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.
NASA's Mars Science Laboratory (MSL) rover mission is planning to make use of advanced software technologies in order to support fulfillment of its ambitious science objectives. The mission plans to adopt the Mission Data System (MDS) as the mission software architecture, and plans to make significant use of on-board autonomous capabilities for the rover software.
This paper presents the development of ICAROUS-2 (Independent Configurable Architecture for Reliable Operation of Unmanned Systems with Distributed Onboard Services), the second generation of a software architecture that integrates several algorithms as distributed onboard services to enable robust autonomous UAS applications. In particular, the ICAROUS architecture defines a framework to perform detect and avoid, geofencing, path monitoring, path planning, and autonomous decision making to ensure safety and mission progress. Most of the core algorithms implemented in ICAROUS are formally verified using an interactive theorem prover. These algorithms are composed together using a plan execution engine, whose operational semantics is formally specified. A description of the integrated architecture, services currently available, and flight test results highlighting the capability of ICAROUS are presented.
This paper presents the development of ICAROUS-2 (Independent Configurable Architecture for Reliable Operation of Unmanned Systems with Distributed Onboard Services), the second generation of a software architecture that integrates several algorithms as distributed onboard services to enable robust autonomous UAS applications. In particular, the ICAROUS architecture defines a framework to perform detect and avoid, geofencing, path monitoring, path planning, and autonomous decision making to ensure safety and mission progress. Most of the core algorithms implemented in ICAROUS are formally verified using an interactive theorem prover. These algorithms are composed together using a plan execution engine, whose operational semantics is formally specified. A description of the integrated architecture, services currently available, and flight test results highlighting the capability of ICAROUS are presented.
NASA's Jet Propulsion Labortory has embarked on the Mission Data System (MDS) project to produce a reusable, integrated flight and ground software architecture.
Mission to Planet Earth (MTPE) is a long-term NASA research mission to study the processes leading to global climate change. The Earth Observing System (EOS) is a NASA campaign of satellite observatories that are a major component of MTPE. The EOS Data and Information System (EOSDIS) is another component of MTPE that will provide the Earth science community with easy, affordable, and reliable access to Earth science data. EOSDIS is a distributed system, with major facilities at six Distributed Active Archive Centers (DAACS) located throughout the United States. The EOSDIS software architecture is being designed to receive, process, and archive several terabytes of science data on a daily basis. Thousands of science users and perhaps several hundred thousands of non-science users are expected to access the system. While there are many segments in EOSDIS (e.g., flight operations, network) this case study discusses the development of the science data processing segment (SDPS). We briefly review the architecture of the system, the goals of the SDPS, and the development progress to date. This study highlights key software development challenges, experiences integrating COTS, and the difficulties of managing a complex system development effort.
At NASA Ames Research Center, the Intelligent Robotics Group (IRG) fields the K9 and K10 class rovers. Both use a mobile robot hardware architecture designed for extensibility and reconfigurability that allows for rapid changes in instrumentation and provides a high degree of modularity. Over the past ssveral years, we have worked with instrument developers at NASA centers, universities, and national laboratories to integrate or partially integrate their instruments onboard the K9 and K10 rovers. Early efforts required considerable interaction to work through integration issues such as power, data protocol and mechanical mounting. These interactions informed the design of our current avionics architecture, and have simplified more recent integration projects. In this paper, we will describe the IRG extensible avionics and software architecture and the effect it has had on our recent instrument integration efforts, including integration of four Mars Instrument Development Program devices.
In 2006, NASA Ames Research Center's (ARC) Intelligent Systems Division, and NASA Johnson Space Centers (JSC) Mission Operations Directorate (MOD) began a collaboration to move user applications for JSC's mission control center to a new software architecture, intended to replace the existing user applications being used for the Space Shuttle and the International Space Station. It must also carry NASA/JSC mission operations forward to the future, meeting the needs for NASA's exploration programs beyond low Earth orbit. Key requirements for the new architecture, called Mission Control Technologies (MCT) are that end users must be able to compose and build their own software displays without the need for programming, or direct support and approval from a platform services organization. Developers must be able to build MCT components using industry standard languages and tools. Each component of MCT must be interoperable with other components, regardless of what organization develops them. For platform service providers and MOD management, MCT must be cost effective, maintainable and evolvable. MCT software is built from components that are presented to users as composable user objects. A user object is an entity that represents a domain object such as a telemetry point, a command, a timeline, an activity, or a step in a procedure. User objects may be composed and reused, for example a telemetry point may be used in a traditional monitoring display, and that same telemetry user object may be composed into a procedure step. In either display, that same telemetry point may be shown in different views, such as a plot, an alpha numeric, or a meta-data view and those views may be changed live and in place. MCT presents users with a single unified user environment that contains all the objects required to perform applicable flight controller tasks, thus users do not have to use multiple applications, the traditional boundaries that exist between multiple heterogeneous applications disappear, leaving open the possibility of new operations concepts that are not constrained by the traditional applications paradigm.
Existing solid state drive (SSD) simulators unfortunately lack hardware and/or software architecture models. Consequently, they are far from capturing the critical features of contemporary SSD devices. More importantly, while the performance of modern systems that adopt SSDs can vary based on their numerous internal design parameters and storage-level configurations, a full system simulation with traditional SSD models often requires unreasonably long runtimes and excessive computational resources. In this work, we propose SimpleSSD, a high-fidelity simulator that models all detailed characteristics of hardware and software, while simplifying the nondescript features of storage internals. In contrast to existing SSD simulators, SimpleSSD can easily be integrated into publicly-available full system simulators. Here, in addition, it can accommodate a complete storage stack and evaluate the performance of SSDs along with diverse memory technologies and microarchitectures. Thus, it facilitates simulations that explore the full design space at different levels of system abstraction.
Although communications-based cognitive engines have been proposed, very few have been implemented in a full system, especially in a space communications system. In this paper, we detail the implementation of a multi-objective reinforcement-learning algorithm and deep artificial neural networks for the use as a radio-resource-allocation controller. The modular software architecture presented encourages re-use and easy modification for trying different algorithms. Various trade studies involved with the system implementation and integration are discussed. These include the choice of software libraries that provide platform flexibility and promote reusability, choices regarding the deployment of this cognitive engine within a system architecture using the DVB-S2 standard and commercial hardware, and constraints placed on the cognitive engine caused by real-world radio constraints. The implemented radio-resource allocation-management controller was then integrated with the larger spaceground system developed by NASA Glenn Research Center (GRC).
Although communications-based cognitive engines have been proposed, very few have been implemented in a full system, especially in a space communications system. In this paper, we detail the implementation of a multi-objective reinforcement-learning algorithm and deep artificial neural networks for the use as a radio-resource-allocation controller. The modular software architecture presented encourages re-use and easy modification for trying different algorithms. Various trade studies involved with the system implementation and integration are discussed. These include the choice of software libraries that provide platform flexibility and promote reusability, choices regarding the deployment of this cognitive engine within a system architecture using the DVB-S2 standard and commercial hardware, and constraints placed on the cognitive engine caused by real-world radio constraints. The implemented radio-resource allocation-management controller was then integrated with the larger spaceground system developed by NASA Glenn Research Center (GRC).
This viewgraph presentation reviews the analysis of the architecture of complex systems and the development of a tool to assist in the analysis. The goal of the project was to research and develop a tool for architecture analysis of dynamic and static data. The new tool, Dyn-SAVE, was an extension of an already existing static tool, Software Architecture Visualization and Evaluation (SAVE).
The Orion Crew Exploration Vehicle is being designed with greater automation capabilities than any other crewed spacecraft in NASA s history. The Guidance, Navigation, and Control (GN&C) flight software architecture is designed to provide a flexible and evolvable framework that accommodates increasing levels of automation over time. Within the GN&C flight software, a data-driven approach is used to configure software. This approach allows data reconfiguration and updates to automated sequences without requiring recompilation of the software. Because of the great dependency of the automation and the flight software on the configuration data, the data management is a vital component of the processes for software certification, mission design, and flight operations. To enable the automated sequencing and data configuration of the GN&C subsystem on Orion, a desktop database configuration tool has been developed. The database tool allows the specification of the GN&C activity sequences, the automated transitions in the software, and the corresponding parameter reconfigurations. These aspects of the GN&C automation on Orion are all coordinated via data management, and the database tool provides the ability to test the automation capabilities during the development of the GN&C software. In addition to providing the infrastructure to manage the GN&C automation, the database tool has been designed with capabilities to import and export artifacts for simulation analysis and documentation purposes. Furthermore, the database configuration tool, currently used to manage simulation data, is envisioned to evolve into a mission planning tool for generating and testing GN&C software sequences and configurations. A key enabler of the GN&C automation design, the database tool allows both the creation and maintenance of the data artifacts, as well as serving the critical role of helping to manage, visualize, and understand the data-driven parameters both during software development and throughout the life of the Orion project.
This paper describes the tools and display capabilities being developed to visualize aircraft flight and trajectory information that is produced during a distributed Live, Virtual, Constructive (LVC) simulation events involving a mix of manned and unmanned aircraft. This paper discusses the design approach and general software architecture of the Remote User Monitor Service (RUMS) application. The simulations and flight tests are conducted over a nationwide network with components communicating via software gateways that translate the messages and provide a location to access and collect data. The data collected is not only archived for post-processing and analysis, but also processed in real-time by RUMS, which allows dynamic viewing of the information via a web browser. This real-time dynamic viewing of the data allows test conductors and researchers to monitor the health of the distributed system and of the data being collected not only in situ where tests are conducted but also at remote locations.
As a result of recent advances in software engineering capabilities, we are now in a more stable environment. De-facto hardware and software standards are emerging. Work on software architecture and design patterns signals a consensus on the importance of early system-level design decisions, and agreements on the uses of certain paradigmatic software structures. We now routinely build systems that would have been risky or infeasible a few years ago. Unfortunately, technological developments threaten to destabilize software design again. Systems designed around novel computing and peripheral devices will spark ambitious new projects that will stress current software design and engineering capabilities. Micro-electro-mechanical systems (MEMS) and related technologies provide the physical basis for new systems with the potential to produce this kind of destabilizing effect. One important response to anticipated software engineering and design difficulties is carefully directed engineering-scientific research. Two specific problems meriting substantial research attention are: A lack of sufficient means to build software systems by generating, extending, specializing, and integrating large-scale reusable components; and a lack of adequate computational and analytic tools to extend and aid engineers in maintaining intellectual control over complex software designs.
There is a desire to design autonomous systems in such a way that capabilities can be easily added or re- combined to produce new behaviors while preserving their safety properties. ICAROUS, a prototype software architecture for building safety-centric autonomous unmanned aircraft applications, is designed to support this type of extensibility and re-configurability. In ICAROUS, core capabilities are implemented as individual soft- ware services, so that enabling access to new capabilities simply requires adding new services. To make use of these capabilities, ICAROUS includes a specialized service that provides a general framework for config- uring the relative priorities, conditions, and rules that govern how different modules should be engaged and disengaged during flight. The inherent complexity of coordinating multiple modules under changing conditions makes it difficult to determine whether a particular configuration could have erroneous behaviors in certain circumstances. A robust set of integration tests can help discover errors, but testing can only realistically cover a relatively small proportion of total system behaviors. Developing good tests and interpreting the results to pinpoint the cause of errors when they arise can also be very time-consuming. To supplement testing, formal methods can be used to model and analyze complex systems, achieving better coverage and simplifying the process of finding, understanding, and fixing errors. To demonstrate these benefits, this paper explores the ap- plication of formal methods to ICAROUS. In particular, the Spin model checker is used to specify requirements for and model portions of the system, then verify whether the model satisfies the requirements and find and fix errors when it does not.
The Payload Operations Integration Center (POIC) at Marshall Space Flight Center (MSFC) supports planning, coordination and scheduling of science activities for the International Space Station (ISS) in coordination with other NASA centers, international partners, and payload developers. The ability to efficiently plan and re-plan in response to change is critical to the flight planning teams. With the achievement of supporting a fourth crew member aboard the ISS and an increasing amount of payload science activities, came the need for a dynamic, more efficient way of building timeline planning reports that could be readily updated as fast as payload science plans could change. This paper addresses software architecture considerations in the successful cross-center development of an automated planning tool with multiple data sources. It also discusses the practical implementation of a time-boxed, hybrid Agile Software Development (ASD) approach to deliver customer-driven value despite changing requirements with respect to low-Earth orbit operational planning activities. The goal of this paper is to open discussion with members of the international community and trade effective strategies for cross-center architectural and customer-developer driven collaborations, to support increasing utilization of planning and conducting science activities in space.
There is a desire to design autonomous systems in such a way that capabilities can be easily added or re-combined to produce new behaviors while preserving their safety properties. ICAROUS, a prototype software architecture for building safety-centric autonomous unmanned aircraft applications, is designed to support this type of extensibility and re-configurability. In ICAROUS, core capabilities are implemented as individual soft- ware services, so that enabling access to new capabilities simply requires adding new services. To make use of these capabilities, ICAROUS includes a specialized service that provides a general framework for config- uring the relative priorities, conditions, and rules that govern how different modules should be engaged and disengaged during flight. The inherent complexity of coordinating multiple modules under changing conditions makes it difficult to determine whether a particular configuration could have erroneous behaviors in certain circumstances. A robust set of integration tests can help discover errors, but testing can only realistically cover a relatively small proportion of total system behaviors. Developing good tests and interpreting the results to pinpoint the cause of errors when they arise can also be very time-consuming. To supplement testing, formal methods can be used to model and analyze complex systems, achieving better coverage and simplifying the process of finding, understanding, and fixing errors. To demonstrate these benefits, this paper explores the ap- plication of formal methods to ICAROUS. In particular, the Spin model checker is used to specify requirements for and model portions of the system, then verify whether the model satisfies the requirements and find and fix errors when it does not.
For high-performance computing (HPC) system designers and users, meeting the myriad challenges of next-generation exascale supercomputing systems requires rethinking their approach to application and system software design. Among these challenges, providing resiliency and stability to the scientific applications in the presence of high fault rates requires new approaches to software architecture and design. As HPC systems become increasingly complex, they require intricate solutions for detection and mitigation for various modes of faults and errors that occur in these large-scale systems, as well as solutions for failure recovery. These resiliency solutions often interact with and affect other system properties, including application scalability, power and energy efficiency. Therefore, resilience solutions for HPC systems must be thoughtfully engineered and deployed.In previous work, we developed the concept of resilience design patterns, which consist of templated solutions based on well-established techniques for detection, mitigation and recovery. In this paper, we use these patterns as the foundation to propose new approaches to designing runtime systems for HPC systems. The instantiation of these patterns within a runtime system enables flexible and adaptable end-to-end resiliency solutions for HPC environments. The paper describes the architecture of the runtime system, named Plexus, and the strategies for dynamically composing and adapting pattern instances under runtime control. This runtime-based approach enables actively balancing the cost-benefit trade-off between performance overhead and protection coverage of the resilience solutions. Based on a prototype implementation of PLEXUS, we demonstrate the resiliency and performance gains achieved by the pattern-based runtime system for a parallel linear solver application.