Search NASA⌕ Search

SEARCH · Search NASA

Results for “regression testing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 343 records · Page 19

Global Analysis of Empirical Relationships Between Annual Climate and Seasonality of NDVI

This study describes the use of satellite data to calibrate a new climate-vegetation greenness function for global change studies. We examined statistical relationships between annual climate indexes (temperature, precipitation, and surface radiation) and seasonal attributes of the AVHRR Normalized Difference Vegetation Index (NDVI) time series for the mid-1980s in order to refine our empirical understanding of intraannual patterns and global abiotic controls on natural vegetation dynamics. Multiple linear regression results using global l(sup o) gridded data sets suggest that three climate indexes: growing degree days, annual precipitation total, and an annual moisture index together can account to 70-80 percent of the variation in the NDVI seasonal extremes (maximum and minimum values) for the calibration year 1984. Inclusion of the same climate index values from the previous year explained no significant additional portion of the global scale variation in NDVI seasonal extremes. The monthly timing of NDVI extremes was closely associated with seasonal patterns in maximum and minimum temperature and rainfall, with lag times of 1 to 2 months. We separated well-drained areas from l(sup o) grid cells mapped as greater than 25 percent inundated coverage for estimation of both the magnitude and timing of seasonal NDVI maximum values. Predicted monthly NDVI, derived from our climate-based regression equations and Fourier smoothing algorithms, shows good agreement with observed NDVI at a series of ecosystem test locations from around the globe. Regions in which NDVI seasonal extremes were not accurately predicted are mainly high latitude ecosystems and other remote locations where climate station data are sparse.

Potter, C. S.↗

Peregrine 100-km Sounding Rocket Project

The Peregrine Sounding Rocket Program is a joint basic research program of NASA Ames Research Center, NASA Wallops, Stanford University, and the Space Propulsion Group, Inc. (SPG). The goal is to determine the applicability of this technology to a small launch system. The approach is to design, build, and fly a stable, efficient liquefying fuel hybrid rocket vehicle to an altitude of 100 km. The program was kicked off in October of 2006 and has seen considerable progress in the subsequent 18 months. This research group began studying liquifying hybrid rocket fuel technology more than a decade ago. The overall goal of the research was to gain a better understanding of the fundamental physics of the liquid layer entrainment process responsible for the large increase in regression rate observed in these fuels, and to demonstrate the effect of increased regression rate on hybrid rocket motor performance. At the time of this reporting, more than 400 motor tests were conducted with a variety of oxidizers (N2O, GOx, LOx) at ever increasing scales with thrust levels from 5 to over 15,000 pounds (22 N to over 66 kN) in order to move this technology from the laboratory to practical applications. The Peregrine program is the natural next step in this development. A number of small sounding rockets with diameters of 3, 4, and 6 in. (7.6, 10.2, and 15.2 cm) have been flown, but Peregrine at a diameter of 15 in. (38.1 cm) and 14,000-lb (62.3-kN) thrust is by far the largest system ever attempted and will be one of the largest hybrids ever flown. Successful Peregrine flights will set the stage for a wide range of applications of this technology.

Zilliac, Gregory↗

Trusted Communication: Utilizing Speech Communication to Enhance Human-Machine Teaming Success

An area of increasing interest for the next generation of aircraft is autonomy and the integration of increasingly autonomous systems into the national airspace. Such an integration requires humans to work closely with autonomous systems, forming teams. Our hypothesis is that a team composed of both humans and autonomous systems will operate better than either entity alone. We have existing procedures for certifying pilots to operate in the national airspace and are currently working on methods for validating the function of autonomous systems, however we have no method in place for assessing the interaction of these two disparate systems. Communication is one avenue. This paper will examine the use of language as a metric for ascertaining human-machine teaming effectiveness. A proof-of-concept of the application of two communication-based analysis techniques, Linguistic Inquiry and Word Count (LIWC) and Latent Semantic Analysis (LSA), for the prediction of success in human/chatbot teaming was conducted. By running these analyses over data from the 2014 and 2015 Loebner Prize competitions of human/chatbot teaming, numerical scores were obtained that can be associated with scores provided by human judges during the competition. Correlating their LIWC and LSA data with the scores provided by the judges, and using linear regression over this correlation, formulae were obtained that predict the score of human/chatbot interaction. These formulae were tested over the 2013 Loebner Prize transcripts, determining that, though there was strong correlation between predicted and actual scores, the predictive success of this method was not strong. However, with specialized topic spaces and lexica, as well as larger data sets, the predictive power of these metrics will improve. Given the importance of providing metrics for human-machine system team success and given the promise shown by the communication-basedLIWCand LSAmethods, continuing research in this area is necessary. After examining the potential for using communication and spoken language as a metric for the success of human/autonomous system teaming, this paper then examines aspects inherent to communication systems that may contribute to unreliability and reduced trust. Modern natural language processing tools rely on deep learning algorithms to create language rules that produce accurate results, but these rules are uninterpretable. The resulting blackbox system lacks transparency necessary for full validation and complete trust. Additionally, speech-based interfaces pose other difficulties to developing coordinated teamwork between humans and autonomous systems. Human communication is infrequently limited to speech only, instead usually relying on a combination of verbal, gestural, and general body language communication. Reducing an analysis of team effectiveness to a study of spoken language alone is problematic as it leaves these other equally important forms of communication out. This paper will examine these problems and the general deficiencies in speech-based metrics for human-machine teaming.

E L Meszaros↗

Vivid Motor Imagery as an Adaptation Method for Head Turns on a Short-Arm Centrifuge

Artificial gravity (AG) has been proposed as a potential countermeasure to the debilitating physiological effects of long duration space flight. The most economical means of implementing AG may be through the use of a short-radius (2m or less) centrifuge. For such a device to produce gravitational forces comparable to those on earth requires rotation rates in excess of 20 revolutions per minute (rpm). Head turns made out of the plane of rotation at these rates, as may be necessary if exercise is combined with AG, result in cross-coupled stimuli (CCS) that cause adverse side effects including motion sickness, illusory sensations of motion, and inappropriate eye movements. Recent studies indicate that people can adapt to CCS and reduce these side effects by making multiple head turns during centrifuge sessions conducted over consecutive days. However, about 25% of the volunteers for these studies have difficulty tolerating the CCS adaptation paradigm and often drop out due to motion sickness symptoms. The goal of this investigation was to determine whether vivid motor imagery could be used as a pseudostimulus for adapting subjects to this unique environment. Twenty four healthy human subjects (14 males, 10 females), ranging in age from 21 to 48 years (mean 33, sd 7 years) took part in this study. The experimental stimuli were produced using the NASA JSC short-arm centrifuge (SAC). Subjects were oriented supinely on this device with the nose pointed toward the ceiling and head centered on the axis of rotation. Thus, centrifuge rotation was in the body roll plane. After ramp-up the SAC rotated clockwise at a constant rate of 23 rpm, producing a centrifugal force of approximately 1 g at the feet. Semicircular canal CCS were produced by having subjects make yaw head turns from the nose up (NU) position to the right ear down (RED) position and from RED to NU. Each head turn was completed in about one second, and a 30 second recovery period separated consecutive head movements. Participants were randomly assigned to one of three groups (n=8 per group): physical adapters (PA), mental adapters (MA), or a control group (CG). Each subject participated in a one hour test session on each of three consecutive days. Each test session consisted of an initial (preadaptation) period during which the subject performed six CCS maneuvers in the dark, followed by an adaptation period with internal lighting on the centrifuge, and a final (postadaptation) period during which six more CCS maneuvers were performed in the dark. For the PA group, the adaptation period consisted of performing 30 additional CCS maneuvers in the light. For the MA and CG group the centrifuge was ramped down to 0 rpm after the pre-adaptation period and ramped back up to 23 rpm before the post-adaptation period. For the both of these groups, the adaptation period consisted of making 30 CCS maneuvers in the light with the centrifuge stationary (so no cross-coupling occurred). MA group subjects were instructed to vividly imagine the provocative sensations produced by the preadaptation CCS maneuvers in terms of magnitude, duration, and direction of illusory body tilt, as well as any accompanying levels of motion sickness. CG group subjects were asked to answer low imagery content questions (trivial pursuit) during each adaptation period head turn. During the 30 second recovery following each head turn, psychophysical data were collected including self reports of motion sickness, magnitude and direction estimates of illusory body tilt, and the overall duration of these sensations. A multilevel mixed effects linear regression analysis performed on all response variables indicated that all three groups experienced some psychophysical adaptation across the three test sessions. For illusory tilt magnitude, the PA group exhibited the most overall adaptation, followed by the MA group, and the CG group. The slopes of these adaptation trajectories by group over day were significantly diffent from one another. For the perceived duration of sensations, the CG group again exhibited the least amount of adaptation. However, the rates of adaptation of the PA and the MA groups were indistinguishable, suggesting that the imagined pseudostimulus appeared to be just as effective a means of adaptation as the actual stimulus. The MA group's rate of adaptation to motion sickness symptoms was also comparable to the PA group. The use of vivid motor imagery may be an effective method for adapting to the illusory sensations and motion sickness symptoms produced by cross-coupled stimuli. For space-based AG applications, this technique may prove quite useful in retaining astronauts considered highly susceptible to motion sickness as it reduces the number of actual CCS required to attain adaptation.

Newby, N. J.↗

Unsteady hovering wake parameters identified from dynamic model tests, part 1

The development of a 4-bladed model rotor is reported that can be excited with a simple eccentric mechanism in progressing and regressing modes with either harmonic or transient inputs. Parameter identification methods were applied to the problem of extracting parameters for linear perturbation models, including rotor dynamic inflow effects, from the measured blade flapping responses to transient pitch stirring excitations. These perturbation models were then used to predict blade flapping response to other pitch stirring transient inputs, and rotor wake and blade flapping responses to harmonic inputs. The viability and utility of using parameter identification methods for extracting the perturbation models from transients are demonstrated through these combined analytical and experimental studies.

Hohenemser, K. H.↗

Evaluation, Analysis, and Application of Internal Strain-Gage Balance Data

Experimental processes, analytical methods, and numerical algorithms are described that may be used to predict the forces and moments of an internal strain-gage balance during a wind tunnel test. First, the control volume model of a strain-gage balance and the concepts of load state, load space, and output space are introduced. These important abstractions provide a better understanding of fundamental characteristics of different balance load prediction approaches. Then, the description of strain-gage balance data and the definition of the primary gage sensitivity is discussed. Afterwards, basic elements of the calibration of a typical six-component balance are reviewed. Two fundamentally different balance load prediction methods, the processing of check loads, and related topics are also discussed. Three real-world balance data examples are reviewed in great detail to illustrate typical analysis results for a variety of strain-gage balance designs. Finally, important observations are summarized and recommendations are provided. Additional information and detailed mathematical derivations can be found in the appendices of the document. They include the following topics: balance terminology, definitions of important statistical metrics, balance axis system conventions, balance load transformations, the combined load diagram, electrical output format options, bi-directional gage output characteristics, determination of the natural zeros, derivation of two balance load prediction methods, description of two tare load iteration algorithms, modeling of balance temperature effects, basics of three-component moment balances, definition of the percent contribution, detection of linear and near-linear dependencies in balance calibration data, a regression model term selection algorithm, and other related topics.

wind tunnel test↗

Physics-Based Machine Learning Methods for U-235 Forensics Signatures

Signatures of low-intensity U-235 sources have been recently studied by utilizing a variety of machine learning (ML) classifiers using features derived from gamma spectral measurements collectedunder structured campaigns. Several ML classifiers, such as ensemble of tress and classification trees, revealed misleadingly-optimistic training error due to over-fitting, and furthermore,their performance is not directly relatable to the physical properties due to their data-driven, opaque designs. We present a regression-based ML method that first estimates the inverse distanceto the source and then utilizes a threshold to infer its presence, by representing the background as a source located at an infinite distance. For the inverse distance estimation, we study the ensembleof trees and Gaussian process regression methods, and a hyper parameter auto-tuning and selection method that employs five regression estimators. These methods avoid the over-fittingobserved in several ML classifiers, while providing the classification error nearly comparable to them based on independent test data. Their error is directly related to estimates of the inversephysical distance to source, and the precision of error determines the seperability property that determines the false alarm and missed detection rates. The property of monotonic decrease of thesource strength with increasing detector distance combined with Poisson distribution of measurements is utilized to analytically validate these methods by deriving the generalization equations ofunderlying regression methods.

Rao, Nageswara↗

Advanced Signal Decomposition Analysis and Anomaly Detection in Photovoltaic Systems

With the rapid expansion of large-scale photovoltaic (PV) plants, it is paramount for solar stakeholders to understand the reliability and efficiency of their plants to inform maintenance decisions, increase production, and understand the design factors that impact performance. Diagnosing underperformance in PV plants is challenging due to the relatively few monitoring points with respect to the large geographic footprint of the plant. This work introduces a cutting-edge method that transforms the analysis and management of key factors influencing PV plant performance, including performance loss rate (PLR), recoverable soiling, and major system changes. Identifying these factors is critical for deriving actionable insights. Leveraging advanced analytical techniques such as wavelet transformation, robust regression, and extreme point analysis, this approach provides a nuanced understanding of these factors. This method has been tested across two synthetic datasets and one real dataset, consistently surpassing existing benchmarks by achieving a lower median mean absolute error and reduced error variability across all comparable components.

14 SOLAR ENERGY↗

The European Southern Observatory-MIDAS table file system

The new and substantially upgraded version of the Table File System in MIDAS is presented as a scientific database system. MIDAS applications for performing database operations on tables are discussed, for instance, the exchange of the data to and from the TFS, the selection of objects, the uncertainty joins across tables, and the graphical representation of data. This upgraded version of the TFS is a full implementation of the binary table extension of the FITS format; in addition, it also supports arrays of strings. Different storage strategies for optimal access of very large data sets are implemented and are addressed in detail. As a simple relational database, the TFS may be used for the management of personal data files. This opens the way to intelligent pipeline processing of large amounts of data. One of the key features of the Table File System is to provide also an extensive set of tools for the analysis of the final results of a reduction process. Column operations using standard and special mathematical functions as well as statistical distributions can be carried out; commands for linear regression and model fitting using nonlinear least square methods and user-defined functions are available. Finally, statistical tests of hypothesis and multivariate methods can also operate on tables.

Peron, M.↗

Productivity Analysis of Public and Private Airports: A Causal Investigation

Around the world, airports are being viewed as enterprises, rather than public services, which are expected to be managed efficiently and provide passengers with courteous customer services. Governments are, increasingly, turning to the private sectors for their efficiency in managing the operation, financing, and development, as well as providing security for airports. Operational and financial performance evaluation has become increasingly important to airport operators due to recent trends in airport privatization. Assessing performance allows the airport operators to plan for human resources and capital investment as efficiently as possible. Productivity measurements may be used as comparisons and guidelines in strategic planning, in the internal analysis of operational efficiency and effectiveness, and in assessing the competitive position of an airport in transportation industry. The primary purpose of this paper is to investigate the operational and financial efficiencies of 22 major airports in the United States and Europe. These airports are divided into three groups based on private ownership (7 British Airport Authority airports), public ownership (8 major United States airports), and a mix of private and public ownership (7 major European Union airports. The detail ownership structures of these airports are presented in Appendix A. Total factor productivity (TFP) model was utilized to measure airport performance in terms of financial and operational efficiencies and to develop a benchmarking tool to identify the areas of strength and weakness. A regression model was then employed to measure the relationship between TFP and ownership structure. Finally a Granger causality test was performed to determine whether ownership structure is a Granger cause of TFP. The results of the analysis presented in this paper demonstrate that there is not a significant relationship between airport TFP and ownership structure. Airport productivity and efficiency is, however dependent upon the level of competition, choice of the market, and regulatory control.

Vasigh, Bijan↗

Post-Flight Back Pain Following International Space Station Missions: Evaluation of Spaceflight Risk Factors

INTRODUCTION Back pain during spaceflight has often been attributed to the lengthening of the spinal column due to the absence of gravity during both short and long-duration missions. Upon landing and re-adaptation to gravity, the spinal column reverts back to its original length thereby causing some individuals to experience pain and muscular spasms, while others experience no ill effects. With International Space Station (ISS) missions, cases of back pain and injury are more common post-flight, but little is known about the potential risk factors. Thus, the purpose of this project was to perform an initial evaluation of reported post-flight back pain and injury cases to relevant spaceflight risk factors in United States astronauts that have completed an ISS mission. METHODS All US astronauts who completed an ISS mission between Expeditions (EXP) 1 and 41 (2000-2015) were included in this evaluation. Forty-five astronauts (36 males and 9 females) completed 50 ISS missions during the study time period, as 5 astronauts completed 2 ISS missions. Researchers queried medical records of the 45 astronauts for occurrences of back pain and injury. A case was defined as any reported event of back pain or injury to the cervical, thoracic, lumbar, sacral, or coccyx spine regions. Data sources for the cases included the Flight Medicine Clinic's electronic medical record; Astronaut Strength, Conditioning and Rehabilitation electronic documentation; the Private Medical Conference tool; and the Space Medicine Operations Team records. Post-flight cases were classified as an early case if reported within 45 days of landing (R + 45) or a late case if reported from R + 46 to R + 365 days after landing (R + 1y). Risk factors in the astronaut population for back pain include age, sex, prior military service, and prior history of back pain. Additionally, spaceflight specific risk factors such as type of landing vehicle and onboard exercise countermeasures were included to evaluate their contribution to post-flight cases. Prior history of back pain included back pain recorded in the medical record within 3 years prior to launch. Landing vehicle was included in the model to discern if more astronauts experienced back pain or injury following a Shuttle or Soyuz landing. Onboard exercise countermeasures were noted for those astronauts who had a mission following 2009 deployment of the Advanced Resistive Exercise Device (aRED) (EXP 19 to 41). T-test and chi-squared tests were performed to evaluate the association between each individual risk factor and post-flight case. Logistic regression was used to evaluate the combined contribution of all the risk factors on post-flight cases. Separate models were calculated for cases reported by R + 45 and R + 1y. RESULTS During the study time period, there were 13 post-flight cases reported by R + 45 and an additional 5 reported by R + 1y. Most of these cases have been reported since EXP 19 with 10 cases by R + 45 and 4 by R + 1y. Individual risk factors of age, sex, landing vehicle, and prior military service were not significantly associated with post-flight cases identified at R + 45 or R + 1y (p greater than 0.05). Having back pain or injury within 3 years prior to launch significantly increased the likelihood of becoming a case by R + 1y (p = 0.041), but not at R+45 (p=0.204). Additionally, astronauts who experienced onboard exercise countermeasures that included aRED had a significantly increased risk of becoming a case at R + 45 (p = 0.024) and R + 1y (p=0.003). Multiple logistic regression evaluating all the risk factors for cases identified no significant risk factors at either the R + 45 or R + 1y time period (p greater than 0.05). Overall model fit was poor for both the R + 45 (R(exp 2) = 0.132) and R + 1y (R(exp 2) = 0.186) cases showing that there are risk factors not represented in our model. CONCLUSIONS Regardless of cause, post-flight cases are reported more often since aRED was deployed in 2009. This may reflect improved documentation or unidentified risk factors. No spaceflight risk factor explains the data fully. Post-flight cases are probably due to multi-faceted factors that are not easily elucidated in the medical data.

Laughlin, M. S.↗

The Use of Absolute-Value Terms in Regression Modeling of Multi-Piece Force Balances

Different aspects of the use of absolute-value terms in regression models of the electrical outputs of multi-piece force balance calibration data are discussed. First, characteristics of a variety of regression model term combinations with absolute-value terms are reviewed that are currently used in the aerospace testing community to fit the gage outputs of a balance. Then, a semi-empirical test is presented that quantifies bidirectional characteristics of the balance bridge outputs. Several diagnostic methods are discussed to assess the severity of near-linear dependencies between regressors of models with absolute-value terms. In particular, connections between the linear, absolute-value, quadratic, signed quadratic, and cubic terms are studied in greater detail. Data from an automated calibration of NASAs MK29B force balance are used to illustrate the most important observations and results. Rules of thumb that variance-inflation factors be less than 10 must be relaxed when using absolute-value terms to describe bidirectional balances.

calibration analysis↗

Evaluation of heart rate changes: electrocardiographic versus photoplethysmographic methods

The heart rate (HR) variation to forced deep breathing (HRDB) and to the Valsalva maneuver (Valsalva ratio; VR) are the two most widely used tests of cardiovagal function in human subjects. The HR is derived from a continuously running electrocardiographic (ECG) recording. Recently, HR derived from the arterial waveform became available on the Finapres device (FinapHR), but its ability to detect rapid changes in HR remains uncertain. We therefore evaluated HRDB and VR derived from FinapHR using ECG-derived HR (ECGHR) recordings as the standard. We also compared the averaged HR on Finapres (Finapav) with beat-to-beat Finapres (FinapBB) values. Studies were undertaken in 12 subjects with large HR variations: age, 34.5 +/- 9.3 (SD) years; six males and six females. FinapBB values were superimposable upon ECGHR for both HRDB and VR. In contrast, Finapav failed to follow ECGHR for HRDB and followed HRECG with a lag for the VR. To evaluate statistically how closely FinapHR approximated ECGHR, we undertook regression analysis, using mean values for each subject. To compare the two methods, we evaluated the significance of the difference between test and standard values. For HRDB, FinapBB reproducibly recorded HR (R2 = 0.998), and was significantly (p = 0.001) better than Finapav (R2 = 0.616; p < 0.001). For VR, HRBB generated a VR that was not significantly different from the correct values, while HRav generated a value that was slightly but consistently lower than the correct values (p < 0.001). We conclude that FinapHR reliably records HR variations in the beat-to-beat mode for cardiovascular HR tests.

Non-NASA Center↗

In-flight demonstration of a Real-Time Flush Airdata Sensing (RT-FADS) system

A prototype real-time flush airdata sensing (RT-FADS) system has been developed and flight tested at the NASA Dryden Flight Research Center. This system uses a matrix of pressure orifices on the vehicle nose to estimate airdata parameters in real time using nonlinear regression. The algorithm is robust to sensor failures and noise in the measured pressures. The RT-FADS system has been calibrated using inertial trajectory measurements that were bootstrapped for atmospheric conditions using meteorological data. Mach numbers as high as 1.6 and angles of attack greater than 45 deg have been tested. The system performance has been evaluated by comparing the RT-FADS to the ship system airdata computer measurements to give a quantitative evaluation relative to an accepted measurement standard. Nominal agreements of approximately 0.003 in Mach number and 0.20 deg in angle of attack and angle of sideslip have been achieved.

Whitmore, Stephen A.↗

An Alternative Flight Software Trigger Paradigm: Applying Multivariate Logistic Regression to Sense Trigger Conditions using Inaccurate or Scarce Information

In late 2014, NASA will fly the Orion capsule on a Delta IV-Heavy rocket for the Exploration Flight Test-1 (EFT-1) mission. For EFT-1, the Orion capsule will be flying with a new GPS receiver and new navigation software. Given the experimental nature of the flight, the flight software must be robust to the loss of GPS measurements. Once the high-speed entry is complete, the drogue parachutes must be deployed within the proper conditions to stabilize the vehicle prior to deploying the main parachutes. When GPS is available in nominal operations, the vehicle will deploy the drogue parachutes based on an altitude trigger. However, when GPS is unavailable, the navigated altitude errors become excessively large, driving the need for a backup barometric altimeter. In order to increase overall robustness, the vehicle also has an alternate method of triggering the drogue parachute deployment based on planet-relative velocity if both the GPS and the barometric altimeter fail. However, this velocity-based trigger results in large altitude errors relative to the targeted altitude. Motivated by this challenge, this paper demonstrates how logistic regression may be employed to automatically generate robust triggers based on statistical analysis. Logistic regression is used as a ground processor pre-flight to develop a classifier. The classifier would then be implemented in flight software and executed in real-time. This technique offers excellent performance even in the face of highly inaccurate measurements. Although the logistic regression-based trigger approach will not be implemented within EFT-1 flight software, the methodology can be carried forward for future missions and vehicles.

Smith, Kelly M.↗

An Alternative Flight Software Trigger Paradigm: Applying Multivariate Logistic Regression to Sense Trigger Conditions Using Inaccurate or Scarce Information

In late 2014, NASA will fly the Orion capsule on a Delta IV-Heavy rocket for the Exploration Flight Test-1 (EFT-1) mission. For EFT-1, the Orion capsule will be flying with a new GPS receiver and new navigation software. Given the experimental nature of the flight, the flight software must be robust to the loss of GPS measurements. Once the high-speed entry is complete, the drogue parachutes must be deployed within the proper conditions to stabilize the vehicle prior to deploying the main parachutes. When GPS is available in nominal operations, the vehicle will deploy the drogue parachutes based on an altitude trigger. However, when GPS is unavailable, the navigated altitude errors become excessively large, driving the need for a backup barometric altimeter to improve altitude knowledge. In order to increase overall robustness, the vehicle also has an alternate method of triggering the parachute deployment sequence based on planet-relative velocity if both the GPS and the barometric altimeter fail. However, this backup trigger results in large altitude errors relative to the targeted altitude. Motivated by this challenge, this paper demonstrates how logistic regression may be employed to semi-automatically generate robust triggers based on statistical analysis. Logistic regression is used as a ground processor pre-flight to develop a statistical classifier. The classifier would then be implemented in flight software and executed in real-time. This technique offers improved performance even in the face of highly inaccurate measurements. Although the logistic regression-based trigger approach will not be implemented within EFT-1 flight software, the methodology can be carried forward for future missions and vehicles.

Smith, Kelly M.↗

An Alternative Flight Software Paradigm: Applying Multivariate Logistic Regression to Sense Trigger Conditions using Inaccurate or Scarce Information

In late 2014, NASA will fly the Orion capsule on a Delta IV-Heavy rocket for the Exploration Flight Test-1 (EFT-1) mission. For EFT-1, the Orion capsule will be flying with a new GPS receiver and new navigation software. Given the experimental nature of the flight, the flight software must be robust to the loss of GPS measurements. Once the high-speed entry is complete, the drogue parachutes must be deployed within the proper conditions to stabilize the vehicle prior to deploying the main parachutes. When GPS is available in nominal operations, the vehicle will deploy the drogue parachutes based on an altitude trigger. However, when GPS is unavailable, the navigated altitude errors become excessively large, driving the need for a backup barometric altimeter to improve altitude knowledge. In order to increase overall robustness, the vehicle also has an alternate method of triggering the parachute deployment sequence based on planet-relative velocity if both the GPS and the barometric altimeter fail. However, this backup trigger results in large altitude errors relative to the targeted altitude. Motivated by this challenge, this paper demonstrates how logistic regression may be employed to semi-automatically generate robust triggers based on statistical analysis. Logistic regression is used as a ground processor pre-flight to develop a statistical classifier. The classifier would then be implemented in flight software and executed in real-time. This technique offers improved performance even in the face of highly inaccurate measurements. Although the logistic regression-based trigger approach will not be implemented within EFT-1 flight software, the methodology can be carried forward for future missions and vehicles

Smith, Kelly↗