NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Collections About News Help Login
Back to Results
Autoplan: A self-processing network model for an extended blocks world planning environmentSelf-processing network models (neural/connectionist models, marker passing/message passing networks, etc.) are currently undergoing intense investigation for a variety of information processing applications. These models are potentially very powerful in that they support a large amount of explicit parallel processing, and they cleanly integrate high level and low level information processing. However they are currently limited by a lack of understanding of how to apply them effectively in many application areas. The formulation of self-processing network methods for dynamic, reactive planning is studied. The long-term goal is to formulate robust, computationally effective information processing methods for the distributed control of semiautonomous exploration systems, e.g., the Mars Rover. The current research effort is focusing on hierarchical plan generation, execution and revision through local operations in an extended blocks world environment. This scenario involves many challenging features that would be encountered in a real planning and control environment: multiple simultaneous goals, parallel as well as sequential action execution, action sequencing determined not only by goals and their interactions but also by limited resources (e.g., three tasks, two acting agents), need to interpret unanticipated events and react appropriately through replanning, etc.
Document ID
19900017210
Acquisition Source
Legacy CDMS
Document Type
Contractor Report (CR)
Authors
Dautrechy, C. Lynne (Maryland Univ. College Park, MD, United States)
Reggia, James A. (Maryland Univ. College Park, MD, United States)
Mcfadden, Frank (Maryland Univ. College Park, MD, United States)
Date Acquired
September 6, 2013
Publication Date
June 1, 1990
Subject Category
Computer Programming And Software
Report/Patent Number
UMIACS-TR-90-78
CS-TR-2483
NASA-CR-186733
NAS 1.26:186733
Accession Number
90N26526
Funding Number(s)
CONTRACT_GRANT: NAG1-885
CONTRACT_GRANT: NSF IRI-84-51430
Distribution Limits
Public
Copyright
Work of the US Gov. Public Use Permitted.

Related Records

There are no records associated with this record.
No Preview Available