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.
APGEN (Activity Plan GENerator - 314), a multi-mission planning tool, must interface with external software to vest serve its users. AP-GEN's original method for incorporating external software, the User-Defined library mechanism, has been very successful in allowing APGEN users access to external software functionality.
Most approaches to robust automony with respect to planning and execution are focused on either providing models that allow for flexibility or providing techniques for changing models to improve performance. We take these techniques into consideration, but focus the majority of our work on robust autonomous planning and execution with imperfect models.
This paper describes the mixed-initiative planning system for MAMM, which dramatically reduced mission-planning costs through automation to just a few scenarios for evaluating mission-design trades.
This paper describes an approach for optimizing over interdependent planning goals. We have implemented a methodology for representing and utilizing information about interdependent goals and their related utilities using the ASPEN planning and scheduling system.
This Science Data Validation Plan describes the plans for validating a subset of the Multi-angle Imaging SpectroRadiometer (MISR) Level 2 algorithms and data products and supplying top-of-atmosphere (TOA) radiances to the In-flight Radiometric Calibration and Characterization (IFRCC) subsystem for vicarious calibration.
This paper describes the application of an automated planning and scheduling system to the NASA Earth Orbitin 1 (EO-1) missions. The planning system, ASPEN, is used to autonomously schedule the daily activites of the satellite.
Plan-IT-II [1,2,3,4,5,6] represents a powerful capability for planning and scheduling of activities to be executed by interplanetary spacecraft. Its development and gradual acceptance by conservative flight projects spans 16 years after a couple of false in the prior 3 years.
Automated planning and scheduling technology - we'll call it automated planning systems, for the sake of brevity-is applicable to a wide spectrum of spaceflight missions, from those with limited onboard computational capabilities, such as Lunar Prospector, to those with highly sophisticated software, such as Cassini.
This paper focuses on the onboard planner and scheduler CASPER, whose core planning engine is based on the ground system ASPEN. Given the challenges of developing flight software, we discuss several of the issues encountered in preparing the planner for flight, including reducing the code image size, determining what data to place within the engineering telemetry packet, and performing long term planning.
This paper describes MEXEC, an implemented micro executive that compiles a device model into an interal structure. Not only does this structure facilitate computing the most likely current device mode from n sets of sensor measurements, but it also facilitates generating an n step reconfiguration plan that is most likely not to result in reaching a target mode - if such a plan exists.
The subject paper shows promise in leading to a useful system for integrating deliberative planning, plan repair, and execution control in a dynamic environment with real-time constraints. The conditions in which the system described would enable such integration seem to be ones that would apply to significant practical problems for an autonomous rover.
This paper describes MEXEC, an implemented micro executive that compiles a device model that can have feedback into a structure for subsequent evaluation. This system computes both the most likely current device mode from n sets of sensor measurements and the n-1 step reconfiguration plan that is most likely to result in reaching a target mode - if such a plan exists. A user tunes the system by increasing n to improve system capability at the cost of real-time performance.
work offers evidence that distributed planning agents can greatly reduce communication costs by reasoning at abstract levels. While it is intuitive that improved search can reduce communication in such cases, there are other decisions about how to communicate plan information that greatly affect communication costs. This paper identifies cases independent of search where communicating at multiple levels of abstraction can exponentially decrease costs and where it can exponentially add costs. We conclude with a process for determining appropriate levels of communication based on characteristics of the domain.
This paper describes the automated mission planning system for MAMM, which dramatically reduced mission-planning costs to just a few workweeks, and enabled rapid generation of 'what-if' scenarios for evaluating mission-design trades. This latter capability informed several critical design decisions and was instrumental in accurately costing the mission.
This paper discusses a proof-of-concept prototype for ground-based automatic generation of validated rover command sequences from highlevel science and engineering activities. This prototype is based on ASPEN, the Automated Scheduling and Planning Environment. This Artificial Intelligence (AI) based planning and scheduling system will automatically generate a command sequence that will execute within resource constraints and satisfy flight rules.
Many traditional planners are built on top of existing execution engines that were not necessarily intended to be operated by a planner. The Mission Data System has been designed from the onset to have both an execution and planning engine and provides a framework for producing state-based models that can be used to coordinate planning and execution. The models provide a basis for ensuring the consistency of assumptions made by the execution engine and planner, and the frameworks provide a basis for run time communications between the planner and execution engines.
The Mars Reconnaissance Orbiter (MRO), launched on August 12, 2005, carries six science instruments, each with unique requirements for repetitive global monitoring, regional or global survey mapping, and/or targeted observations of Mars. Some prefer nadir-only observations, while other instruments require many off-nadir observations (especially for stereo viewing). Because the operations requirements are often incompatible, an interactive science planning process has been developed. This process is more complex than in some recent NASA Mars missions, but less complex (and more repetitive) than processes used by many large planetary missions. It takes full advantage of MRO's novel onboard processing capabilities, and uses simple electronic interactions between geographically distributed teams. This paper describes the process used during MRO's Primary Science Phase (PSP) to plan both interactive and non-interactive observations of Mars, and what has already been learned in the tests and rehearsals preparing for PSP.
With each new rover mission to Mars, rovers are traveling significantly longer distances. This distance increase raises not only the opportunities for science data collection, but also amplifies the amount of environment and rover state uncertainty that must be handled in rover operations. This paper describes how planning, scheduling and execution techniques can be used onboard a rover to autonomously generate and execute rover activities and in particular to handle new science opportunities that have been identified dynamically. We also discuss some of the particular challenges we face in supporting autonomous rover decision-making. These include interaction with rover navigation and path-planning software and handling large amounts of uncertainty in state and resource estimations. Finally, we describe our experiences in testing this work using several Mars rover prototypes in a realistic environment.