Flight deck human-automation issue detection via intent inference

Lee, Sang-Jin; Hwang, Inseok; Leiden, Kenneth

doi:10.1145/2669592.2669690

Cited by 7 publications

(8 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…16 For example, issues in flight deck human-automation interactions may be detected using anomaly detection. 17,18 Our prior work involved finding precursors to go-arounds, 4 highenergy approach and landing issues 19 and stall hazard 6 using the ADOPT algorithm discussed in this paper. This paper is an extension of the previous works that includes several scenarios of safety precursors under different problem setups.…”

Section: Literature Reviewmentioning

confidence: 99%

Using ADOPT Algorithm and Operational Data to Discover Precursors to Aviation Adverse Events

Janakiraman

Matthews

Oza

2018

2018 AIAA Information Systems-Aiaa Infotech @ Aerospace

View full text Add to dashboard Cite

In this paper, we describe a data mining method called ADOPT (Automatic Discovery of Precursors in Time series data) to identify precursors to aviation adverse events. An adverse event may refer to any unsafe event ranging from a negligible safety hazard to a catastrophic accident, depending on the scope of the analysis. A precursor is an early indicator of an increasing likelihood of the adverse event. Identifying precursors is important in the context of a proactive safety management because precursors detect the increasing severity of the underlying hazard much earlier, giving sufficient time to identify, analyze and implement corrective actions. ADOPT analyzes large volumes of historical data to find complex trends among several sensory variables simultaneously to find precursors. ADOPT's data mining approach captures real-world effects such as human factors, weather, geographic constraints, operating procedures, airline strategies etc that are difficult to capture using first-principle models. This paper describes the algorithm using two case studies including a takeoff stall hazard and RNAV adherence. While the case studies are not intended to discuss the critical safety risks in aviation, they are used to demonstrate the various steps involved in ADOPT including data preparation, variable selection, parameter tuning, experiment setup and analyzing the results. The results show that ADOPT can be a powerful tool to identify and analyze performance and safety issues in Aviation.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Using ADOPT Algorithm and Operational Data to Discover Precursors to Aviation Adverse Events

Janakiraman

Matthews

Oza

2018

2018 AIAA Information Systems-Aiaa Infotech @ Aerospace

View full text Add to dashboard Cite

show abstract

“…For example, in the dataset consisting of 20000 flights, used in Section 6.2.2, only 673 unique modes exist (0.3% of 262144). The number of hidden phases was set to n x = 5, after some preliminary experiments in which we tested several n x in range [3,20] and selected n x balancing a good prediction accuracy with low computational complexity of algorithm. We have selected flights with landings at the same destination airport with aircrafts of the same fleet and type, so that we eliminate potential differences related to aircraft dynamics or landing patterns.…”

Section: Real Flight Datamentioning

confidence: 99%

Semi-Markov Switching Vector Autoregressive Model-Based Anomaly Detection in Aviation Systems

Melnyk

Banerjee

Matthews

et al. 2016

Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

In this work we consider the problem of anomaly detection in heterogeneous, multivariate, variablelength time series datasets. Our focus is on the aviation safety domain, where data objects are flights and time series are sensor readings and pilot switches. In this context the goal is to detect anomalous flight segments, due to mechanical, environmental, or human factors in order to identifying operationally significant events and provide insights into the flight operations and highlight otherwise unavailable potential safety risks and precursors to accidents. For this purpose, we propose a framework which represents each flight using a semi-Markov switching vector autoregressive (SMS-VAR) model. Detection of anomalies is then based on measuring dissimilarities between the model's prediction and data observation. The framework is scalable, due to the inherent parallel nature of most computations, and can be used to perform online anomaly detection. Extensive experimental results on simulated and real datasets illustrate that the framework can detect various types of anomalies along with the key parameters involved.

show abstract

“…However, these approaches base their decisions of anomalies on data itself and not on the models that generate the data. 12 Taking a model-driven detection approach is computationally beneficial because detecting deviations of the actual data from the true model is much efficient.…”

Section: A Literature Surveymentioning

confidence: 99%

“…The aim of the proposed research is to implement a general framework as initially proposed by Lee et al, 12 for UI validation purposes using mode confusion detection (by validating on various mode confusion incidents/accidents) from the control-theoretic perspective. An intent-based framework for mode confusion detection is proposed because the mode confusion is the divergence in expectations or intents of the automation and the pilot about the actual behavior of the aircraft.…”

Section: B Proposed Approachmentioning

confidence: 99%

“…The architecture for UI validation using mode confusion detection is shown in Figure 2 inspired by our previous work. 12 Our approach develops modular models of automation (hybrid system) and pilot (discrete-event system), and models the UI as a filter of flight information, unlike existing works. 1,[7][8][9][10][11] Mode confusion could happen when the actual behavior of the automation does not match with the expected goal of the pilot.…”

Section: Mode Confusion Detection Frameworkmentioning

confidence: 99%

See 1 more Smart Citation

User Interface Validation using Mode Confusion Detection

Nandiganahalli

Lee

Hwang

et al. 2014

AIAA Modeling and Simulation Technologies Conference

Self Cite

View full text Add to dashboard Cite

Due to the rapid advancements in the digital avionics systems, the aircraft operation is getting more automated. However, this does not completely exclude the operations carried out by the pilot via interaction with the automation through the User Interface (UI) (or more commonly known as flight deck instruments). At times, this interaction raises situational awareness issues to the pilot about the behavior of the automation. One such critical issue is mode confusion where a mismatch occurs between the behavior of the automation expected by the pilot and the actual behavior of the automation. This paper examines the mode confusion detection problem from the UI design perspective by taking a control-theoretic perspective on dynamical systems. The automation is modeled as a hybrid system and the pilot is modeled as an intent-based discrete-event system for the given UI which is modeled as a filter of flight information. By inferring and comparing the intents of automation and pilot using a host of information such as the aircraft's continuous states and flight modes, and automation's and pilot's inputs, this paper attempts to validate the given UI via mode confusion detection. The causes responsible for the detected mode confusion are used to suggest improvements to the existing UI design, thereby improving the situational awareness of the pilot and aircraft safety. The framework is tested on two representative mode confusion incidents/accidents and the associated UIs are validated.

show abstract

Flight deck human-automation issue detection via intent inference

Cited by 7 publications

References 12 publications

Using ADOPT Algorithm and Operational Data to Discover Precursors to Aviation Adverse Events

Using ADOPT Algorithm and Operational Data to Discover Precursors to Aviation Adverse Events

Semi-Markov Switching Vector Autoregressive Model-Based Anomaly Detection in Aviation Systems

User Interface Validation using Mode Confusion Detection

Contact Info

Product

Resources

About