Pub Date : 2009-03-07DOI: 10.1109/AERO.2009.4839504
G. Roosevelt, D. Bueno, J. Haque, Weston Roper, Thomas Romanko
SerDes (Serializer/Deserializer) is a key component of serial communication architecture for high-speed servers and communications networking systems and point to point communication links. It is a vital building block for space-based high-speed data communications. Honeywell is enabling optimized communication systems with a SerDes macro-cell as part of their HX5000 Rad-Hard ASIC design platform and a SerDes standard part for next-generation aerospace systems. Communication networks developed with SerDes embedded in an Application Specific Integrated Circuits (ASIC) achieves significantly lower power, higher data throughput and more efficient ASIC area usage than traditional approaches for high-speed backplanes and box-to-box networking systems. When implemented on a Space qualified ASIC platform, SerDes offers the communication reliability and efficiency required in harsh space environments.
{"title":"Rad-Hard high speed serial communication using Honeywell SerDes macros","authors":"G. Roosevelt, D. Bueno, J. Haque, Weston Roper, Thomas Romanko","doi":"10.1109/AERO.2009.4839504","DOIUrl":"https://doi.org/10.1109/AERO.2009.4839504","url":null,"abstract":"SerDes (Serializer/Deserializer) is a key component of serial communication architecture for high-speed servers and communications networking systems and point to point communication links. It is a vital building block for space-based high-speed data communications. Honeywell is enabling optimized communication systems with a SerDes macro-cell as part of their HX5000 Rad-Hard ASIC design platform and a SerDes standard part for next-generation aerospace systems. Communication networks developed with SerDes embedded in an Application Specific Integrated Circuits (ASIC) achieves significantly lower power, higher data throughput and more efficient ASIC area usage than traditional approaches for high-speed backplanes and box-to-box networking systems. When implemented on a Space qualified ASIC platform, SerDes offers the communication reliability and efficiency required in harsh space environments.","PeriodicalId":117250,"journal":{"name":"2009 IEEE Aerospace conference","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125365378","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-03-07DOI: 10.1109/AERO.2009.4839555
L. Dreyer
SpaceX is committed to revolutionizing access to space by providing highly reliable, low cost launch services. To this end, SpaceX has been developing a family of orbital transportation solutions comprised of SpaceX-developed proprietary technologies, as well as off-the-shelf and minimally modified hardware from leading aerospace subcontractors. SpaceX is currently offering launch services on the Falcon 1, Falcon 9, and Falcon 9 Heavy launch vehicles as well as commercial flights on the Dragon spacecraft. Founded in mid-2002, SpaceX has been busy. This paper provides the latest developments and progress, including recent launches and future plans.
{"title":"Latest developments on SpaceX's Falcon 1 and Falcon 9 launch vehicles and Dragon spacecraft","authors":"L. Dreyer","doi":"10.1109/AERO.2009.4839555","DOIUrl":"https://doi.org/10.1109/AERO.2009.4839555","url":null,"abstract":"SpaceX is committed to revolutionizing access to space by providing highly reliable, low cost launch services. To this end, SpaceX has been developing a family of orbital transportation solutions comprised of SpaceX-developed proprietary technologies, as well as off-the-shelf and minimally modified hardware from leading aerospace subcontractors. SpaceX is currently offering launch services on the Falcon 1, Falcon 9, and Falcon 9 Heavy launch vehicles as well as commercial flights on the Dragon spacecraft. Founded in mid-2002, SpaceX has been busy. This paper provides the latest developments and progress, including recent launches and future plans.","PeriodicalId":117250,"journal":{"name":"2009 IEEE Aerospace conference","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125374473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-03-07DOI: 10.1109/AERO.2009.4839345
A. Sengupta, J. Kulleck, S. Sell, John W. Van Norman, M. Mehta, Mark Pokora
The Mars Science Laboratory (MSL) Mission will land a 900-kg rover on the surface of Mars in 2010. Four Mars Lander Engines (MLE's) will be fired during the final propulsive descent to maintain a 0.75 m/s vertical rate of descent, in support of a tethered landing approach referred to as the “Sky-Crane”. At 20 m above the surface the rover will be lowered on a bridle as it continues to descend. At touch-down, a minimum of 6.5 m of vertical separation are provided between the engines nozzle exit plane and the ground-surface below [1]. This maneuver was chosen in part to minimize the ground/soil interaction that occurs when rocket engine plumes are fired into a soil media. In spite of the 6.5 m altitude above the surface, surface impingement pressures are expected to reach in excess of 2000 Pa, a metric previously established by the Viking program to mitigate soil bearing capacity failure. Plume-ground interaction has been a concern of Lunar and Mars propulsive landings for some time, but was not an issue for the Mars Pathfinder and Mars Explorer Rover era due to their use of airbag landing systems [2][3].This was also a concern of the Phoenix lander program, which fired twelve pulsed hydrazine monopropellant thrusters for its final descent and touch-down [4]. Phoenix was concerned with plume impingement soil interaction due to its high surface impingement pressure and potential for diffused gas eruptions. Phoenix was also concerned with landing site alteration due to its lack of mobility as well as instrument and solar array contamination issues. As MSL will operate in a regime that will result in ground-soil erosion a plume-ground interaction program has been undertaken to quantify the amount of soil erosion, namely the trajectory and number flux of particulates and the contamination and erosion this can impart to sensitive instruments and thermal surface coatings.
{"title":"Mars Lander Engine plume impingement environment of the Mars Science Laboratory","authors":"A. Sengupta, J. Kulleck, S. Sell, John W. Van Norman, M. Mehta, Mark Pokora","doi":"10.1109/AERO.2009.4839345","DOIUrl":"https://doi.org/10.1109/AERO.2009.4839345","url":null,"abstract":"The Mars Science Laboratory (MSL) Mission will land a 900-kg rover on the surface of Mars in 2010. Four Mars Lander Engines (MLE's) will be fired during the final propulsive descent to maintain a 0.75 m/s vertical rate of descent, in support of a tethered landing approach referred to as the “Sky-Crane”. At 20 m above the surface the rover will be lowered on a bridle as it continues to descend. At touch-down, a minimum of 6.5 m of vertical separation are provided between the engines nozzle exit plane and the ground-surface below [1]. This maneuver was chosen in part to minimize the ground/soil interaction that occurs when rocket engine plumes are fired into a soil media. In spite of the 6.5 m altitude above the surface, surface impingement pressures are expected to reach in excess of 2000 Pa, a metric previously established by the Viking program to mitigate soil bearing capacity failure. Plume-ground interaction has been a concern of Lunar and Mars propulsive landings for some time, but was not an issue for the Mars Pathfinder and Mars Explorer Rover era due to their use of airbag landing systems [2][3].This was also a concern of the Phoenix lander program, which fired twelve pulsed hydrazine monopropellant thrusters for its final descent and touch-down [4]. Phoenix was concerned with plume impingement soil interaction due to its high surface impingement pressure and potential for diffused gas eruptions. Phoenix was also concerned with landing site alteration due to its lack of mobility as well as instrument and solar array contamination issues. As MSL will operate in a regime that will result in ground-soil erosion a plume-ground interaction program has been undertaken to quantify the amount of soil erosion, namely the trajectory and number flux of particulates and the contamination and erosion this can impart to sensitive instruments and thermal surface coatings.","PeriodicalId":117250,"journal":{"name":"2009 IEEE Aerospace conference","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126626001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-03-07DOI: 10.1109/AERO.2009.4839647
E. Konieczny, R. Ashcraft, D. Cunningham, S. Maripuri
As Service Oriented Architectures continue to gain prominence as a mechanism to realize standards-based, distributed computing paradigms, the ability for traditional implementations to support bandwidth disadvantaged and runtime composition scenarios has been questioned. Traditional approaches leverage centralized registry platforms to enable service discovery functionality, but this inherently introduces the possibility of stale metadata and registry information that does not reflect actual operating conditions. Service presence offers a fresh opportunity to redefine service discovery; while not typically viewed as a crucial element of SOA's runtime discovery solution space, service presence significantly enhances the real-time monitoring of services by introducing an omnipresent mechanism for capturing a service's state. This paper focuses on analyzing and evaluating the feasibility of utilizing Peer-to-Peer (P2P) approaches to better facilitate service presence and dynamic service discovery through discussion of experimentation conducted using an eXtensible Messaging and Presence Protocol (XMPP) driven prototype.
{"title":"Establishing presence within the service-oriented environment","authors":"E. Konieczny, R. Ashcraft, D. Cunningham, S. Maripuri","doi":"10.1109/AERO.2009.4839647","DOIUrl":"https://doi.org/10.1109/AERO.2009.4839647","url":null,"abstract":"As Service Oriented Architectures continue to gain prominence as a mechanism to realize standards-based, distributed computing paradigms, the ability for traditional implementations to support bandwidth disadvantaged and runtime composition scenarios has been questioned. Traditional approaches leverage centralized registry platforms to enable service discovery functionality, but this inherently introduces the possibility of stale metadata and registry information that does not reflect actual operating conditions. Service presence offers a fresh opportunity to redefine service discovery; while not typically viewed as a crucial element of SOA's runtime discovery solution space, service presence significantly enhances the real-time monitoring of services by introducing an omnipresent mechanism for capturing a service's state. This paper focuses on analyzing and evaluating the feasibility of utilizing Peer-to-Peer (P2P) approaches to better facilitate service presence and dynamic service discovery through discussion of experimentation conducted using an eXtensible Messaging and Presence Protocol (XMPP) driven prototype.","PeriodicalId":117250,"journal":{"name":"2009 IEEE Aerospace conference","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114930699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-03-07DOI: 10.1109/AERO.2009.4839513
B. Lameres, Clint Gauer
This paper presents the design and prototyping of a computing architecture which dynamically reconfigures itself depending on the environment in which it resides. The system switches among three modes of operation (parallel processing, low power, and radiation tolerant) depending on an external radiation sensor and application input from the user. The system was prototyped on a Xilinx Virtex-5 FPGA to verify its feasibility when controlling a series of peripherals under the three modes of operation. This type of system is ideal for robust, real-time applications such as spacecraft control systems.
{"title":"Dynamic reconfigurable computing architecture for aerospace applications","authors":"B. Lameres, Clint Gauer","doi":"10.1109/AERO.2009.4839513","DOIUrl":"https://doi.org/10.1109/AERO.2009.4839513","url":null,"abstract":"This paper presents the design and prototyping of a computing architecture which dynamically reconfigures itself depending on the environment in which it resides. The system switches among three modes of operation (parallel processing, low power, and radiation tolerant) depending on an external radiation sensor and application input from the user. The system was prototyped on a Xilinx Virtex-5 FPGA to verify its feasibility when controlling a series of peripherals under the three modes of operation. This type of system is ideal for robust, real-time applications such as spacecraft control systems.","PeriodicalId":117250,"journal":{"name":"2009 IEEE Aerospace conference","volume":"61 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115097835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-03-07DOI: 10.1109/AERO.2009.4839628
P. von Allmen, Seungwon Lee, L. Kamp, S. Gulkis
We have implemented and validated a model for the radiative transfer in molecular clouds and studied the water emission line for Comet 67P/Churyumov-Gerasimenko. Results are reported that show that the spectral profiles for water emission lines depend on the physical properties of the coma. This tool will be used to interpret observations by MIRO and will help build hydrodynamics models of comets.
{"title":"Molecular excitation and radiative transfer model for MIRO","authors":"P. von Allmen, Seungwon Lee, L. Kamp, S. Gulkis","doi":"10.1109/AERO.2009.4839628","DOIUrl":"https://doi.org/10.1109/AERO.2009.4839628","url":null,"abstract":"We have implemented and validated a model for the radiative transfer in molecular clouds and studied the water emission line for Comet 67P/Churyumov-Gerasimenko. Results are reported that show that the spectral profiles for water emission lines depend on the physical properties of the coma. This tool will be used to interpret observations by MIRO and will help build hydrodynamics models of comets.","PeriodicalId":117250,"journal":{"name":"2009 IEEE Aerospace conference","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122449068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-03-07DOI: 10.1109/AERO.2009.4839671
P. Ballal, A. Ramani, Matthew B. Middleton, Christopher D. McMurrough, A. Athamneh, Weijen Lee, C. Kwan, F. Lewis
This paper has three contributions. First, we develop a low-cost test-bed for simulating bearing faults in a motor. In Aerospace applications, it is important that motor fault signatures are identified before a failure occurs. It is known that 40% of mechanical failures occur due to bearing faults. Bearing faults can be identified from the motor vibration signatures. Second, we develop a wireless sensor module for collection of vibration data from the test-bed. Wireless sensors have been used because of their advantages over wired sensors in remote sensing. Finally, we use a novel two-stage neural network to classify various bearing faults. The first stage neural network estimates the principal components using the Generalized Hebbian Algorithm (GHA). Principal Component Analysis is used to reduce the dimensionality of the data and to extract the fault features. The second stage neural network uses a supervised learning vector quantization network (SLVQ) utilizing a self organizing map approach. This stage is used to classify various fault modes. Neural networks have been used because of their flexibility in terms of online adaptive reformulation. At the end, we discuss the performance of the proposed classification method.
{"title":"Mechanical fault diagnosis using wireless sensor networks and a two-stage neural network classifier","authors":"P. Ballal, A. Ramani, Matthew B. Middleton, Christopher D. McMurrough, A. Athamneh, Weijen Lee, C. Kwan, F. Lewis","doi":"10.1109/AERO.2009.4839671","DOIUrl":"https://doi.org/10.1109/AERO.2009.4839671","url":null,"abstract":"This paper has three contributions. First, we develop a low-cost test-bed for simulating bearing faults in a motor. In Aerospace applications, it is important that motor fault signatures are identified before a failure occurs. It is known that 40% of mechanical failures occur due to bearing faults. Bearing faults can be identified from the motor vibration signatures. Second, we develop a wireless sensor module for collection of vibration data from the test-bed. Wireless sensors have been used because of their advantages over wired sensors in remote sensing. Finally, we use a novel two-stage neural network to classify various bearing faults. The first stage neural network estimates the principal components using the Generalized Hebbian Algorithm (GHA). Principal Component Analysis is used to reduce the dimensionality of the data and to extract the fault features. The second stage neural network uses a supervised learning vector quantization network (SLVQ) utilizing a self organizing map approach. This stage is used to classify various fault modes. Neural networks have been used because of their flexibility in terms of online adaptive reformulation. At the end, we discuss the performance of the proposed classification method.","PeriodicalId":117250,"journal":{"name":"2009 IEEE Aerospace conference","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122759629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-03-07DOI: 10.1109/AERO.2009.4839438
N. Karafolas, J. Armengol, I. Mckenzie
Photonic Technologies in the form of fiber optics, integrated optics and micro-photonics offer some compelling advantages when considered for use in spacecraft. Since 2002 the European Space Agency has engaged in a comprehensive Research and Development program in Photonics that covers applications in communications, sensing, signal processing as well as in some specialized applications. The Research and Development program has been accompanied with the first in-flight demonstrations as well as with the first operational use of fiber optics as critical element of a satellite payload. Fiber optic digital communications for all types of data rates is the first application of Photonics that will reach space qualification. Analog signal communication will follow together with fiber optic sensing. Signal processing applications including Rf down-conversion, switching and analog to digital conversion with electro-photonic means are also under development and their potentials remain to be assessed in comparison with the evolving electronic approaches.
{"title":"Introducing photonics in spacecraft engineering: ESA's strategic approach","authors":"N. Karafolas, J. Armengol, I. Mckenzie","doi":"10.1109/AERO.2009.4839438","DOIUrl":"https://doi.org/10.1109/AERO.2009.4839438","url":null,"abstract":"Photonic Technologies in the form of fiber optics, integrated optics and micro-photonics offer some compelling advantages when considered for use in spacecraft. Since 2002 the European Space Agency has engaged in a comprehensive Research and Development program in Photonics that covers applications in communications, sensing, signal processing as well as in some specialized applications. The Research and Development program has been accompanied with the first in-flight demonstrations as well as with the first operational use of fiber optics as critical element of a satellite payload. Fiber optic digital communications for all types of data rates is the first application of Photonics that will reach space qualification. Analog signal communication will follow together with fiber optic sensing. Signal processing applications including Rf down-conversion, switching and analog to digital conversion with electro-photonic means are also under development and their potentials remain to be assessed in comparison with the evolving electronic approaches.","PeriodicalId":117250,"journal":{"name":"2009 IEEE Aerospace conference","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123001388","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-03-07DOI: 10.1109/AERO.2009.4839450
S. Jordan
Range estimation algorithms have been applied to simulated 3-D flash laser radar data to test for accuracy and bias. Simulated data is modeled after hit mode performance of the Advanced Scientific Concepts 3-D flash laser radar camera. Hit mode is a mode of operation that stores buffered samples into memory only after a set number of photoelectrons have been observed by the detector. In hit mode, waveforms may not be centered within the range gate and may not contain the true peak of original waveform. Under these conditions traditional range estimation techniques could prove ineffective. A peak estimator, matched filter, and maximum likelihood estimator were tested for performance as waveforms shift position within the range gate. This paper suggests the best scenario for implementing each algorithm and shows the overall effectiveness of the matched filter when incorporated in the time domain. 1000 trials with noise were conducted for each waveform position and performance was judged based on mean square error and standard deviation of the range estimations.
{"title":"Range estimation algorithms comparison in simulated 3-D flash LADAR data","authors":"S. Jordan","doi":"10.1109/AERO.2009.4839450","DOIUrl":"https://doi.org/10.1109/AERO.2009.4839450","url":null,"abstract":"Range estimation algorithms have been applied to simulated 3-D flash laser radar data to test for accuracy and bias. Simulated data is modeled after hit mode performance of the Advanced Scientific Concepts 3-D flash laser radar camera. Hit mode is a mode of operation that stores buffered samples into memory only after a set number of photoelectrons have been observed by the detector. In hit mode, waveforms may not be centered within the range gate and may not contain the true peak of original waveform. Under these conditions traditional range estimation techniques could prove ineffective. A peak estimator, matched filter, and maximum likelihood estimator were tested for performance as waveforms shift position within the range gate. This paper suggests the best scenario for implementing each algorithm and shows the overall effectiveness of the matched filter when incorporated in the time domain. 1000 trials with noise were conducted for each waveform position and performance was judged based on mean square error and standard deviation of the range estimations.","PeriodicalId":117250,"journal":{"name":"2009 IEEE Aerospace conference","volume":"316 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121900413","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-03-07DOI: 10.1109/AERO.2009.4839666
A. Saxena, J. Celaya, B. Saha, S. Saha, K. Goebel
Prognostics has taken center stage in Condition Based Maintenance (CBM) where it is desired to estimate Remaining Useful Life (RUL) of a system so that remedial measures may be taken in advance to avoid catastrophic events or unwanted downtimes. Validation of such predictions is an important but difficult proposition and a lack of appropriate evaluation methods renders prognostics meaningless. Evaluation methods currently used in the research community are not standardized and in many cases do not sufficiently assess key performance aspects expected out of a prognostics algorithm. In this paper we introduce several new evaluation metrics tailored for prognostics and show that they can effectively evaluate various algorithms as compared to other conventional metrics. Four prognostic algorithms, Relevance Vector Machine (RVM), Gaussian Process Regression (GPR), Artificial Neural Network (ANN), and Polynomial Regression (PR), are compared. These algorithms vary in complexity and their ability to manage uncertainty around predicted estimates. Results show that the new metrics rank these algorithms in a different manner; depending on the requirements and constraints suitable metrics may be chosen. Beyond these results, this paper offers ideas about how metrics suitable to prognostics may be designed so that the evaluation procedure can be standardized.
{"title":"Evaluating algorithm performance metrics tailored for prognostics","authors":"A. Saxena, J. Celaya, B. Saha, S. Saha, K. Goebel","doi":"10.1109/AERO.2009.4839666","DOIUrl":"https://doi.org/10.1109/AERO.2009.4839666","url":null,"abstract":"Prognostics has taken center stage in Condition Based Maintenance (CBM) where it is desired to estimate Remaining Useful Life (RUL) of a system so that remedial measures may be taken in advance to avoid catastrophic events or unwanted downtimes. Validation of such predictions is an important but difficult proposition and a lack of appropriate evaluation methods renders prognostics meaningless. Evaluation methods currently used in the research community are not standardized and in many cases do not sufficiently assess key performance aspects expected out of a prognostics algorithm. In this paper we introduce several new evaluation metrics tailored for prognostics and show that they can effectively evaluate various algorithms as compared to other conventional metrics. Four prognostic algorithms, Relevance Vector Machine (RVM), Gaussian Process Regression (GPR), Artificial Neural Network (ANN), and Polynomial Regression (PR), are compared. These algorithms vary in complexity and their ability to manage uncertainty around predicted estimates. Results show that the new metrics rank these algorithms in a different manner; depending on the requirements and constraints suitable metrics may be chosen. Beyond these results, this paper offers ideas about how metrics suitable to prognostics may be designed so that the evaluation procedure can be standardized.","PeriodicalId":117250,"journal":{"name":"2009 IEEE Aerospace conference","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121901588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：