This installment of the AI in Space department looks at the process NASA uses to plan and execute an astronaut's "day in the life." As I do so, I will examine new planning and scheduling technologies recently presented at the Third International NASA Workshop on Planning and Scheduling for Space, held in Houston last October, and show how they might affect future Space Station astronauts. Pre-increment planning phaseStarting a year before the Space Station crew arrives, the pre-increment mission-planning phase involves accepting input from a variety of sources, including the international partners, public affairs department, and scientists with experimental payloads, among others. A detailed schedule of the crew's activities results. At the heart of all plan development is a single relational database at NASA Johnson Space Center (see Figure 1). A planning and scheduling software application called the consolidated planning system provides resource and constraint checking of the database's data. The CPS tracks such resources as crew, power, communications bandwidth, and consumables. The construction of a pre-increment plan is still largely a manual process, consuming huge numbers of work hours.Changing this process from a mostly manual to a more automated one will require significant advances in planning and scheduling software. Because humans will always be part of the planning process, especially to resolve political issues, any deployed system must exhibit mixedinitiative interaction, whereby automated software and human experts collaborate to create a plan. 1 During the recent NASA Workshop on Planning and Scheduling for Space, James Allen of the University of Rochester presented a paper on "Human-Machine Collaborative Planning" that discussed an explicit problem-solving level to mediate between the human-computer interaction and the underlying automated plan reasoners, bridging the gap between human and automated planning. 2 In his approach, mission planners would use natural language and graphical interfaces to input objectives they wish to achieve and constraints on possible solutions. Mission planners could build a plan incrementally, adding or subtracting constraints as necessary or suggesting potential solutions. The underlying plan reasoners would then decompose objectives into tasks, check constraints, and optimize resources, returning the incremental plan to the mission planners.In "Passat: A User-Centric Planning Framework," Karen Myers from SRI International discussed a framework that provides a set of plan editing and manipulation capabilities to support novice users in creating and modifying plans. 3 The Passat system builds upon a library of predefined templates that encode task networks describing standard operating procedures and previous cases. Users can select from these templates during plan development, with the system providing various forms of automated Previously in these pages I've made the observation that planning and scheduling technology often provides the core capability of a spa...
This paper summarizes some of the work performed by the NASA Exploration Team (NEXT) Human-Robotics Working Group. The team had representation from across NASA and performed a number of key studies and experiments between 2001 and early 2003.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.