Browse Topic: Flight recorders
Rotorcrafts are generally subject to a higher fatal accident rate than other segments of aviation, including commercial and general aviation. The safety improvement for rotorcrafts would directly improve the efficiency of air traffic control, since rotorcrafts operate primarily within low-level airspace; an area that is becoming increasingly complex with new entrants, such as unmanned aircraft systems and urban air mobility. The recent impact of artificial intelligence and deep learning algorithms on various aspects of our lives has led to the investigation of the application of these algorithms in the aviation domain; as it may offer a prime opportunity to enhance safety within the aviation community. In this research, we explore the efficacy, reliability, and, more importantly, the explainability of modern deep learning algorithms. We use machine learning models to predict the attitude (pitch and yaw) of rotorcrafts using video data recorded with ordinary cameras. The cameras were mounted inside the helicopter cockpit and recorded outside view through windshield continually during the flight. We train four different architectures of convolutional neural networks (CNNs), i.e., VGG16, VGG19, ResNet50, and Xception. The models achieved 90%, 91%, 88%, and 88%, respectively, average attitude prediction accuracy on the test video dataset. Furthermore, we use gradient class activation maps (grad-CAM) to ascertain the features and regions of the image that influenced the model to make a specific prediction. We show that CNNs learn to focus on similar features as human operators (pilots), i.e., the natural horizon curve. Our findings demonstrate the feasibility of using deep learning models for attitude prediction from f light videos recorded using ordinary inexpensive cameras. The proposed video analytics framework provides a cost-effective means to supplement traditional Flight Data Recorders (FDR); a technology that is often beyond the financial reach of most general aviation rotorcraft operators.
Spatial Disorientation (SD) mishaps account for the greatest loss of lives in both military and civilian aviation worldwide. When no mechanical cause of a mishap is identified, mishap investigators can use flight data recorder information to populate perceptual models with aircraft flight parameters in order to confirm or deny that pilot SD was the probable cause of the mishap. Current perceptual model weaknesses include the inability to analyze hover and hover-transition mishaps and not accounting for sensory inputs from the auditory and somatosensory systems. The authors have conducted in-flight helicopter perceptual threshold studies to extend the model envelop to include hover as well as a series of tactile cueing in-flight studies in fixed-wing aircraft to permit the inclusion of somatosensory information into the model. This expanded model, by including all sensory modalities, now provides a probable solution to prevention of SD mishaps by continuously maintaining spatial orientation via multisensory cueing. Examples of application of the model to recent high-profile mishaps are included.
Over the last 100 years, the automobile has become integrated in a fundamental way into the broader economy. A broad and deep ecosystem has emerged, and critical components of this ecosystem include insurance, after-market services, automobile retail sales, automobile lending, energy suppliers (e.g., gas stations), medical services, advertising, lawyers, banking, public planners, and law enforcement. These components - which together represent almost $2 trillion of the U.S. economy - are in equilibrium based on the current capabilities of automotive technology. However, the advent of autonomous vehicles (AVs) and technologies like electrification have the potential to significantly disrupt the automotive ecosystem. The critical cog governing the rate and pace of this shift is the management of the test and verification of AVs. In this SAE EDGE™ report, six senior industry leaders in the impacted ecosystems essay articles which describe sectors of the current automotive ecosystem and the manner in which AV technology can potentially reshape them - providing a mosaic of the massive infrastructure shifts which will be required to absorb AV technologies. NOTE: SAE EDGE™ Research Reports are intended to identify and illuminate key issues in emerging, but still unsettled, technologies of interest to the mobility industry. The goal of SAE EDGE™ Research Reports is to stimulate discussion and work in the hope of promoting and speeding resolution of identified issues. SAE EDGE™ Research Reports are not intended to resolve the issues they identify or close any topic to further scrutiny. Click here to access the full SAE EDGETM Research Report portfolio.
As the premier agency for promoting and insuring aviation safety, the Federal Aviation Administration (FAA) continues to promote and highlight the importance of participating in aviation Flight Data Monitoring (FDM) programs to improve flight safety and operational efficiency. Indeed, recorder safety is one of the agency's top 10 most wanted list of safety improvements in 2017-2018. The FAA, National Transportation Safety Board (NTSB), and the United States Helicopter Safety Team (USHST) are strong proponents of recorder use. These organizations and other industry partners are working together to implement a helicopter safety enhancement that promotes the use of flight data recorders as a mechanism to reduce the helicopter fatal accident rate. However, despite these best efforts to reduce the fatal accident rate with this lifesaving technology, barriers to implementation exist. These include initial costs of flight data recorders which can range from 9,000 - 50,000, on average. These costs can be significant for small operators and they combine to prohibit the widespread adoption of FDM by the rotorcraft community. Thus, rotorcraft, in general, typically have a lower participation rate in FDM programs than other forms of aviation (i.e. commercial fixed-wing or part 121 airline operations). On the other hand, even small helicopter operators often have access to or the financial means to purchase one or more off-the-shelf video cameras, which can be mounted inside the cockpit. These cameras offer an alternative to traditional flight data recorders as well as a means to augment them with supplementary data not always available depending on the type of Flight Data Recorder (FDR) installed in the helicopter. On board video data offers several possibilities for improving safety including flight replay, as well as the ability to extract information from the external scene such as readings of instrument panel gauges. As part of our research approach, we analyzed video data from cameras recording the instrument panel and compared these values against ground truth data from the flight data recorder. These values formed the training dataset for our video analytic framework. To analyze this information, we first cropped the gauge of interest (i.e. airspeed indicator, tachometer, engine oil temperature/pressure) in each frame of every video. The gauge image, extracted from all videos, were subsequently fed to train a deep Convolutional Neural Network (CNN) using the FDR measurements as ground truth. We trained Resnet50 CNN models for airspeed, engine oil temperature/pressure, and tachometer gauges. These models obtained 78%, 89%, 89%, and 88% validation accuracy on airspeed, engine oil temperature/pressure, and tachometer gauges, respectively. To further demonstrate the feasibility, we used the trained models to retrieve airspeed and engine oil values from the complete flight profile. We observed that the our models predicted trajectories for gauges closely follow the actual sensory values recorded by FDR. Such solution results in an effective flight data analysis tool as well as improved safety and operational efficiency of rotorcraft. These results demonstrate the feasibility of an inexpensive cockpit camera solution that would facilitate participation in FDM programs even for legacy helicopters that may otherwise require significant installation work.
The CH146 Griffon helicopter is a Federal Aviation Administration (FAA) and Transport Canada certified commercial helicopter used by Canadian Armed Forces in military role, which is distinguished from the original design intent of the helicopter for commercial use. A regime based Structural Usage Monitoring (SUM) program is developed by Bell to meet the Canadian Government requirements documented in the Technical Airworthiness Manual (TAM). As part of the CH146 SUM, helicopter usage data recorded by the Flight Data Recorder (FDR) system is collected and flight conditions (regimes) are defined by using dedicated software tools. Helicopter usage is then compared with the baseline and by using the actual helicopter data on selected critical components and fatigue damage accumulation is performed. Finally, the calculated damage is evaluated to define recommended structural maintenance actions. Development of the CH146 SUM is completed and more than 50,000 flight hours accumulated FDR data is processed. The acceptance of the SUM program by the Canadian Department of Defence (DND) is expected in the summer of 2019.
This paper presents the results from several load estimation methods developed at the National Research Council Canada (NRC) which enable the estimation of helicopter loads and tracking load exceedances and fatigue damage for a targeted component using computational intelligence techniques. The approach relies only on flight state and control system (FSCS) parameters, such as those recorded by a flight data recorder (FDR), and can also be applied to legacy aircraft or to those aircraft not equipped with HUMS. The methodologies adapt to the input data available so are not constrained to one particular system or platform, and enable the estimation of loads through the duration of a manoeuvre instead of assuming a constant load for an entire manoeuvre. So far, the three methods have been tested on data obtained from two different helicopter platforms, the S-70A-9 Australian Black Hawk and the CH-146 Griffon (Bell 412). Significant improvements are made over previous results presented for the S-70A-9 Black Hawk using the NRC developed Signal Approximation Method (SAM) while using uniquely FSCS parameters obtained from a FDR to obtain full manouevre dynamic load signals in time. Furthermore, the application of the technology to a different platform and different original equipment manufacturer (OEM), namely the CH-146 Griffon, demonstrates the possibility of applying this methodology and its adaptability across different platforms.
ABSTRACT The US Army Condition Based Maintenance program collects data from Health and Usage Monitoring Systems, Flight Data Recorders, Maintenance Records, and Reliability Databases. These data sources are not integrated, but decisions regarding the health of aircraft components are dependent upon the information stored within them. The Army has begun an effort to bring these data sources together using Machine Learning algorithms. Two prototypes will be built using decision-making machines: one for an engine output gearbox and another for a turbo-shaft engine. This paper will discuss the development of these prototypes and provide the path forward for implementation. The importance of determining applicable error penalty methods for machine learning algorithms for aerospace applications is explored. The foundations on which the applicable dataset is built are also explored, showing the importance of cleaning disparate datasets. The assumptions necessary to generate the dataset for learning are documented. The dataset is built and ready for unsupervised and supervised learning techniques to be applied.
Items per page:
50
1 – 50 of 102