Analysis Techniques for Automatic Flight Maneuver Detection and Evaluation

Citations Over TimeTop 20% of 2010 papers

Abstract

Information obtained from analyses of data from aircraft onboard data collection systems can be used to enhance the objective evaluation of aircrew and aircraft systems performance. The analyses of raw and processed flight data enable capabilities such as Condition Based Maintenance (CBM) and Flight Operational Quality Assurance (FOQA) to provide users with quantifiable and actionable information to improve safety and operational efficiencies. Specific flight maneuvers and predefined events can be detected and evaluated on an individual and multi-flight basis. Aggregate analyses can also provide trends with regards to the frequency of occurrence of maneuvers and events of interest, and possible correlation to causal and contributing factors as well as other in-depth enterprise level insight. However, rule-based analysis techniques have practical limits with respect to the accurate detection and evaluation of certain flight maneuvers and events. Whereas rulebased algorithms are adequate for some analyses, maneuvers and events of greater complexity may require more advanced analysis approaches to ensure reliable results. In addition, analysis techniques need to include the ability to detect and mitigate data anomalies that typically occur to varying degrees with different flight data collection systems. Anomalous data can cause erroneous and misleading analytical results that could lead to inappropriate actions by users and justifiable loss of confidence in the analysis capabilities provided. Therefore, analysis techniques that exceed the practical limits of rulebased methods, as well as detect and correct anomalous data, are required to provide users reliable information. The Department of the Navy (DON) is implementing a knowledge management process known as Military Flight Operations Quality Assurance (MFOQA) based on the principles of commercial FOQA. In order to mature MFOQA capabilities for Naval aviation, analysis techniques that go beyond the predominantly rule-based algorithms currently used are being investigated and developed in order to maximize the availability, reliability, and accuracy of actionable information obtained from flight data analyses. The objective is to enhance the MFOQA process with more autonomous, accurate, and timely functionalities that approximate the human reasoning process and include appropriate feedback mechanisms for evolving detection and evaluation capabilities as well as to successfully mitigate anomalous data. In addition to improving the accuracy of analytical results, comprehensive automated analysis techniques provide the capability to analyze large volumes of data that would otherwise be impractical with human analysts. This paper discusses several analysis techniques intended to automatically detect discrete multivariable events in recorded flight data for future MFOQA enhancements.

Related Papers

• → Self-monitoring to improve robustness of 3D object tracking for robotics(2011)4 cited
• → Object-oriented stripe structured-light vision-guided robot(2017)2 cited
• → 6-DOF object localization by combining monocular vision and robot arm kinematics(2017)1 cited
• → 3-D tracking of a moving object by an active stereo vision system(2002)9 cited
• → Robust object tracking based on RGB-D camera(2014)