Richard Doyle - Profile on Academia.edu (original) (raw)

Papers by Richard Doyle

In real-time monitoring situations, more information is not necessarily better. When faced with c... more In real-time monitoring situations, more information is not necessarily better. When faced with complex emergency situations, operators can experience information overload and a compromising of their ability to react quickly and correctly. We describe an approach to focusing operator attention in real-time systems monitoring based on a set of empirical and model-based measures for detecting different kinds of anomalies and for determining the relative importance of sensor data. This approach has been evaluated on data from the life support system testbed of Space Station Freedom.

PIPE: An intelligent scientific data preparation assistant

AIP Conference Proceedings, 1993

Scientific data preparation is the process of extracting usable scientific data from raw instrume... more Scientific data preparation is the process of extracting usable scientific data from raw instrument data. This task involves noise detection (and subsequent noise classification and flagging or removal), extracting data from compressed forms, and construction of derivative or aggregate data (e.g., spectral densities or running averages). This paper describes the PIPE system. PIPE provides intelligent assistance developing scientific data preparation

Sensor placement for diagnosability in space-borne systems - A model-based reasoning approach

30th Aerospace Sciences Meeting and Exhibit, 1992

... Authors: Chien, Steve; Doyle, Richard; Rouquette, Nicolas. Affiliation: AA(JPL, Pasadena, CA)... more ... Authors: Chien, Steve; Doyle, Richard; Rouquette, Nicolas. Affiliation: AA(JPL, Pasadena, CA), AB(JPL, Pasadena, CA), AC(JPL, Pasadena, CA). Publication: AIAA, Aerospace Sciences Meeting and Exhibit, 30th, Reno, NV, Jan. 6-9, 1992. 10 p. Publication Date: 01/1992. ...

9th Computing in Aerospace Conference, 1993

McI)oi~l)cll-Dot]glas Aerospace lloustoll, TX 77059-3556 Abstract Monitoring of c.omplcx continuo... more McI)oi~l)cll-Dot]glas Aerospace lloustoll, TX 77059-3556 Abstract Monitoring of c.omplcx continuous physical systems has been traditionally accomplished in compulcr-based process control soft ware by onc or bolt] ofthcfollowing rnctbods: l)cslal~lisllit] glit]~it checks for sensors and raising an alarm whcncvcr a sensor'svaluccrosscs oi]cofttlcsc tllrcsl]olds, :i!l{l 2) comparing prcdickxt vahrcs from a simulation toaclual scnsorvalucs and flagging discrcpancics. 'lhcsc anomaly dclcction techniques ,arc not as robusl as lhcynccd lobe. Failures can manifcsl in ways which arc nol captured by lhcsc Iradilional mcihods. Fur(tlcrn]orc,son] canon] alousbchwiors arc more naturally dclcckd at lhc lCVCI of global interactions affccfing mulliplc sensors, Wc dcscribc cxlcnsions 10 lhc traditional lcchniqucs for anomaly dclcclion, as WCII as ncw anomaly dc(cc(ion techniques based on altcrna(c nmdc]s of what dis(inguishcs normal from abnormal bchfivior. Somcof thcsc(cchniqucs arcdcsigncd (o cap-Iurcanomlics at individual sensors; sorncdclccl anomalies across collcdionsof sensors. 10 assist in rcasrming about complex global behaviors, wc cons[rucl and simulate a causal model of Ihc physical syslcm being n]onitorcd. These Icchniqucs bavc been Icskd on da(a from lhc hvironmcnlal Control and Life Support Syslcm (lX1.

New directions in science-enabling autonomy for planetary missions

1997 IEEE Aerospace Conference, 1997

Autonomy concepts now exert a powerful influence on the development of a wide range of aerospace ... more Autonomy concepts now exert a powerful influence on the development of a wide range of aerospace systems. Their value for automating spacecraft functions such as guidance, navigation and control is well-recognized. However, relatively little effort has been devoted in the past to their use in the context of payload data processing and understanding. In this paper, we focus on the

[1991] Proceedings. The Second Annual Conference on AI, Simulation and Planning in High Autonomy Systems, 1991

For complex systems with large sensor complements, we develop a selective monitoring strategy to ... more For complex systems with large sensor complements, we develop a selective monitoring strategy to avoid information overload on system operators. We describe an approach to determining from moment to moment which subset of the available sensor data for a system is most informative about the state of the system and about interactions occurring within the system. We term this process sensor selection. Our approach draws on concepts from causal reasoning and information theory. After describing our NASA test domain, we conclude with a report on the status of our implementation of the SELMON system. 154 0-8186-2162-1/91/0000/0154$01 .OO 0 1991 IEEE

Machine Learning In Engineering Automation

Machine Learning Proceedings 1991, 1991

--Autonomy,concepts,now,exert a powerful,influcncc on,the dcvcloprncnt of,a wide range of,aerospa... more --Autonomy,concepts,now,exert a powerful,influcncc on,the dcvcloprncnt of,a wide range of,aerospace systems. Their value for automating,spacecraft,ftmctions such,as guidance, navigation and control is , recognized. I lowcvcr, relatively little effort has been,devoted,in the past,to their usc in the context of payload data processing,and understanding. In this paper, we focus on the usc,of autonomy,to enable,and,enhance scienti~lc goals,for spaceborne,missions. We argue that the general notions of autonomy can bc applied,directly to a broad,range of scientific problems,in ways that have,never before been considered.Wc begin by outlining our general philosophy, and by describing the ways,in which,autonomy,can,bc used,t o transform,the ways,in which,spaceborne science is conducted. This is followed by a description of two novel systems that wc have developed,to exploit this philosophy,for planetaly missions. 1. IN1RODUCTION The development,of autonomous,spacecraft

MESA: An Interactive Modeling and Simulation Environment for Intelligent Systems Automations

Hawaii International Conference on System Sciences, 1994

Describes MESA, a domain-independent interactive tool for the development of reusable causal mode... more Describes MESA, a domain-independent interactive tool for the development of reusable causal models and model-based-reasoning applications. Our current efforts are focused on developing automated sensor monitoring applications for NASA flight projects. MESA supports model development from a component-centered approach and provides a graphical editor for rapidly prototyping components and connections from a small set of modeling primitives that describe structure

The Future of AI in Space

IEEE Intelligent Systems, 2006

Recent AI history in space Flight technology experiments are focused activities that support NASA... more Recent AI history in space Flight technology experiments are focused activities that support NASA&amp;amp;amp;amp;#x27;s pursuit of the capabilities it needs to explore space frontiers. The experiments validate new spacecraft capabilities, exploring their potential and value, and mitigating the inherent risks. ...

IEEE Intelligent Systems, 1999

The Autonomous Sciencecraft Experiment flight demonstration (ASE) will fly onboard the US Air For... more The Autonomous Sciencecraft Experiment flight demonstration (ASE) will fly onboard the US Air Force's TechSat-21 constellation, an unclassified mission scheduled for launch in 2004. ASE will use onboard science analysis, replanning, robust execution, and formation flying to radically increase science return by enabling intelligent downlink selection and autonomous retargeting. Demonstration of these capabilities in a flight environment will open up tremendous new opportunities in planetary science, space physics, and earth science that would be unreachable without this technology.

situations, more information is not necessarily better.

Acta Astronautica, 2006

Space exploration missions are undergoing a significant transformation as are the expectations of... more Space exploration missions are undergoing a significant transformation as are the expectations of their scientific investigators and the public who participate in these great voyages of exploration. The early reconnaissance missions are giving way to a new data-intensive era of long duration observational outposts, landed vehicles, sample returns, and multi-spacecraft fleets and constellations. Mars exploration has already become a special case of the new operational mode; other destinations will follow. These changes will require orders of magnitude increases in data rates, highly automated and standardized data communications between the remote locations and Earth, more transparent and responsive mission operations procedures, and the ability to engage the public by giving them Intemet-based visibility into the missions as they unfold. The new area will demand a new paradigm for the Deep Space Network, with increased emphasis on data networking and the data processing applications that allow users to become more intimately engaged with the conduct of the mission. We call this new paradigm the Interplanetary Network. Its vision is seamless connectivity between scientists and their instruments, new data analysis and visualization tools that will greatly enhance and enable new modes of space exploration, and the involvement of the public via web-based "telepresence."

Journal of Aerospace Computing, Information, and Communication, 2009

Aerospace technology, driven by military and civilian applications, is increasingly moving toward... more Aerospace technology, driven by military and civilian applications, is increasingly moving towards a new generation of sophisticated manned and unmanned systems.As warfare gets asymmetric and uncertain, and space exploration more bold and daring, the desire is for machines to emulate a human's ability to adapt quickly and effectively.