Results for “Autonomous Operation” — Page 15

NASA NTRSDate not supplied

Optical Modem Enabling Broadband Datacom Links for Crewed Cis-Lunar Missions

We report on the design, development, and testing of our high-power broadband optical modem supporting NASA’s crewed Artemis-2 mission. The O2O modem will be mounted in the crewed Orion module and provide a broadband 505,000 km bi-directional optical link back to earth while en route to the moon. The full-duplex modem consists of a high-power optical transmitter and receiver optimized for serially-concatenated pulse-position modulation (SCPPM). The transmitter is a master-oscillator power-amplifier optical architecture using efficient cladding-pumped amplification in erbium-ytterbium co-doped fiber. The transmitter outputs up to 1 W at ≈1550 nm (limited for eye safety) and supports 6 different user-rates ranging from 20.39 Mb/s to 260.95 Mb/s using PPM16 and PPM32 modulation formats. The optical receiver supports two user-rates: 10.19 Mb/s and 20.39 Mb/s with both rates employing PPM32. The narrowband receiver filtering is optimized to simultaneously accept four separate wavelength channels to mitigate atmospherics through spatial diversity. A configurable interleaver provides additional protection against atmospherics-based signal fading and a powerful soft-decision error correction scheme enables highly sensitive detection. The measured sensitivities at the two bit-rates are -73.8 and -71.8dBm, respectively. The architecture was designed for reliable operation in space, featuring automatic hardware interlocks, pump sparing for the amplifiers, and autonomous operation of all internal hardware and software control loops. The protoflight unit (PFU) was put through rigorous environmental testing which included pyroshock, vibration, electromagnetic interference/compatibility, and thermal-vacuum testing. The modem successfully passed all the environmental screening and has been declared at Technology Readiness Level (TRL) 6.

space communications↗

NASA NTRSDate not supplied

Autonomous Navigation, Guidance, and Control Software in a Low SWaP Box

Onboard autonomy is a necessity for responsive space operations. Autonomous navigation, guidance, and control (NGC) enables space missions to reduce their dependence on high demand ground assets and costly ground personnel. It also allows for in-situ decision making and higher return on mission data. A flight software and hardware system providing this capability, called “autoNGC,” is currently being developed at NASA Goddard Space Flight Center for infusion into multiple future missions. The autoNGC flight software is built on the plug-and-play architecture of the core Flight System (cFS) consisting of the standard cFS apps and newly developed autoNGC interface apps and libraries. The various apps cooperate through communication over the message-based software bus. With the plug-and-play architecture of autoNGC, cFS apps can easily be added and replaced to meet the needs of different missions, even after launch. The first flight software release of autoNGC is targeted for Summer 2024 to provide autonomous navigation at the Moon and beyond. It can perform sensor fusion of multiple measurement types including pseudo-range from a Global Navigation Satellite System (GNSS) receiver (including weak signal), 1-way and 2-way range and Doppler from ground stations (i.e., direct to Earth (DTE)), bearing and range from optical camera images, and accelerometer data. Accurate onboard navigation and timing is obtained through the Goddard Enhanced Onboard Navigation System (GEONS) software library which fuses different measurement types through an extended Kalman filter (EKF) framework. Optical measurements that are ingested in GEONS are first extracted from optical images by the cFS Goddard Image Analysis and Navigation Tool (cGIANT) app. If the imaged body is far enough away that it appears as a pixel or cluster of pixels, then bearing angles to the body centroid can be provided. If the body is close enough and the shape is known coarsely, then bearing angles and range to the body centroid can be derived from the limb. Bearing angles to individual surface features can also be extracted (i.e., terrain relative navigation (TRN)). Onboard guidance and control capabilities are being developed for a future release to perform autonomous station-keeping and trajectory correction maneuvers in multiple orbital regimes. Capabilities to enable distributed systems missions and constellations, such as crosslink measurements, and onboard time management are being developed as well. The first hardware implementation of autoNGC is a minimal size, weight, and power (SWaP) design allowing for inclusion into CubeSats and SmallSat-size buses. Advancements in miniaturized space processors, such as the SpaceCube 3.0 Mini and the SpaceCube Mini-Z are utilized for low SWaP while maintaining a high level of performance. The current enclosure design is 12 cm x 17 cm x 13.5 cm. The box mass is expected to be less than 2 kg, and the nominal power is 21 W. In order to accommodate a wide range of missions, the hardware interfaces are designed for flexibility with a variety of sensor inputs. Through comprehensive testing in the software-in-the-loop, processor-in-the-loop, and hardware-in-the-loop test beds that are concurrently being developed, autoNGC is expected to achieve TRL 6 by late 2024.

Sun Hur-Diaz↗

NASA NTRS2008

Video Guidance Sensor for Surface Mobility Operations

Robotic systems and surface mobility will play an increased role in future exploration missions. Unlike the LRV during Apollo era which was an astronaut piloted vehicle future systems will include teleoperated and semi-autonomous operations. The tasks given to these vehicles will run the range from infrastructure maintenance, ISRU, and construction to name a few. A common task that may be performed would be the retrieval and deployment of trailer mounted equipment. Operational scenarios may require these operations to be performed remotely via a teleoperated mode,or semi-autonomously. This presentation describes the on-going project to adapt the Automated Rendezvous and Capture (AR&C) sensor developed at the Marshall Space Flight Center for use in an automated trailer pick-up and deployment operation. The sensor which has been successfully demonstrated on-orbit has been mounted on an iRobot/John Deere RGATOR autonomous vehicle for this demonstration which will be completed in the March 2008 time-frame.

Fernandez, Kenneth R.↗

DOE OSTI2024

Enabling a Physical Twin for Control Methods Evaluation

Advanced nuclear reactors play an important role in the energy future of the United States and the rest of the world. They are designed and operated based on a different model than that of the current operating fleet, thus enabling deployment in remote locations and allowing for safe semi-autonomous or autonomous operations. Such characteristics require the development of a new reactor control paradigm. A significant factor in the development of control technologies and methods is integration of the various technologies and methods with each other and with hardware (both reactor system hardware and control hardware). A recent workshop on control of advanced reactors identified the lack of a flexible, expandable software/hardware infrastructure to enable such integration as a key gap. A previous phase of the current effort involved developing and demonstrating the Control and Optimization Modular Modeling Application for Nuclear Deployment (COMMAND) platform, which is capable of integrating autonomous-control-enabling technologies and methods, without the constraints imposed by existing software solutions. This platform will enable advanced reactor developers to deploy and test any developed technologies and methods by employing a common framework, and to couple them with their own models and hardware. The present phase of this effort entails using the Microreactor Automated Control System (MACS) platform, which was developed by the U.S. Department of Energy (DOE) Microreactor Program, to serve as a control method testbed. MACS can be used by advanced reactor developers to integrate their control related research activities with any reactor system. For the present effort, MACS was customized to mirror Idaho National Laboratory (INL)’s Microreactor Applications Research Validation and Evaluation (MARVEL) microreactor, and COMMAND was leveraged to enable MACS to emulate the physics of MARVEL, thus positioning MACS as a physical twin of MARVEL.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

NASA NTRS2004

Design of a Multi-mode Flight Deck Decision Support System for Airborne Conflict Management

NASA Langley has developed a multi-mode decision support system for pilots operating in a Distributed Air-Ground Traffic Management (DAG-TM) environment. An Autonomous Operations Planner (AOP) assists pilots in performing separation assurance functions, including conflict detection, prevention, and resolution. Ongoing AOP design has been based on a comprehensive human factors analysis and evaluation results from previous human-in-the-loop experiments with airline pilot test subjects. AOP considers complex flight mode interactions and provides flight guidance to pilots consistent with the current aircraft control state. Pilots communicate goals to AOP by setting system preferences and actively probing potential trajectories for conflicts. To minimize training requirements and improve operational use, AOP design leverages existing alerting philosophies, displays, and crew interfaces common on commercial aircraft. Future work will consider trajectory prediction uncertainties, integration with the TCAS collision avoidance system, and will incorporate enhancements based on an upcoming air-ground coordination experiment.

Barhydt, Richard↗

DOE OSTI2023

Integrating Control Methods and Digital Twins for Advanced Nuclear Reactors

Advanced nuclear reactors offer new capabilities such as the ability to adapt to variable energy demand, operate autonomously with remote supervision, be deployable in rural locations, be compact in size, afford lower power ratings, and rely on novel techniques for increased operational safety. However, realizing these capabilities requires intelligent control systems that can track changing power demands and make autonomous decisions based on these demands. The unique aspects of advanced reactors (e.g., strict regulatory requirements, harsh operating environments, high consequences, highly coupled dynamics, evolving knowledge, and limited operating histories) directly impact the design and deployment of control systems for these reactors. The present work identifies and evaluates these aspects so as to develop a set of control system requirements to guide future research and development. To meet these requirements, a layered control system approach is proposed that integrates digital twins with different control paradigms. This work aims to demonstrate how different methods interface with enabling solutions, and to identify any gaps that need to be researched.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

NASA NTRS2018

Starling1 - Mission Technologies Overview

The Starling series of demonstration missions will test technologies required to achieve affordable, distributed spacecraft ("swarm") missions that: are scalable to at least 100 spacecraft for applications that include synchronized multipoint measurements; involve closely coordinated ensembles of two or more spacecraft operating as a single unit for interferometric, synthetic aperture, or similar sensor architectures; or use autonomous or semi-autonomous operation of multiple spacecraft functioning as a unit to achieve science or other mission objectives with low-cost small spacecraft.Starling1 will focus on developing technologies that enable scalability and deep space application. The mission goals include the demonstration of a Mobile Ad-hoc NETwork (MANET) through an in-space communication experiment, vision based relative navigation through the Starling Formation-flying Optical eXperiment (StarFOX), and demonstration of autonomous spacecraft reconfiguration using technologies developed by the Distributed System Autonomy (DSA) project.

Cannon, Howard↗

NASA NTRS2015

AMO EXPRESS: A Command and Control Experiment for Crew Autonomy

NASA is investigating a range of future human spaceflight missions, including both Mars-distance and Near Earth Object (NEO) targets. Of significant importance for these missions is the balance between crew autonomy and vehicle automation. As distance from Earth results in increasing communication delays, future crews need both the capability and authority to independently make decisions. However, small crews cannot take on all functions performed by ground today, and so vehicles must be more automated to reduce the crew workload for such missions. NASA's Advanced Exploration Systems Program funded Autonomous Mission Operations (AMO) project conducted an autonomous command and control demonstration of intelligent procedures to automatically initialize a rack onboard the International Space Station (ISS) with power and thermal interfaces, and involving core and payload command and telemetry processing, without support from ground controllers. This autonomous operations capability is enabling in scenarios such as a crew medical emergency, and representative of other spacecraft autonomy challenges. The experiment was conducted using the Expedite the Processing of Experiments for Space Station (EXPRESS) rack 7, which was located in the Port 2 location within the U.S Laboratory onboard the International Space Station (ISS). Activation and deactivation of this facility is time consuming and operationally intensive, requiring coordination of three flight control positions, 47 nominal steps, 57 commands, 276 telemetry checks, and coordination of multiple ISS systems (both core and payload). The autonomous operations concept includes a reduction of the amount of data a crew operator is required to verify during activation or de-activation, as well as integration of procedure execution status and relevant data in a single integrated display. During execution, the auto-procedures provide a step-by-step messaging paradigm and a high level status upon termination. This messaging and high level status is the only data generated for operator display. To enhance situational awareness of the operator, the Web-based Procedure Display (WebPD) provides a novel approach to the issues of procedure display and execution tracking. For this demonstration, the procedure was initiated and monitored from the ground. As the Timeliner sequences executed, their high level execution status was transmitted to ground, for WebPD consumption.

Stetson, Howard K.↗

NASA NTRS2015

Technology Advancement for Active Remote Sensing of Carbon Dioxide from Space Using the ASCENDS CarbonHawk Experiment Simulator: First Results

The ASCENDS CarbonHawk Experiment Simulator (ACES) is a newly developed lidar developed at NASA Langley Research Center and funded by NASA's Earth Science Technology Office (ESTO) Instrument Incubator Program (IIP) that seeks to advance technologies critical to measuring atmospheric column carbon dioxide (CO2) mixing ratios in support of the NASA Active Sensing of CO2 Emissions over Nights, Days, and Seasons (ASCENDS) mission. The technology advancements targeted include: (1) increasing the power-aperture product to approach ASCENDS mission requirements by implementing multi-aperture telescopes and multiple co-aligned laser transmitters; (2) incorporating high-efficiency, high-power Erbium-Doped Fiber Amplifiers (EDFAs); (3) developing and incorporating a high-bandwidth, low-noise HgCdTe detector and transimpedence amplifier (TIA) subsystem capable of long-duration autonomous operation on Global Hawk aircraft, and (4) advancing algorithms for cloud and aerosol discrimination. The ACES instrument architecture is being developed for operation on high-altitude aircraft and will be directly scalable to meet the ASCENDS mission requirements. These technologies are critical towards developing not only spaceborne instruments but also their airborne simulators, with lower platform requirements for size, mass, and power, and with improved instrument performance for the ASCENDS mission. ACES transmits five laser beams: three from commercial EDFAs operating near 1.57 microns, and two from the Exelis oxygen (O2) Raman fiber laser amplifier system operating near 1.26 microns. The three EDFAs are capable of transmitting up to 10 watts average optical output power each and are seeded by compact, low noise, stable, narrow-linewidth laser sources stabilized with respect to a CO2 absorption line using a multi-pass gas absorption cell. The Integrated-Path Differential Absorption (IPDA) lidar approach is used at both wavelengths to independently measure the CO2 and O2 column number densities and retrieve the average column CO2 mixing ratio. The ACES receiver uses three fiber-coupled 17.8-cm diameter athermal telescopes. The transmitter assembly consists of five fiber-coupled laser collimators and an associated Risley prism pair for each laser to co-align the outgoing laser beams and to align them with the telescope field of view. The backscattered return signals collected by the three telescopes are combined in a fiber bundle and sent to a single low noise detector. The detector/TIA development has improved the existing detector subsystem by increasing its bandwidth to 4.7 MHz from 500 kHz and increasing the duration of autonomous, service-free operation periods from 4 hours to >24 hours. The new detector subsystem enables the utilization of higher laser modulation rates, which provides greater flexibility for implementing advanced thin-cloud discrimination algorithms as well as improving range-determination resolution and error reduction. The cloud/aerosol discrimination algorithm development by Langley and Exelis features a new suite of algorithms for the minimization/elimination of bias errors in the return signal induced by the presence of intervening thin clouds. Multiple laser modulation schemes are being tested in an effort to significantly mitigate the effects of thin clouds on the retrieved CO2 column amounts. Full instrument development concluded in the spring of 2014. After ground range tests of the instrument, ACES successfully completed six test flights on the Langley Hu-25 aircraft in July, 2014, and recorded data at multiple altitudes over land and ocean surfaces with and without intervening clouds. Preliminary results from these test flights will be presented in this paper.

Obland, Michael D.↗

NASA NTRS1989

Automation and robotics technology for intelligent mining systems

The U.S. Bureau of Mines is approaching the problems of accidents and efficiency in the mining industry through the application of automation and robotics to mining systems. This technology can increase safety by removing workers from hazardous areas of the mines or from performing hazardous tasks. The short-term goal of the Automation and Robotics program is to develop technology that can be implemented in the form of an autonomous mining machine using current continuous mining machine equipment. In the longer term, the goal is to conduct research that will lead to new intelligent mining systems that capitalize on the capabilities of robotics. The Bureau of Mines Automation and Robotics program has been structured to produce the technology required for the short- and long-term goals. The short-term goal of application of automation and robotics to an existing mining machine, resulting in autonomous operation, is expected to be accomplished within five years. Key technology elements required for an autonomous continuous mining machine are well underway and include machine navigation systems, coal-rock interface detectors, machine condition monitoring, and intelligent computer systems. The Bureau of Mines program is described, including status of key technology elements for an autonomous continuous mining machine, the program schedule, and future work. Although the program is directed toward underground mining, much of the technology being developed may have applications for space systems or mining on the Moon or other planets.

Welsh, Jeffrey H.↗

NASA NTRS1989

Concept synthesis of an equipment manipulation and transportation system EMATS

The European Columbus Scenario is established. One of the Columbus Elements, the Man Tended Free Flyer will be designed for fully autonomous operation in order to provide the environment for micro gravity facilities. The Concept of an autonomous automation system which perform servicing of facilities and deals with related logistic tasks is discussed.

Depeuter, W.↗

NASA NTRS1996

SEDSAT 1 Technologies

The original goals of this project were to support the development of SEDSAT 1 for a tethered launch in July of 1997. This specifically required: (1) Monitoring development progress against a comprehensive delivery plan; (2) Incremental development and release of CDS and SEASIS software; (3) Supporting the integration of version 1.0 SEASIS software that will allow minimal autonomous operation without a software reload. These algorithms would include image quality evaluation, attitude determination, and autonomous earth imaging; and (4) Developing software requirements and design for ground segment software, concentrating on command and data download capability; and interface to external development efforts for a more comprehensive software suite to be used after the initial mission. Because of an unfavorable space shuttle safety review of the SEDS-3 tether deployer, and cost and schedule problems in upgrading the deployer, the mission was changed to an independent launch of SEDSAT. The original plan was to do a tether-less deployment from the space shuttle. Since this would have resulted in an unacceptable orbital lifetime, the mission was changed again to a tethered launch from a Delta II in June 1998. As a result of Marshall Space Flight Center's redirection of the SEDS-3 mission away from a tether launch, the whole question of a tether endmass had to be reconsidered. The net result of these multiple changes was twofold. First, we completed work needed to define some aspects of ground software on SEDSAT 1 that would remain constant no matter the launch mode. Second, we developed a set of concepts for using SEDSAT 1 technology to support alternative endmass missions on SEDS-3. Both of these are included.

Maier, Mark W.↗

NASA NTRS2014

Technologies for Human Exploration

Access to Space, Chemical Propulsion, Advanced Propulsion, In-Situ Resource Utilization, Entry, Descent, Landing and Ascent, Humans and Robots Working Together, Autonomous Operations, In-Flight Maintenance, Exploration Mobility, Power Generation, Life Support, Space Suits, Microgravity Countermeasures, Autonomous Medicine, Environmental Control.

Drake, Bret G.↗

NASA NTRS2015

AMO EXPRESS: A Command and Control Experiment for Crew Autonomy Onboard the International Space Station

NASA is investigating a range of future human spaceflight missions, including both Mars-distance and Near Earth Object (NEO) targets. Of significant importance for these missions is the balance between crew autonomy and vehicle automation. As distance from Earth results in increasing communication delays, future crews need both the capability and authority to independently make decisions. However, small crews cannot take on all functions performed by ground today, and so vehicles must be more automated to reduce the crew workload for such missions. NASA's Advanced Exploration Systems Program funded Autonomous Mission Operations (AMO) project conducted an autonomous command and control experiment on-board the International Space Station that demonstrated single action intelligent procedures for crew command and control. The target problem was to enable crew initialization of a facility class rack with power and thermal interfaces, and involving core and payload command and telemetry processing, without support from ground controllers. This autonomous operations capability is enabling in scenarios such as initialization of a medical facility to respond to a crew medical emergency, and representative of other spacecraft autonomy challenges. The experiment was conducted using the Expedite the Processing of Experiments for Space Station (EXPRESS) rack 7, which was located in the Port 2 location within the U.S Laboratory onboard the International Space Station (ISS). Activation and deactivation of this facility is time consuming and operationally intensive, requiring coordination of three flight control positions, 47 nominal steps, 57 commands, 276 telemetry checks, and coordination of multiple ISS systems (both core and payload). Utilization of Draper Laboratory's Timeliner software, deployed on-board the ISS within the Command and Control (C&C) computers and the Payload computers, allowed development of the automated procedures specific to ISS without having to certify and employ novel software for procedure development and execution. The procedures contained the ground procedure logic and actions as possible to include fault detection and recovery capabilities. The autonomous operations concept includes a reduction of the amount of data a crew operator is required to verify during activation or de-activation, as well as integration of procedure execution status and relevant data in a single integrated display. During execution, the auto-procedures (via Timerliner) provide a step-by-step messaging paradigm and a high-level status upon termination. This messaging and high-level status is the only data generated for operator display. To enhance situational awareness of the operator, the Web-based Procedure Display (WebPD) provides a novel approach to the issues of procedure display and execution tracking. WebPD is a web based application that serves as the user interface for electronic procedure execution. It incorporates several aspects of the HTML5 standard. Procedures are written in a dialect of XML called Procedure Representation Language (PRL). WebPD tracks execution status in the procedure or procedures being displayed. WebPD aggregates and simplifies the auto-sequence execution status information, and formatted to be easily followed and understood by an operator who is not dedicated to actively monitoring the task. WebPD also provides an integrated data and control interface to pause or halt the execution in order to provide a check point of operation and to examine progress before starting the next sequence of activities. For this demonstration, the procedure was initiated and monitored from the ground. As the Timeliner sequences executed, their high-level execution status was written to PLMDM memory. This memory is read and downlinked via Ku-Band at a 1 Hz rate. The data containing the high-level execution status is de-commutated on the ground, and rebroadcast for WebPD consumption. A future demonstration will be performed onboard, with ISS astronauts initiating the operations instead of ground controllers. The AMO EXPRESS experiment demonstrated activation and de-activation of EXPRESS rack 7, providing the capability of future single button activations and deactivations of facility class racks. The experiment achieved numerous technical and operations 'firsts' for the ISS

Stetson, Howard K.↗

NASA NTRS1993

A fault-tolerant intelligent robotic control system

This paper describes the concept, design, and features of a fault-tolerant intelligent robotic control system being developed for space and commercial applications that require high dependability. The comprehensive strategy integrates system level hardware/software fault tolerance with task level handling of uncertainties and unexpected events for robotic control. The underlying architecture for system level fault tolerance is the distributed recovery block which protects against application software, system software, hardware, and network failures. Task level fault tolerance provisions are implemented in a knowledge-based system which utilizes advanced automation techniques such as rule-based and model-based reasoning to monitor, diagnose, and recover from unexpected events. The two level design provides tolerance of two or more faults occurring serially at any level of command, control, sensing, or actuation. The potential benefits of such a fault tolerant robotic control system include: (1) a minimized potential for damage to humans, the work site, and the robot itself; (2) continuous operation with a minimum of uncommanded motion in the presence of failures; and (3) more reliable autonomous operation providing increased efficiency in the execution of robotic tasks and decreased demand on human operators for controlling and monitoring the robotic servicing routines.

Marzwell, Neville I.↗

NASA NTRSDate not supplied

Assessing Performance of Radar and Visual Sensing Techniques for Ground-To-Air Surveillance in Advanced Air Mobility

The safe integration of Unmanned Aircraft Vehicles (UAV) within the civil airspace is of great interest to NASA’s Advanced Air Mobility project, which envisions high density of operations in and around urban areas that include both UAV and AAM aircraft. To enable safe autonomous operations of both platforms, reliable airspace surveillance strategies must be designed and experimentally validated in relevant scenarios, where multiple small UAV operate flying in low altitude conditions. An example of such a scenario is described in this paper which provides performance assessment of various sensing strategies experimentally tested during flight campaigns with four UAV completing simultaneous missions from vertiports. Such campaigns are performed by the High Density Vertiplex subproject which assesses a prototype of Urban Air Mobility ecosystem. For the purposes of this work, the flights are observed from multiple sensing nodes each with radar and camera sensors. The visual detection and tracking algorithms achieved 96.1% to 99.9% average tracking coverage of the UAV above the horizon, reaching detection ranges larger than 1 kilometer for octocopter. Radar-based tracking shows a lower coverage mainly due to ground clutter removal challenges but provides comparable detection ranges and meter-level range accuracy.

Federica Vitiello↗

NASA NTRSDate not supplied

Tradeoffs When Considering Deep Reinforcement Learning for Contingency Management in Advanced Air Mobility

Air transportation is undergoing a rapid evolution globally with the introduction of Advanced Air Mobility (AAM) and with it comes novel challenges and opportunities for transforming aviation. As AAM operations introduce increasing heterogeneity in vehicle capabilities and density, increased levels of automation are likely necessary to achieve operational safety and efficiency goals. This paper focuses on one example where increased automation has been suggested. Autonomous operations will need contingency management systems that can monitor evolving risk across a span of interrelated (or interdependent) hazards and, if necessary, execute appropriate control interventions via supervised or automated decision making. Accommodating this complex environment may require automated functions (autonomy) that apply artificial intelligence (AI) techniques that can adapt and respond to a quickly changing environment. This paper explores the use of Deep Reinforcement Learning (DRL) which has shown promising performance in complex and high-dimensional environments where the objective can be constructed as a sequential decision-making problem. An extension of a prior formulation of the contingency management problem as a Markov Decision Process (MDP) is presented and uses a DRL framework to train agents that mitigate hazards present in the simulation environment. A comparison of these learning-based agents and classical techniques is presented in terms of their performance, verification difficulties, and development process.

machine learning↗

NASA NTRSDate not supplied

Tradeoffs When Considering Deep Reinforcement Learning for Contingency Management in Advanced Air Mobility

Air transportation is undergoing a rapid evolution globally with the introduction of Advanced Air Mobility (AAM) and with it comes novel challenges and opportunities for transforming aviation. As AAM operations introduce increasing heterogeneity in vehicle capabilities and density, increased levels of automation are likely necessary to achieve operational safety and efficiency goals. This paper focuses on one example where increased automation has been suggested. Autonomous operations will need contingency management systems that can monitor evolving risk across a span of interrelated (or interdependent) hazards and, if necessary, execute appropriate control interventions via supervised or automated decision making. Accommodating this complex environment may require automated functions (autonomy) that apply artificial intelligence (AI) techniques that can adapt and respond to a quickly changing environment. This paper explores the use of Deep Reinforcement Learning (DRL) which has shown promising performance in complex and high-dimensional environments where the objective can be constructed as a sequential decision-making problem. An extension of a prior formulation of the contingency management problem as a Markov Decision Process (MDP) is presented and uses a DRL framework to train agents that mitigate hazards present in the simulation environment. A comparison of these learning-based agents and classical techniques is presented in terms of their performance, verification difficulties, and development process.

machine learningautonomous systems; flight simulat↗