigate th... more The aim of this paper is to describe an ongoing research summarized as follows: to investigate the conjecture “Speaking implies Reasoning.” Indeed, in both professional situations and everyday life, any conversation requires hypothetico-deductive capabilities that are more or less ignored or relegated to a so-called context space. The purpose of the current study is then to formally describe these cognitive functionalities (ie, to design a theoretical model) as reasoning is a permanent invisible process involved in any interaction between human agents.
In the field of cyber-security, software performance optimization is a major focus of research to... more In the field of cyber-security, software performance optimization is a major focus of research to better prevent cyber threats. However, once threats are detected, they have to be managed by a human operator or more often by human operators' joint actions. The purpose of this study is to show that in these collaborative situations, the interpersonal trust level between these actors shapes their handling of the threat. Forty-five participants performed, with twenty-eight different fictive teammates, a collaborative counting task that included aleatory phases of jamming. Each fictive teammate was described through two adjectives selected to induce a predefined level of interpersonal trust (low or high). The subject and his collaborator worked on different systems with different objects to count and different jamming phases. Nevertheless, each participant had the possibility of supervising his teammate's work by checking out his task and modifying his answers (number of targets...
This paper presents the outlines of the STARE project that aims at developing the basis of a Fati... more This paper presents the outlines of the STARE project that aims at developing the basis of a Fatigue Risk Management System (FRMS) that will enable the French regional airlines to comply with the new European regulation on flight and duty time limitations. More specifically, the new regulation requires for an operator to implement a FRMS in the case of reduced rests. The FRMS is a data-driven system integrated in the future Safety Management System (SMS) of the airline. The whole process of the FRMS is described, including the use of predictive models of fatigue, a systematic and focused monitoring. The principles of the FRMS are discussed with regards to resilience engineering.
The aim of this article is to describe a Contextual Rule Induction Model based on Formal Concept ... more The aim of this article is to describe a Contextual Rule Induction Model based on Formal Concept Analysis. This model is the third layer of two algebraic structures previously designed. First layer: a lattice structure - called the Cube model - has been defined so as to capture the extension of the natural set operators to a lattice on conjunctions of first order literals. Second layer: an extension (Cubical-Formal Concept Analysis) to first order contexts (i.e. a mapping from a set of objects onto first order properties) has been built on the Cube model In the paper, the rule status is first defined and algorithms are proposed so as to produce checked symbolic rules. Second, we show how to gradate the rule space using again FCA in a bootstrap step. This rule clustering allows to evaluate the scope of each rule among the set of objects. Indeed, one can have a global view of the syntactic causal relations, the stratification of this set and a graphical representation (via the rule co...
ONERA and IMASSA are currently involved in a research program which aims at defining a methodolog... more ONERA and IMASSA are currently involved in a research program which aims at defining a methodology for the analysis of military aircrews' activity during everyday operations. The proposed methodology is different from the flight analysis processes currently used by airlines, for instance, as it is clearly focused on the human activity : it is based on a model of generic activity, built from aircrews' interviews and flight simulations. This model is used as a reference to identify the possible differences between the generic activity and the actions actually performed during the flight. Those differences are then interpreted using a typology of typical safety relevant events, established through an extensive review of incidents cases, together with the advanced functionalities of constrained logic programming. This paper presents the rationale and the objectives for this research program. It describes the various models of aircrew activity and the methodology used to build th...
Uploads
Papers by Nicolas Maille