Pub Date : 2023-03-04DOI: 10.1109/AERO55745.2023.10115596
A. Utter, Mark Kubiak, E. Grayver
There is growing interest in optical communications for satellite crosslinks. These crosslinks can be simultaneously used for data exchange, time synchronization, and even position determination. Use cases range from dense constellations of small satellites flying in formation to deep space links. It is difficult to emulate such channels under laboratory conditions because the range between any two satellites is both large and constantly changing. Channel emulators for cis-lunar space, for example, must provide delays for link ranges up ~1.5 light-seconds that change by many kilometers per second. Both requirements far exceed the capabilities of off-the-shelf solutions. This paper describes an FPGA-based binary channel emulator that applies a dynamic delay and supports a variety of noncoherent free-space optical (FSO) waveforms. The delay is applied at the physical layer, in contrast with packet-level delays implemented by network emulators. We present two approaches: one based on blind oversampling designed to work with any high-rate transceiver, and another using specific features of the transceivers in Xilinx FPGAs to allow delay adjustment in single-picosecond increments. A detailed implementation is described, addressing issues from initial digitization, accurate delay calibration, and dealing with enormous memory bandwidth. The delay emulator is entirely electronic but can be integrated with other equipment for end-to-end optical testing. Using Ethernet (SGMII) as a placeholder for the over-the-air waveform, the prototype demonstrates relay of gigabit user traffic interleaved with Precision Time Protocol (PTP) messages that are used to measure the channel delay in real-time. Future efforts will include support for coherent optical waveforms.
{"title":"Dynamic Binary Channel Delay Emulation with Picosecond-Scale Precision","authors":"A. Utter, Mark Kubiak, E. Grayver","doi":"10.1109/AERO55745.2023.10115596","DOIUrl":"https://doi.org/10.1109/AERO55745.2023.10115596","url":null,"abstract":"There is growing interest in optical communications for satellite crosslinks. These crosslinks can be simultaneously used for data exchange, time synchronization, and even position determination. Use cases range from dense constellations of small satellites flying in formation to deep space links. It is difficult to emulate such channels under laboratory conditions because the range between any two satellites is both large and constantly changing. Channel emulators for cis-lunar space, for example, must provide delays for link ranges up ~1.5 light-seconds that change by many kilometers per second. Both requirements far exceed the capabilities of off-the-shelf solutions. This paper describes an FPGA-based binary channel emulator that applies a dynamic delay and supports a variety of noncoherent free-space optical (FSO) waveforms. The delay is applied at the physical layer, in contrast with packet-level delays implemented by network emulators. We present two approaches: one based on blind oversampling designed to work with any high-rate transceiver, and another using specific features of the transceivers in Xilinx FPGAs to allow delay adjustment in single-picosecond increments. A detailed implementation is described, addressing issues from initial digitization, accurate delay calibration, and dealing with enormous memory bandwidth. The delay emulator is entirely electronic but can be integrated with other equipment for end-to-end optical testing. Using Ethernet (SGMII) as a placeholder for the over-the-air waveform, the prototype demonstrates relay of gigabit user traffic interleaved with Precision Time Protocol (PTP) messages that are used to measure the channel delay in real-time. Future efforts will include support for coherent optical waveforms.","PeriodicalId":344285,"journal":{"name":"2023 IEEE Aerospace Conference","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124169541","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 : 2023-03-04DOI: 10.1109/AERO55745.2023.10115732
Raunak Srivastava, Rolif Lima, Roshan Sah, K. Das
Use of autonomous space robots show promising potential for precise in-orbit proximity operations like in-orbit servicing and debris capture. However, manipulators mounted on board a satellite present a highly complex and nonlinear dynamic system, which is hence difficult to control for precise in-orbit tasks. We had, in our previous work, presented a Non-linear Model Predictive Controller (NMPC) for Free Floating and Rotation Floating space robots in order to design an optimal path that the end-effector can follow while being controlled to reach the target. However, the MPC optimization problem has to be solved online with the requirement of obtaining the solution within the specified loop rate for a stable performance. Due to the high computational time taken by the MPC's optimization routine, the update frequency of MPC becomes a limiting factor when deployed even on moderately complex hardware systems. This led us to modify the existing controller and use a parameterized Neural Network based controller which learns the optimal policy from the MPC solution. Accordingly, in this work, we solve the optimal control problem via Iterative Linear Quadratic Regulator (iLQR) and use it as means to train a Neural Network (NN) policy online. The final control value for the space robot is hence a weighted combination of the control efforts obtained from the iLQR and NN policy. The accuracy of the proposed modification to a conventional Model Predictive controller and its ability to perform the control objective is demonstrated.
{"title":"In-Orbit Control of Floating Space Robots using a Model Dependant Learning Based Methodology","authors":"Raunak Srivastava, Rolif Lima, Roshan Sah, K. Das","doi":"10.1109/AERO55745.2023.10115732","DOIUrl":"https://doi.org/10.1109/AERO55745.2023.10115732","url":null,"abstract":"Use of autonomous space robots show promising potential for precise in-orbit proximity operations like in-orbit servicing and debris capture. However, manipulators mounted on board a satellite present a highly complex and nonlinear dynamic system, which is hence difficult to control for precise in-orbit tasks. We had, in our previous work, presented a Non-linear Model Predictive Controller (NMPC) for Free Floating and Rotation Floating space robots in order to design an optimal path that the end-effector can follow while being controlled to reach the target. However, the MPC optimization problem has to be solved online with the requirement of obtaining the solution within the specified loop rate for a stable performance. Due to the high computational time taken by the MPC's optimization routine, the update frequency of MPC becomes a limiting factor when deployed even on moderately complex hardware systems. This led us to modify the existing controller and use a parameterized Neural Network based controller which learns the optimal policy from the MPC solution. Accordingly, in this work, we solve the optimal control problem via Iterative Linear Quadratic Regulator (iLQR) and use it as means to train a Neural Network (NN) policy online. The final control value for the space robot is hence a weighted combination of the control efforts obtained from the iLQR and NN policy. The accuracy of the proposed modification to a conventional Model Predictive controller and its ability to perform the control objective is demonstrated.","PeriodicalId":344285,"journal":{"name":"2023 IEEE Aerospace Conference","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125189829","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 : 2023-03-04DOI: 10.1109/AERO55745.2023.10115656
Annmarie Spexet, Jessica LaRocco-Olszewski, D. Alvord
In the aviation space, maintenance is the main driver in the push for Internet of Things (loT) device management systems, artificial intelligence (AI)/machine learning (ML) re-search, and cloud infrastructure. The potential for this approach to reduce downtime, maximize component lifetime, re-duce man-hours on diagnosis and repair, and optimize supply chains and scheduling has driven massive investments across the industry. And yet, the challenges in delivering on these promises with the available data and technology should also not be minimized. To reach its full potential, maintenance program implementers must understand what predictions can be derived from the available data, what maintenance actions may be driven by those predictions, and how the predictions should be presented to the appropriate decision makers in ground operations and the logistics chain. This report examines the current state of data within the aviation maintenance space, variations in component level coverage, and how that translates to the type, volume, and timeliness of data and computational infrastructure necessary to provide right time predictions and analytics to maintainers, supply chain managers, and operators. This report also addresses some of the specific challenges in the aviation space with respect to data availability, equipment variability, use variability, and maintenance action coding that can affect the ability of operators to derive value from a data science program.
{"title":"Data Pipeline Considerations for Aviation Maintenance","authors":"Annmarie Spexet, Jessica LaRocco-Olszewski, D. Alvord","doi":"10.1109/AERO55745.2023.10115656","DOIUrl":"https://doi.org/10.1109/AERO55745.2023.10115656","url":null,"abstract":"In the aviation space, maintenance is the main driver in the push for Internet of Things (loT) device management systems, artificial intelligence (AI)/machine learning (ML) re-search, and cloud infrastructure. The potential for this approach to reduce downtime, maximize component lifetime, re-duce man-hours on diagnosis and repair, and optimize supply chains and scheduling has driven massive investments across the industry. And yet, the challenges in delivering on these promises with the available data and technology should also not be minimized. To reach its full potential, maintenance program implementers must understand what predictions can be derived from the available data, what maintenance actions may be driven by those predictions, and how the predictions should be presented to the appropriate decision makers in ground operations and the logistics chain. This report examines the current state of data within the aviation maintenance space, variations in component level coverage, and how that translates to the type, volume, and timeliness of data and computational infrastructure necessary to provide right time predictions and analytics to maintainers, supply chain managers, and operators. This report also addresses some of the specific challenges in the aviation space with respect to data availability, equipment variability, use variability, and maintenance action coding that can affect the ability of operators to derive value from a data science program.","PeriodicalId":344285,"journal":{"name":"2023 IEEE Aerospace Conference","volume":"105 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125389861","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 : 2023-03-04DOI: 10.1109/AERO55745.2023.10115742
Aaron B. Hoskins, R. Alvarez
Satellites are a valuable resource in monitoring the Earth for scientific and military tasks. However, the initial orbital parameters determine the ground track of the satellite and, thus which ground locations can be imaged. For a mission designer, selecting the orbital parameters that will maximize the collection of the desired data is imperative. This work investigates the optimal selection of initial orbital parameters for a satellite to monitor a user-supplied list of prioritized ground locations. The ground locations' priorities decrease for subsequent images of a location as a means of encouraging image diversity and prioritizing more valuable locations. The objective function is the summation of the prioritized images collected. The dynamics of the problem are simulated using General Mission Analysis Tool (GMAT). Using GMAT, a robust framework is created where the dynamics can be easily altered to include (or disregard) any perturbation forces; it is also possible to easily include constraints such as lighting or topography that could prevent an image from being collected. The optimization problem is solved using Coyote Optimization Algorithm (COA). COA is a relatively new metaheuristic with promising potential, and it is compared to the more traditional metaheuristic Particle Swarm Optimization (PSO). The results show that COA performs better than PSO in terms of computational time while finding virtually identical initial orbital parameters. The two primary benefits of this work are the creation of a robust framework for initial orbital parameters for a list of user-supplied prioritized ground locations and introducing COA to this class of problems.
{"title":"Initial Orbit Selection for Prioritized Ground Targets Using Coyote Optimization Algorithm","authors":"Aaron B. Hoskins, R. Alvarez","doi":"10.1109/AERO55745.2023.10115742","DOIUrl":"https://doi.org/10.1109/AERO55745.2023.10115742","url":null,"abstract":"Satellites are a valuable resource in monitoring the Earth for scientific and military tasks. However, the initial orbital parameters determine the ground track of the satellite and, thus which ground locations can be imaged. For a mission designer, selecting the orbital parameters that will maximize the collection of the desired data is imperative. This work investigates the optimal selection of initial orbital parameters for a satellite to monitor a user-supplied list of prioritized ground locations. The ground locations' priorities decrease for subsequent images of a location as a means of encouraging image diversity and prioritizing more valuable locations. The objective function is the summation of the prioritized images collected. The dynamics of the problem are simulated using General Mission Analysis Tool (GMAT). Using GMAT, a robust framework is created where the dynamics can be easily altered to include (or disregard) any perturbation forces; it is also possible to easily include constraints such as lighting or topography that could prevent an image from being collected. The optimization problem is solved using Coyote Optimization Algorithm (COA). COA is a relatively new metaheuristic with promising potential, and it is compared to the more traditional metaheuristic Particle Swarm Optimization (PSO). The results show that COA performs better than PSO in terms of computational time while finding virtually identical initial orbital parameters. The two primary benefits of this work are the creation of a robust framework for initial orbital parameters for a list of user-supplied prioritized ground locations and introducing COA to this class of problems.","PeriodicalId":344285,"journal":{"name":"2023 IEEE Aerospace Conference","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126212497","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 : 2023-03-04DOI: 10.1109/AERO55745.2023.10115940
C. Whiting
The Next Generation Radioisotope Thermoelectric Generator (Next Gen RTG) is a project aimed at restarting production of the high performing General Purpose Heat Source (GPHS) RTG. Unfortunately, the performance of most GPHS-RTG missions is not expected to be representative of the Next Gen RTG. Heat sources used by the U.S. today are the slightly larger Step-2 GPHS module. This will reduce the amount of $^{238}mathbf{PuO_{2}}$ fuel that can fit within the RTG down to $boldsymbol{sim 3900mathrm{W}_{text{th}} (text{from}sim 4400 mathrm{W}_{text{th}})}$. This reduced thermal inventory will lower RTG temperatures, which will reduce performance. New Horizons is a current GPHS-RTG mission that was loaded with 3948 $mathbf{W}_{mathbf{th}}$ of $^{238}mathbf{PuO_{2}}$, suggesting that New Horizons performance may be similar to the future Next Gen RTG. The rate law analysis method has been successfully used to convert RTG telemetry into equations that describe behavior. A rate law analysis of New Horizons was used to determine the degree of thermoelectric degradation and thermal inventory losses. This analysis showed that the thermoelectric degradation mechanism on New Horizons changed after 8.0 years, and that this change in mechanism is likely tied to the thermal inventory of the fuel at that time. These equations were then used to produce a preliminary estimate the performance of the Next Gen RTG. Based on this estimate, Next Gen RTG will produce $boldsymbol{184 mathrm{W}_{e}}$ at the end-of-design-life (EODL) if the unit is not stored before launch, and $boldsymbol{177 mathrm{W}_{mathrm{e}}}$ if the unit is stored for 3 years before launch. Performance curves were also produced. It is important to realize that this preliminary estimate is based on a lot of assumptions, including: how the Step-2 module will affect performance, mission specific parameters, and potential changes to the Next Gen RTG as the design evolves. A discussion of these assumptions is provided. It may be necessary to update this estimate as additional Next Gen RTG design information becomes public.
{"title":"Estimating the Next Gen RTG Mod 1 Performance Based on Analysis of the New Horizons Mission","authors":"C. Whiting","doi":"10.1109/AERO55745.2023.10115940","DOIUrl":"https://doi.org/10.1109/AERO55745.2023.10115940","url":null,"abstract":"The Next Generation Radioisotope Thermoelectric Generator (Next Gen RTG) is a project aimed at restarting production of the high performing General Purpose Heat Source (GPHS) RTG. Unfortunately, the performance of most GPHS-RTG missions is not expected to be representative of the Next Gen RTG. Heat sources used by the U.S. today are the slightly larger Step-2 GPHS module. This will reduce the amount of $^{238}mathbf{PuO_{2}}$ fuel that can fit within the RTG down to $boldsymbol{sim 3900mathrm{W}_{text{th}} (text{from}sim 4400 mathrm{W}_{text{th}})}$. This reduced thermal inventory will lower RTG temperatures, which will reduce performance. New Horizons is a current GPHS-RTG mission that was loaded with 3948 $mathbf{W}_{mathbf{th}}$ of $^{238}mathbf{PuO_{2}}$, suggesting that New Horizons performance may be similar to the future Next Gen RTG. The rate law analysis method has been successfully used to convert RTG telemetry into equations that describe behavior. A rate law analysis of New Horizons was used to determine the degree of thermoelectric degradation and thermal inventory losses. This analysis showed that the thermoelectric degradation mechanism on New Horizons changed after 8.0 years, and that this change in mechanism is likely tied to the thermal inventory of the fuel at that time. These equations were then used to produce a preliminary estimate the performance of the Next Gen RTG. Based on this estimate, Next Gen RTG will produce $boldsymbol{184 mathrm{W}_{e}}$ at the end-of-design-life (EODL) if the unit is not stored before launch, and $boldsymbol{177 mathrm{W}_{mathrm{e}}}$ if the unit is stored for 3 years before launch. Performance curves were also produced. It is important to realize that this preliminary estimate is based on a lot of assumptions, including: how the Step-2 module will affect performance, mission specific parameters, and potential changes to the Next Gen RTG as the design evolves. A discussion of these assumptions is provided. It may be necessary to update this estimate as additional Next Gen RTG design information becomes public.","PeriodicalId":344285,"journal":{"name":"2023 IEEE Aerospace Conference","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125709887","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 : 2023-03-04DOI: 10.1109/AERO55745.2023.10116022
Kalani R. Danas Rivera, D. Sternberg, Kevin Lo, S. Mohan
The Small Satellite Dynamics Testbed (SSDT) at the NASA Jet Propulsion Lab (JPL) is a facility dedicated to the verification and validation of small satellite guidance, navigation, and control (GNC) hardware and software. The combination of the SSDT with NASA JPL's Formation Control Testbed (FCT) provides a testbed for close-proximity operations between small satellites and larger, traditional spacecraft. Previously, the SSDT housed a single thrusting air-bearing platform that affords position control on a planar surface and rotational control about the axis perpendicular to the translation plane. This work outlines the design modifications to the previous air-bearing platform and implements the design to two planar air-bearing platforms. The extension to two planar air-bearing platforms enables testing of close-proximity operations between two small satellites. This work on the SSDT platforms may be combined with the capabilities of the FCT's larger thrusting planar air bearings to produce a multi-platform testbed for close proximity operations between multiple small satellites and multiple traditional satellites. Consequently, this work fundamentally transforms the capabilities of the SSDT to enable multi-platform research and development.
{"title":"Multi-Platform Small Satellite Dynamics Testbed","authors":"Kalani R. Danas Rivera, D. Sternberg, Kevin Lo, S. Mohan","doi":"10.1109/AERO55745.2023.10116022","DOIUrl":"https://doi.org/10.1109/AERO55745.2023.10116022","url":null,"abstract":"The Small Satellite Dynamics Testbed (SSDT) at the NASA Jet Propulsion Lab (JPL) is a facility dedicated to the verification and validation of small satellite guidance, navigation, and control (GNC) hardware and software. The combination of the SSDT with NASA JPL's Formation Control Testbed (FCT) provides a testbed for close-proximity operations between small satellites and larger, traditional spacecraft. Previously, the SSDT housed a single thrusting air-bearing platform that affords position control on a planar surface and rotational control about the axis perpendicular to the translation plane. This work outlines the design modifications to the previous air-bearing platform and implements the design to two planar air-bearing platforms. The extension to two planar air-bearing platforms enables testing of close-proximity operations between two small satellites. This work on the SSDT platforms may be combined with the capabilities of the FCT's larger thrusting planar air bearings to produce a multi-platform testbed for close proximity operations between multiple small satellites and multiple traditional satellites. Consequently, this work fundamentally transforms the capabilities of the SSDT to enable multi-platform research and development.","PeriodicalId":344285,"journal":{"name":"2023 IEEE Aerospace Conference","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130115284","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 : 2023-03-04DOI: 10.1109/AERO55745.2023.10115584
Sixiao Wei, Li Li, Genshe Chen, Erik Blasch, K. Chang, T. Clemons, K. Pham
Current radar-based Air Traffic Service (ATS) providers lack the preservation of privacy for airspace operations of selected flight plans, positions, and state data; requiring security assurance. Recent events have shown that modern unmanned aerial vehicles (UAVs) are vulnerable to attack and subversion through software flaws or sometimes malicious devices that are present on urban air mobility (UAM) communication networks, which increases the need for cyber awareness including the risk of cyber intrusion. Many stealthy attacks (such as False Data Injection Attacks (FDIAs)) are hard to detect on avionics systems, as they can compromise measurements from sensors and bypass the sensor's basic “faulty data” detection mechanism and remain undetected. Such attacks on a UAM system may not even present their impact but propagate from the sensor to fool the system by predicting a delayed asset failure or maintenance interval. In this paper, we develop a Resilience Oriented Security Inspection System (ROSIS) to maximize UAM capability to secure data accessing and sharing among aircraft and Air Traffic Service (ATS) service providers. Specifically, we collect and demonstrate the effect of generalized FDIAs on wireless sensors of a turbofan engine using NASA's C-MAPSS simulator, and develop data-driven based deep learning methods (Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU)) for detecting abnormal features. A graphical physics-informed Bayesian Network model is developed to represent the dynamic nature of the engine to predict health conditions accordingly. The ROSIS model characterizes the condition-symptom relationships of different engine components and sensors. A hybrid software-in-the-loop (SITL) and hardware-in-the-loop (HITL) design is also developed to evaluate the effectiveness of the ROSIS defense mechanisms. Our experiments validate the performance of ROSIS in detection accuracy and efficiency against FDIAs.
{"title":"ROSIS: Resilience Oriented Security Inspection System against False Data Injection Attacks","authors":"Sixiao Wei, Li Li, Genshe Chen, Erik Blasch, K. Chang, T. Clemons, K. Pham","doi":"10.1109/AERO55745.2023.10115584","DOIUrl":"https://doi.org/10.1109/AERO55745.2023.10115584","url":null,"abstract":"Current radar-based Air Traffic Service (ATS) providers lack the preservation of privacy for airspace operations of selected flight plans, positions, and state data; requiring security assurance. Recent events have shown that modern unmanned aerial vehicles (UAVs) are vulnerable to attack and subversion through software flaws or sometimes malicious devices that are present on urban air mobility (UAM) communication networks, which increases the need for cyber awareness including the risk of cyber intrusion. Many stealthy attacks (such as False Data Injection Attacks (FDIAs)) are hard to detect on avionics systems, as they can compromise measurements from sensors and bypass the sensor's basic “faulty data” detection mechanism and remain undetected. Such attacks on a UAM system may not even present their impact but propagate from the sensor to fool the system by predicting a delayed asset failure or maintenance interval. In this paper, we develop a Resilience Oriented Security Inspection System (ROSIS) to maximize UAM capability to secure data accessing and sharing among aircraft and Air Traffic Service (ATS) service providers. Specifically, we collect and demonstrate the effect of generalized FDIAs on wireless sensors of a turbofan engine using NASA's C-MAPSS simulator, and develop data-driven based deep learning methods (Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU)) for detecting abnormal features. A graphical physics-informed Bayesian Network model is developed to represent the dynamic nature of the engine to predict health conditions accordingly. The ROSIS model characterizes the condition-symptom relationships of different engine components and sensors. A hybrid software-in-the-loop (SITL) and hardware-in-the-loop (HITL) design is also developed to evaluate the effectiveness of the ROSIS defense mechanisms. Our experiments validate the performance of ROSIS in detection accuracy and efficiency against FDIAs.","PeriodicalId":344285,"journal":{"name":"2023 IEEE Aerospace Conference","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129659954","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 : 2023-03-04DOI: 10.1109/AERO55745.2023.10115916
Andrew Harris, Kedar R. Naik
Autonomy is a highly sought-after feature for future civil, commercial, and military Earth-observation space missions. Deep Reinforcement Learning (DRL) techniques have demonstrated state-of-the-art performance for on-line decision making in complex domains such as competitive real-time strat-egy games and black-box control. DRL allows for the direct uti-lization of high-fidelity whole-system simulators without inter-mediate approximations or representations. As a result, agents trained with DRL can explore and exploit subtle dynamics that may be lost when applying approximation-driven techniques for tasking and planning, thereby enabling greater performance. This work describes efforts at Ball Aerospace in developing a DRL-driven solution to the single-satellite, arbitrary-target, single-ground-station planning problem. A design reference mission for this problem, based on the Compact Infrared Ra-diometer in Space (CIRiS) cubesat, is described alongside an im-plementation of this mission using the Basilisk simulation frame-work. The resulting simulator is used as a training environment for the DRL-based agent. This simulator is also used for performance evaluation. The DRL-based agent's performance is compared against the results of a rule-based agent. The results of these approaches are compared using multiple figures of merit, including objective performance, decision-making time, and mission-level resource utilization. The resulting comparison demonstrates the relative merits of DRL as an approach versus heuristic, rule-based command and control architectures.
{"title":"Autonomous Command and Control for Earth-Observing Satellites using Deep Reinforcement Learning","authors":"Andrew Harris, Kedar R. Naik","doi":"10.1109/AERO55745.2023.10115916","DOIUrl":"https://doi.org/10.1109/AERO55745.2023.10115916","url":null,"abstract":"Autonomy is a highly sought-after feature for future civil, commercial, and military Earth-observation space missions. Deep Reinforcement Learning (DRL) techniques have demonstrated state-of-the-art performance for on-line decision making in complex domains such as competitive real-time strat-egy games and black-box control. DRL allows for the direct uti-lization of high-fidelity whole-system simulators without inter-mediate approximations or representations. As a result, agents trained with DRL can explore and exploit subtle dynamics that may be lost when applying approximation-driven techniques for tasking and planning, thereby enabling greater performance. This work describes efforts at Ball Aerospace in developing a DRL-driven solution to the single-satellite, arbitrary-target, single-ground-station planning problem. A design reference mission for this problem, based on the Compact Infrared Ra-diometer in Space (CIRiS) cubesat, is described alongside an im-plementation of this mission using the Basilisk simulation frame-work. The resulting simulator is used as a training environment for the DRL-based agent. This simulator is also used for performance evaluation. The DRL-based agent's performance is compared against the results of a rule-based agent. The results of these approaches are compared using multiple figures of merit, including objective performance, decision-making time, and mission-level resource utilization. The resulting comparison demonstrates the relative merits of DRL as an approach versus heuristic, rule-based command and control architectures.","PeriodicalId":344285,"journal":{"name":"2023 IEEE Aerospace Conference","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129754011","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 : 2023-03-04DOI: 10.1109/AERO55745.2023.10115730
Daryna Budiakivska, Jakub Fabisiak
Experimental copper beryllium suspension consisting of 4 units (quasi-rocker arms) was designed for the Sirius II rover, created by Students' Space Association at Warsaw University of Technology. The aim was to create a suspension system for multi-purpose wheeled robot, that will be easy to manufacture, assemble and maintain, have no kinematic pairs, and fulfil the goal of absorbing shocks and vibrations, while traversing through rough and uneven terrain. Copper beryllium alloys have high fatigue strength, excellent wear and corrosion resistance, can operate in wide range of temperatures, from low cryogenic (-200°C) up to 250°C. The 3D model was created and then numerically analysed with the usage of Ansys Mechanical, with the static load based on the rover geometry and masses, collisions with different obstacles, dynamical load cases with eigenfrequencies values and many more. Based on the consecutive analysis, the design was optimised until the desired strength and deflection parameters were obtained and the safety factor fulfilled our needs ensuring that the structure will withstand standard operating conditions but will also cope with unexpected accidents or falls. Final structure has a look of a spring with 10 folds. Complex spring-like structure and subtle differences in radii on quasi-rocker arm folds resulted in that the part could not be manufactured by most manual methods, so from the beginning the preferred ones were computer controlled. Toxicity of beryllium was another factor considered and dismissed by manual manufacturing. In the end, the 23 mm thick plate of copper beryllium was introduced to the water jet cutter, which proved to be the most fitting method for this design cost- and quality-wise. After manufacturing, the suspension was mounted to the rover. Several tests were performed, among them - riding the plain and rocky terrain, riding up and down steep hills, pulling high masses (100 kg - with 45 kg rover total mass for comparison), collisions and abrupt change of load. The suspension performed perfectly well under every condition, ensuring great traction and stability for the rover without any sign of damage during and after continuous and extensive loads. Final design derived from student competition construction resulted in an innovative solution, which could be adapted in numerous vehicles for both Earth and other planets' environments, ensuring great traversal capabilities combined with relatively low-cost materials and low to no maintenance needs. Absence of kinematic pairs dismiss the need of lubrication and reduces material wear, while high-strength of used material lowers the risk of fatigue failure. Ability to operate in aggressive environments is also present, due to materials resistance to corrosion and chemical substances. Although simple, the proposed idea of design proved to be reliable and provides possibilities of further development to serve well in various wheeled vehicles in multiple types of harsh enviro
{"title":"Experimental Suspension for Wheeled Vehicles and Robots Aimed for Harsh and Off-Earth Environments","authors":"Daryna Budiakivska, Jakub Fabisiak","doi":"10.1109/AERO55745.2023.10115730","DOIUrl":"https://doi.org/10.1109/AERO55745.2023.10115730","url":null,"abstract":"Experimental copper beryllium suspension consisting of 4 units (quasi-rocker arms) was designed for the Sirius II rover, created by Students' Space Association at Warsaw University of Technology. The aim was to create a suspension system for multi-purpose wheeled robot, that will be easy to manufacture, assemble and maintain, have no kinematic pairs, and fulfil the goal of absorbing shocks and vibrations, while traversing through rough and uneven terrain. Copper beryllium alloys have high fatigue strength, excellent wear and corrosion resistance, can operate in wide range of temperatures, from low cryogenic (-200°C) up to 250°C. The 3D model was created and then numerically analysed with the usage of Ansys Mechanical, with the static load based on the rover geometry and masses, collisions with different obstacles, dynamical load cases with eigenfrequencies values and many more. Based on the consecutive analysis, the design was optimised until the desired strength and deflection parameters were obtained and the safety factor fulfilled our needs ensuring that the structure will withstand standard operating conditions but will also cope with unexpected accidents or falls. Final structure has a look of a spring with 10 folds. Complex spring-like structure and subtle differences in radii on quasi-rocker arm folds resulted in that the part could not be manufactured by most manual methods, so from the beginning the preferred ones were computer controlled. Toxicity of beryllium was another factor considered and dismissed by manual manufacturing. In the end, the 23 mm thick plate of copper beryllium was introduced to the water jet cutter, which proved to be the most fitting method for this design cost- and quality-wise. After manufacturing, the suspension was mounted to the rover. Several tests were performed, among them - riding the plain and rocky terrain, riding up and down steep hills, pulling high masses (100 kg - with 45 kg rover total mass for comparison), collisions and abrupt change of load. The suspension performed perfectly well under every condition, ensuring great traction and stability for the rover without any sign of damage during and after continuous and extensive loads. Final design derived from student competition construction resulted in an innovative solution, which could be adapted in numerous vehicles for both Earth and other planets' environments, ensuring great traversal capabilities combined with relatively low-cost materials and low to no maintenance needs. Absence of kinematic pairs dismiss the need of lubrication and reduces material wear, while high-strength of used material lowers the risk of fatigue failure. Ability to operate in aggressive environments is also present, due to materials resistance to corrosion and chemical substances. Although simple, the proposed idea of design proved to be reliable and provides possibilities of further development to serve well in various wheeled vehicles in multiple types of harsh enviro","PeriodicalId":344285,"journal":{"name":"2023 IEEE Aerospace Conference","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128638751","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 : 2023-03-04DOI: 10.1109/AERO55745.2023.10115959
M. Mottaghi, R. Chhabra, Wei Huang
Six-wheel autonomous rovers with skid-steering rear wheels have been designed for Lunar exploration programs due to their lightweight and their enhanced traction and stability. In this paper, a Software-in-the-loop Simulation (SILS) is presented for such a system containing a controller coded in MAT-LAB and a digital twin of the system modeled in Vortex Studio. The controller is developed based on the system's governing equations and static state feedback linearization to perform an output-tracking control task. The equations of motion are derived using Lagrange-d’ Alembert principle subject to ideal nonholonomic constraints and under the point-contact assumption at all wheels. Such simplifying assumptions are commonly considered in proposed control strategies for autonomous rover systems in the literature. The digital twin of the rover is modeled as a multi-body system with realistic parameters moving on 3-dimensional soft/rough terrains with arbitrary tire models provided by Vortex Studio. The results of the developed SILS are compared to those of a 2-dimensional simulation that is fully coded in MATLAB under the simplifying assumptions (ideal plant). This comparison discloses often existing discrepancies between real rover systems and their commonly used mathematical models. This study reveals the effects of isolated physical phenomena, e.g., wheel-slip and tractive force distribution, on the control performance, and can be utilized to design enhanced controllers for rover systems.
{"title":"High-fidelity Software-in-the-loop Simulation of a Six-wheel Lunar Rover using Vortex Studio for Output-tracking Control Design","authors":"M. Mottaghi, R. Chhabra, Wei Huang","doi":"10.1109/AERO55745.2023.10115959","DOIUrl":"https://doi.org/10.1109/AERO55745.2023.10115959","url":null,"abstract":"Six-wheel autonomous rovers with skid-steering rear wheels have been designed for Lunar exploration programs due to their lightweight and their enhanced traction and stability. In this paper, a Software-in-the-loop Simulation (SILS) is presented for such a system containing a controller coded in MAT-LAB and a digital twin of the system modeled in Vortex Studio. The controller is developed based on the system's governing equations and static state feedback linearization to perform an output-tracking control task. The equations of motion are derived using Lagrange-d’ Alembert principle subject to ideal nonholonomic constraints and under the point-contact assumption at all wheels. Such simplifying assumptions are commonly considered in proposed control strategies for autonomous rover systems in the literature. The digital twin of the rover is modeled as a multi-body system with realistic parameters moving on 3-dimensional soft/rough terrains with arbitrary tire models provided by Vortex Studio. The results of the developed SILS are compared to those of a 2-dimensional simulation that is fully coded in MATLAB under the simplifying assumptions (ideal plant). This comparison discloses often existing discrepancies between real rover systems and their commonly used mathematical models. This study reveals the effects of isolated physical phenomena, e.g., wheel-slip and tractive force distribution, on the control performance, and can be utilized to design enhanced controllers for rover systems.","PeriodicalId":344285,"journal":{"name":"2023 IEEE Aerospace Conference","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130951848","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：