ALBERT

Description: A scheduling and resource management system named MAESTRO was interfaced with a Space Station Module Power Management and Distribution (SSM/PMAD) breadboard at MSFC. The combined system serves to illustrate the integration of planning, scheduling, and control in a realistic, complex domain. This paper briefly describes the functional elements of the combined system, including normal and contingency operational scenarios, then focusses on the method used by the scheduler to handle real-time contingencies.

Description: A set of viewgraphs on managing temporal relations in the MAESTRO scheduling system are given. The viewgraphs present information on such topics as why scheduling is hard, managing temporal relations, constraints on the placement of a single activity, constraints between activities, soft constraints (such as preferences), and contingency handling.

Description: The dark fine grained matrix of gas-rich ordinary chondrites replicates many of the physical, morphological, and spectral characteristics of the highly shocked and optically altered black chondrites. Spectral mixture modeling shows that the darkening and spectral attenuation seen in the dark matrix can be simulated with realistic mass fractions of light host material and black chondritic material. All these factors point to the conclusion that the dark matrix of gas-rich ordinary chondrites is dark due to the same processes that darkens black chondrites, shock distributed small particle size FeNi metal and troilite. Because the darkening is not seen in any of the non-gas-rich light portions and is only seen in the gas rich grains of the meteorite, the shock darkening would have to occur as part of the matrix's exposure to regolith processes. Since all gas-rich grains are darkened, it follows that darkening is not only common, but pervasive in asteroidal regoliths. These results imply that the upper, optically active layer of an ordinary chondrite parent body should have the spectral characteristics of a black chondrite, which are a dark, relatively featureless spectrum with modest red slope in the infrared. These are the characteristics of special type C asteroids.

Description: The Phobos 2 mission provided multispectral observations of Phobos over a large wavelength range and with relatively high spectral resolution. Here, researchers integrate results from three multispectral detectors by determining the ultraviolet-visible near infrared spectral properties of color and brightness features recognized in VSK TV images. Researchers present evidence that there are two fundamental spectral units within the region of overlapping coverage by the detectors. They describe the units' spectral and reflectance properties and discuss the implications of these results for the composition of Phobos.

Description: Every spacecraft, whether in orbit around the earth or an a deep space flight, has at its disposal limited amounts of the resources for it to accomplish its mission. Activity scheduling is currently a costly, human intensive task which requires a great deal of expertise. It belongs to a class of problems whose complexity increases exponentially with the number of operations. NASA has in the past accomplished this task by using a great deal of manpower, a large number of negotiating sessions, interminable bouts of phone tag, and mountains of paperwork. Lately the situation has improved with the introduction of automated scheduling techniques, but these to date still require expert involvement and fall short in some important ways. A prototype activity scheduler, MAESTRO, is introduced which is capable of meeting the needs of many NASA missions, eventually to include the Space Station. The approach to resource constrained scheduling is first discussed, then the intended domain for MAESTRO is described along with its design and current capabilities. A description of planned enhancements and revisions to the systems is also presented.

Description: Spacecraft activity scheduling was a focus of attention in artificial intelligence recently. Several scheduling systems were devised which more-or-less successfully address various aspects of the activity scheduling problem, though most of these are not yet mature, with the notable expection of NASA's ESP. Few current scheduling systems, however, make any attempt to deal fully with the problem of modifying a schedule in near-real-time in the event of contingencies which may arise during schedule execution. These contingencies can include resources becoming unavailable unpredictably, a change in spacecraft conditions or environment, or the need to perform an activity not scheduled. In these cases it becomes necessary to repair an existing schedule, disrupting ongoing operations as little as possible. Normal scheduling is just a part of that which must be accomplished during contingency rescheduling. A prototype system named MAESTRO was developed for spacecraft activity scheduling. MAESTRO is briefly described with a focus on recent work in the area of real-time contingency handling. Included is a discussion of some of the complexities of the scheduling problem and how they affect contingency rescheduling, such as temporal constraints between activities, activities which may be interrupted and continued in any of several ways, and different ways to choose a resource complement which will allow continuation of an activity. Various heuristics used in MAESTRO for contingency rescheduling is discussed, as are operational concerns such as interaction of the scheduler with spacecraft subsystems controllers.

Description: By combining the best of automated and human decision-making in scheduling many advantages can accrue. The joint performance of the user and system is potentially much better than either alone. Features of the MAESTRO scheduling system serve to illustrate concepts of user/software cooperation. MAESTRO may be operated at a user-determinable and dynamic level of autonomy. Because the system allows so much flexibility in the allocation of decision-making responsibilities, and provides users with a wealth of information and other support for their own decision-making, better overall schedules may result.

Description: The relationship between ordinary chondrites and S-type asteroids is an unresolved issue in meteorite science. S-type asteroids exhibit a positively red-sloped spectrum that is interpreted to indicate the presence of elemental iron on the surfaces. The characteristic red-sloped spectrum of iron-rich meteorites is produced by only the specular component of the reflectance. Complex metallic surfaces can be modeled as linear mixtures of specular and nonspecular components. It is the geometry of the metal on a surface and its interaction with surrounding material, rather than the absolute amount of metal, that determine the redness of resulting spectra. In order to distinguish between ordinary chondrite and differentiated parent bodies it is important to understand how regolith processes affect the nature and form of metal on asteroid surfaces.

Description: A prototype scheduling system named MAESTRO currently under development is being used to explore possible approaches to the spacecraft operations scheduling problem. Results indicate that the appropriate combination of heuristic and other techniques can provide an acceptable solution to the scheduling problem over a wide range of operational scenarios and management approaches. These can include centralized or distributed instrument or systems control, batch or incremental scheduling, scheduling loose resource envelopes or exact profiles, and scheduling with varying degrees of user intervention. Techniques used within MAESTRO to provide this flexibility and power include constraint propagation mechanisms, multiple asynchronous processes, prioritized transaction-based command management, resource opportunity calculation, user-alterable selection and placement mechanisms, and maintenance of multiple schedules and resource profiles. These techniques and scheduling complexities requiring them are discussed.

Description: Artificial Intelligence (AI) techniques provide good solutions for many of the problems which are characteristic of scheduling applications. However, scheduling is a large, complex heterogeneous problem. Different applications will require different solutions. Any individual application will require the use of a variety of techniques, including both AI and conventional software methods. The operational context of the scheduling system will also play a large role in design considerations. The key is to identify those places where a specific AI technique is in fact the preferable solution, and to integrate that technique into the overall architecture.