SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
This paper describes three autonomy architectures for a system that continuously plans to control a fleet of spacecraft using collective mission goals instead of goals or command sequences for each spacecraft.
The Deep Space One (DS1) mission, scheduled to fly in 1998, will be the first NASA spacecraft to feature an on-board planner. The planner is part of an artificial intelligence based control architecture that comprises the planner/scheduler, a plan execution engine, and a model-based fault diagnosis and reconfiguration engine...This paper describes the on-board planning and scheduling component of the DS1 autonomy architecture.
Telerobotics continues to provide a means for extending human presence in space as an integral part of the control architecture of unmanned spacecraft. Supervised autonomy for control of space manipulators provides a safe and effective means for time delayed remote control of unmanned spacecraft.
Presentation offers methodology and approaches for increasing autonomy in aviation and uses car industry (self-driving cars) as comparison.
This report focuses on the role of the Fleet Manager (FM) and, in particular, the ways in which automation could support their position as they manage multiple aircraft and operators in a highly dynamic, advanced air mobility (e.g., air taxi) environment, specifically "high-density vertiport" (HDV) regions. Similar to terminal area operations for traditional aviation, operations involving HDVs will need to be highly structured while also remaining resilient to the various contingencies that can happen in that environment. The role of the FM is consistent with an “m:N” architecture, where “m” number of operators cooperatively manage “N” number of vehicles (where “N” is always larger than “m”). In such a paradigm it is critical to provide the operator with the tools and information necessary to manage their fleet safely and navigate the known pitfalls with highly automated, complex systems (e.g., brittleness, insufficient situation awareness, skill degradation). As the field of m:N has expanded as an area of study, a set of higher-level automation concepts have emerged—namely “plays,” “working agreements,” and “human-autonomy teaming” (HAT)—that could support operators in this new role.
The NASA Office of the Chief Health and Medical Officer (OCHMO) requested HRP conduct a survey of industry and government standards and guidelines for the interaction of automation/autonomy with humans. Guidance for verification testing of human-automation designs was also requested. A rapid response was requested so that the deliverables could be provided as part of a future NASA solicitation for automated systems for HLS. The project began with a broad review and analysis of prior NASA work in this area, as well as external standards related to human interaction with automation/autonomy. The team reviewed approximately 200 documents, and interviewed multiple industry and government subject matter experts (SMEs). “Gold standard” documents were selected for focus and key categories were identified. A cross-walk of standards was created, and common standards and guidelines were selected and rephrased (where necessary) to be information-rich, concise, and usable. Verification guidelines and examples were also developed. The final report includes an introduction to automation and autonomy, selected standards and guidelines, verification method guidance, and research gaps. In addition to the report, the team delivered a list of candidate standards for autonomous vehicles, as well as a list of candidate standards for a future NASA-STD-3001 update on automation/autonomy. The presentation will describe the rapid response project approach and methods, examples of standards and guidelines identified and delivered to OCHMO, and research needed to develop additional automation/autonomy design guidance where gaps currently exist.
To enable scalability of air travel for use cases such as cargo delivery, it is anticipated that future air traffic operations will involve unmanned aircraft operated by remote pilots. Of particular interest are schemes where a small number of pilots operate a large number of vehicles, mitigating high cost and pilot shortage issues. Such architectures require increased levels of automation and supervisory control modes. They also require ensuring safe operations when the command and control link to the vehicle is degraded or lost completely, rendering the vehicle autonomous. To evaluate these variable and dynamic architectures, this paper will present a framework for decomposing the functions necessary to ensure safe, orderly, and expeditious air travel, assessing the agents in the system, and classifying the levels of autonomy. Then, an example allocation to agents of roles for the function of separation assurance is presented, highlighting the dependency of the allocation on three main factors: time criticality of a potential separation violation, the ratio of pilots to vehicles, and the loss of the command and control link.
To enable scalability of air travel for use cases such as cargo delivery, it is anticipated that future air traffic operations will involve unmanned aircraft operated by remote pilots. Of particular interest are schemes where a small number of pilots operate a large number of vehicles, mitigating high cost and pilot shortage issues. Such architectures require increased levels of automation and supervisory control modes. They also require ensuring safe operations when the command and control link to the vehicle is degraded or lost completely, rendering the vehicle autonomous. To evaluate these variable and dynamic architectures, this paper will present a framework for decomposing the functions necessary to ensure safe, orderly, and expeditious air travel, assessing the agents in the system, and classifying the levels of autonomy. Then, an example allocation to agents of roles for the function of separation assurance is presented, highlighting the dependency of the allocation on three main factors; time criticality of a potential separation violation, the ratio of pilots to vehicles, and the loss of the command and control link.
This talk describes the need for human autonomy teaming (HAT), the basic principles of HAT and the application of those principles in various domains.
To achieve the scalability envisioned for many Advanced Air Mobility (AAM) applications, uncrewed aerial system (UAS) concepts are being pursued with the goal of enabling fewer human operators to manage more increasingly autonomous vehicles. NASA’s Transformational Tools and Technologies – Revolutionary Aviation Mobility (T3-RAM) subproject has identified human-autonomy teaming (HAT) as a critical area of research required to support these operations. Under T3-RAM, the HAT Foundational Research Activity has been tasked with providing basic research to identify HAT and human-automation interaction (HAI) principles that can be used to achieve scalable multi-vehicle UAS operations. This paper first outlines a research model to produce ecologically relevant basic research, then contextualizes completed and planned research and development activities within this model. Proposed research threads are presented, along with their practical and theoretical implications.
Envisioned future Advanced Air Mobility (AAM) operations will require a transition of aircraft command and control from onboard pilots to remote operators. The National Aeronautics and Space Administration (NASA) has developed a research ground control station (GCS) software called MPATH (Measuring Performance for Autonomy Teaming with Humans) to study the human factors of remote operators in a representative AAM environment, where small uncrewed aerial systems (sUAS; simulated or real) act as surrogates for larger AAM aircraft. A primary focus of the research being conducted with the MPATH GCS is scalability (i.e., one human managing multiple vehicles). MPATH has demonstrated to be a useful capability for human factors research and remote operations. Two initial usability studies and a multi-vehicle control assessment were recently conducted, and usability data and operator feedback were collected. Generally, participants rated MPATH high on usability and interface quality, whereas information quality was rated slightly lower. These results were supported by specific feedback. Several generalized GCS design recommendations are proposed based on the results and feedback from these flight activities. Future updates to MPATH will incorporate the proposed recommendations. In practice, these recommendations could be used by any GCS software designer or developer to promote usability and safety.
An exploratory human-in-the-loop simulation was conducted to investigate and characterize a Human-Autonomy Teaming (HAT) Assistant to support a remote operator of multiple small Uncrewed Aircraft Systems (sUAS) using a ground control station (GCS) in the context of a wildland fire surveillance mission. Operator performance using the GCS with the HAT Assistant (Assisted Mode) was compared to operator performance using the GCS without the HAT Assistant (Unassisted Mode) during two types of contingency-event scenarios (Low and High Complexity). In the Assisted Mode, the HAT Assistant provided updates to the level of risk to the mission along with recommendations for risk mitigation, which were not provided in the Unassisted Mode. No significant differences in objective performance and subjective ratings of workload, situation awareness, and trust in automation between the Assisted and Unassisted Modes were detected, however there were indications that participants preferred the Assisted GCS over the Unassisted GCS and directions for further development were explored. Additional work is necessary to further refine the HAT Assistant and better characterize its effects on remote operator performance while managing multiple sUAS assets. Future work is recommended to optimize the implementation of an assistant to support operator performance during different missions and across vehicle classes.
An exploratory human-in-the-loop simulation was conducted to investigate and characterize a Human-Autonomy Teaming (HAT) Assistant to support a remote operator of multiple small Uncrewed Aircraft Systems (sUAS) using a ground control station (GCS) in the context of a wildland fire surveillance mission. Operator performance using the GCS with the HAT Assistant (Assisted Mode) was compared to operator performance using the GCS without the HAT Assistant (Unassisted Mode) during two types of contingency-event scenarios (Low and High Complexity). In the Assisted Mode, the HAT Assistant provided updates to the level of risk to the mission along with recommendations for risk mitigation, which were not provided in the Unassisted Mode. No significant differences in objective performance and subjective ratings of workload, situation awareness, and trust in automation between the Assisted and Unassisted Modes were detected, however there were indications that participants preferred the Assisted GCS over the Unassisted GCS and directions for further development were explored. Additional work is necessary to further refine the HAT Assistant and better characterize its effects on remote operator performance while managing multiple sUAS assets. Future work is recommended to optimize the implementation of an assistant to support operator performance during different missions and across vehicle classes
NASA’s Distributed Spacecraft Autonomy (DSA) Project developed and demonstrated software to enhance multi-spacecraft mission adaptability, efficiently allocate tasks between spacecraft using ad-hoc networking, and enable human-swarm commanding of Distributed Space Missions.
The Lunar Autonomy Challenge is a competition for university students to develop algorithms to autonomously control systems in a realistic lunar environment.
The potential development of autonomous onboard navigation systems for future space missions is discussed. Motivations for autonomy in the near-earth satellite program and in deep-space navigation are examined, emphasizing lower mission-operation costs for the first, and reduction of ground-system complexity as well as encounter accuracy for the latter. Uses of navigation parameters by the Information Adaptive System, including sensor-pointing computation and image correction, are noted, as are the accuracy parameters of the NAVSTAR Global Positioning System. The role of autonomous onboard navigation in contact exploration of Neptune's and Uranus's satellites as well as in such extreme-high-velocity encounters as those with retrograde comets is investigated. The Automated Optical Navigation program is described
Design guidelines and functional systems being considered in the process of defining the configuration of the automated systems for a manned space station are outlined. The requirements are dependent on life-cycle costing and will set the necessary level of automation, as well as autonomy from outside commands. Fault protection routines have been largely devised according to successful programming on the Voyager spacecraft. An analysis is still needed of the housekeeping functions, including human necessities, machine functions, and mission objectives. A data base will result, defining the functions that have historically been delegated to either man or machine. Care must be taken to coordinate and document stationkeeping functions that might interface with mission functions. A data management system that is flexible with regards to changing mission objectives and to the MTBF factors, which will determine the level of technology to be used is required. Expert systems will be integrated into the automation to guide the machines in problem solving, including ensuring adequate management of the battery subsystem.
Potential applications for autonomous space platforms (SP) are discussed. The platforms are assumed to have long in-service lifetimes and therefore be flexible as to configuration modification and payload changeout. Higher degrees of autonomy, particularly from ground control, are made possible because of the rapid increase of microprocessor power and artificial intelligence advances. Functioning independently, the platforms are to rely only on periodic refurbishment visits by, e.g., the Orbiter. The Manned Space Station (MSS) will be the most complex structure, involving multifacted man-machine interfaces. The SP can be subsystems of the MSS (or other platforms), handling communications enunciation, data acquisition, analysis and telemetry, fault detection and isolation, systems monitoring and control, etc. The SP adopted will depend in all cases on costs vs benefits analyses to determine the worth of removing the function(s) from direct, regular human intervention.