PRS (Procedural Reasoning System) was developed in the Artificial Intelligence Center at SRI International, Menlo Park, CA, USA, during the 1980s, by the research group including Michael Peter Georgeff, Amy L. Lansky, and François Félix Ingrand, et al., as a framework responsible for exploiting and popularizing the BDI model in software for control of an intelligent agent.
PRS-CL: A Procedural Reasoning System | See also other projects.
Australian Artificial Intelligence Institute (AAII)
Carlton, Victoria, Australia
E-mail: Этот адрес электронной почты защищён от спам-ботов. У вас должен быть включен JavaScript для просмотра.
Michael Peter Georgeff is a computer scientist and entrepreneur who has made contributions in the areas of Intelligent Software Agents and eHealth. Georgeff is a former program director in the Artificial Intelligence Center at SRI International, Menlo Park, California, and former director of the Australian Artificial Intelligence Institute Ltd., at the University of Melbourne.
Myers, Karen Program Director & Principal Scientist:
PRS-CL: A Procedural Reasoning System
Intelligent Mixed-initiative Planning and Control Technologies (IMPACT)
Artificial Intelligence Center, SRI International, Menlo Park, CA, USA
Email: Этот адрес электронной почты защищён от спам-ботов. У вас должен быть включен JavaScript для просмотра. | profile on LinkedIn
Общие сведения
Когнитивная архитектура PRS разработана в первоначальном варианте исследовательской группой под руководством Майкла Георгиева (Michael P. Georgeff) в Центре искусственного интеллекта (Artificial Intelligence Center, SRI International, Менло-Парк, Калифорния, США) при Стенфордском университете.
Архитектура используется в серии других проектов (смотри ниже).
Архитектура PRS (Procedural Reasoning System) [Ingrand, Georgeff, & Rao, 1992] была одной из первых архитектур, разработанных в парадигме BDI (Beliefs–Desires–Intentions) "Намерения-Желания-Убеждения".
Эта структура хранит иерархические процедуры с условиями, эффектами и предписанными шагами, которые вызывают подпроцедуры.
Динамические структуры памяти включают в себя убеждения об окружающей среде, желания, которые агент хочет достигнуть, и намерения, которые агент планирует выполнить.
В каждом цикле система PRS решает, продолжить ли исполнять ее текущее намерение или выбрать новое намерение, которому необходимо следовать.
Description
The development of systems capable of handling and diagnosing malfunctions in real time has long been of considerable practical importance.
Procedural Reasoning System (PRS) ongoing development and application of PRS in collaboration with SRI International.
The PRS concept was developed by the Artificial Intelligence Center at SRI International during the 1980s, by many workers including Michael Georgeff, Amy L. Lansky, and François Félix Ingrand.
The seminal application of the framework was a fault detection system for the reaction control system of the NASA Space Shuttle Discovery. Development on this PRS continued at the Australian Artificial Intelligence Institute through to the late 1990s, which lead to the development of a C++ implementation and extension called dMARS.
The Procedural Reasoning System (PRS) is a system for controlling and carrying out the high-level reasoning of a robot that combines traditional means-end reasoning with the abilities to react to unanticipated events and to change goals and intentions as situations warrant. PRS has been used to control the movements of a real robot operating in an single-agent, indoor navigation domain.
A PRS agent comprises a set of changing beliefs (facts about the world), a set of current goals or desires, a set of procedural plan schemas or KnowledgeAreas (KAs) which describe how to achieve its goals and how to react to particular events, an interpreter for manipulating each of these components, and a process stack of currently active KAs or intentions. PRS operates as a partial, hierarchical planner, its interpreter permitting the interleaving of planning and execution. As KA interpretation is time-bounded, a guaranteed level of responsiveness can always be maintained.
The set of KAs also includes metalevel KAs. Interpreted in the same manner as regular KAs, these metalevel routines contain application-specific knowledge which instruct the agent how to manipulate its own beliefs, desires, and intentions; how to prioritise and select among conflicting KAs; and how to make best use of the available resources given the system’s changing realtime constraints. The agent’s beliefs and goals determine which KAs are to be considered for execution and, as KAs are executed, new subgoals will be posted and new beliefs derived.
The Procedural Reasoning System (PRS) is based on the notion of a rational agent that can reason and plan under possibly stringent constraints on both time and information. This approach provides the system with the ability to reason in complex ways about dynamic processes, while still maintaining the reactivity required to ensure appropriate responsiveness and control. By considering two large-scale applications in aerospace and telecommunications, it is shown how PRS meets many of the critical requirements for real-time malfunction, handling and diagnostic systems.
1) a Database containing current facts and beliefs about the world, represented using first order predicate calculus,
2) a Set of Goals to be realized by the system as conditions over an interval of time on internal and external state descriptions (desires).
3) a set of Plans or procedures describing how certain sequences of conditional tests and actions may be performed to achieve certain goals or to react to certain situations - the Knowledge areas (KAs) that define sequences of low-level actions toward achieving a goal in specific situations, and Intentions that include those KAs that have been selected for current and eventual execution.
4) an Interpreter or inference mechanism that manipulates these components to select and execute appropriate plans for achieving the system's goals.
A system built in PRS-CL is intended to simultaneously achieve any goals it might have based on its current beliefs about the world while noticing and responding to new events.
See User Guide for the Procedural Reasoning System, K. L. Myers, Technical Report, Artificial Intelligence Center, SRI International, Menlo Park, CA, 1997. [postscript] [PDF]
See also other related projects:
Continuous Planning and Execution Automated Capture of Design Rationale
Multiagent Planning Architecture (MPA)
Mixed-initiative Planning and Scheduling for Science Missions
Taskable Reactive Agent Communities
Distributed Multi-Agent Reasoning System (dMARS) an agent-oriented development and implementation environment for building complex, distributed, time-critical systems. Developed as a C++ extension to PRS.
PRS-CL: A Procedural Reasoning System An extension to PRS maintained by SRI International
Selected publications
View All publicationsSelected Publications of Karen L. Myers
A Procedural Knowledge Approach to Task-Level Control, K. L. Myers, in Proceedings of the Third International Conference on AI Planning Systems, 1996.
Ingrand, F. F., Georgeff, M. P., & Rao, A. S. (1992). An architecture for real-time reasoning and system control. IEEE Expert, 7, 34–44.
Georgeff, M. P.; F. F. Ingrand (1990). "Real-time reasoning: the monitoring and control of spacecraft systems". Proceedings of the sixth conference on Artificial intelligence applications. pp. 198–204.
Georgeff, M. P., Lansky, A. L. Reactive Reasoning and Planning, in AAAI-87 Proceedings, American Association of Artificial Intelligence, pp. 677-682, 1987. [PDF, Details]
Georgeff, Michael; Bonollo, Umberto. Procedural Expert Systems, Technical Note 314. AI Center, SRI International, 333 Ravenswood Ave., Menlo Park, CA 94025, Dec 1983. [PDF, Details]
