Perceptual control theory: Difference between revisions

Revision as of 00:28, 2 November 2024

Latest revision as of 23:29, 18 June 2025

(4 intermediate revisions by 4 users not shown)

Line 10:

==History==

PCT has roots in the 19th-century physiological insights of [[Claude Bernard]] and in 20ththe 20th-century inresearch of Peter Putnam<ref name="Gefter 2025">{{cite web | last=Gefter | first=Amanda | title=Finding Peter Putnam | website=Nautilus | date=June 17, 2025 | url=https://nautil.us/finding-peter-putnam-1218035/ | access-date=June 18, 2025}}</ref><ref name="Putnam Fuller 1964">{{cite web | title=Outline of a Functional Model of the researchNervous bySystem, [[WalterPutnam/Fuller Bradford1964 Cannon|Walterfirst1=Peter|last1=Putnam|last2=Fuller|first2=Robert| Bwebsite=The Peter Putnam Papers | date=October 30, 1970 | url=https://www.peterputnam.org/outline-of-a-functional-model-of-the-nervous-system-putnam/fuller-1964 | access-date=June 18, 2025}}</ref> and [[Walter Bradford Cannon]], and in the fields of [[control systems engineering]] and [[cybernetics]]. Classical negative feedback control was worked out by engineers in the 1930s and 1940s,<ref name="black93">Harold Black and the Negative-Feedback Amplifier, Ronald Kline, IEEE Control Systems Magazine, Aug 1993, Volume 13, Issue 4, Pages 82-85</ref><ref name=Bennett>{{cite journal | last =Bennett | first =Stuart | title =A brief history of automatic control | journal =IEEE Control Systems Magazine | volume =16 | issue =3 | pages =17–25 | date =June 1996 | url =http://ieeecss.org/CSM/library/1996/june1996/02-HistoryofAutoCtrl.pdf | doi =10.1109/37.506394 | access-date =18 July 2016 | archive-url =https://web.archive.org/web/20160809050823/http://ieeecss.org/CSM/library/1996/june1996/02-HistoryofAutoCtrl.pdf | archive-date =9 August 2016 | url-status =dead }}</ref> and further developed by [[Norbert Wiener|Wiener]],<ref name=Wiener48>{{cite book | title =Cybernetics: Or Control and Communication in the Animal and the Machine | publisher =Hermann & Cie | date =1948 | ___location =Paris }} 2nd revised ed. 1961, MIT Press, Cambridge, MA. {{ISBN|978-0-262-73009-9}}.</ref> [[William Ross Ashby|Ashby]],<ref name=Ashby.DFB>{{cite book | last =Ashby | first =William Ross | title =Design for a Brain | publisher =Chapman & Hall | date =1952 | ___location =London | url =https://archive.org/details/designforbrainor00ashb }}</ref> and others in the early development of the field of [[cybernetics]]. Beginning in the 1950s, [[William T. Powers]] applied the concepts and methods of engineered control systems to biological control systems, and developed the experimental methodology of PCT.<ref name=Runkel.cast>{{cite book | last =Runkel | first =Philip J. | title =Casting nets and testing specimens: Two grand methods of psychology | publisher =Praeger | date =1990 | ___location =New York | page =103 | isbn =978-0-275-93533-7 }}</ref><ref name=Cziko.TWD>{{Citation | last =Cziko | first =Gary | title =The things we do: Using the lessons of Bernard and Darwin to understand the what, how, and why of our behavior | place =Cambridge, MA | publisher =MIT Press | year =2000 | page =[https://archive.org/details/thingswedousingl0000czik/page/9 9] | isbn =978-0-262-03277-3 | url =https://archive.org/details/thingswedousingl0000czik/page/9 }}</ref>

A key insight of PCT is that the controlled variable is not the output of the system (the behavioral actions), but its input, that is, a sensed and transformed function of some state of the environment that the control system's output can affect. Because these sensed and transformed inputs may appear as consciously perceived aspects of the environment, Powers labelled the controlled variable "perception". The theory came to be known as "Perceptual Control Theory" or PCT rather than "Control Theory Applied to Psychology" because control theorists often assert or assume that it is the system's output that is controlled.<ref name=Astrom>{{cite book | last1 =Astrom | first1 =Karl J. | last2 =Murray | first2 =Richard M. | title =Feedback Systems: An Introduction for Scientists and Engineers | publisher =Princeton University Press | date =2008 | url =http://www.cds.caltech.edu/~murray/books/AM08/pdf/am08-complete_28Sep12.pdf | isbn =978-0-691-13576-2 }}</ref> In PCT it is the internal representation of the state of some variable in the environment—a "perception" in everyday language—that is controlled.<ref>For additional information about the history of PCT, see:

Line 138:

==Robotics==

PCT has significant implications for Robotics and Artificial Intelligence. W.T. Powers introduced the application of PCT to robotics in 1978, early in the availability of home computers.<ref>{{cite magazine |last=Powers |first=William T. |date=1978 |title=The Nature of Robots: Part I: Defining Behavior |url=http://www.livingcontrolsystems.com/appendixes/byte_june_1979.pdf |url-status=dead |magazine=Byte: The Small Systems Journal |___location=Peterborough, NH |publisher=McGraw-Hill |issue=4 |pages=132-144132–144 |archive-url=https://web.archive.org/web/20070604100935/http://www.livingcontrolsystems.com/appendixes/byte_june_1979.pdf |archive-date=June 4, 2007 |access-date=}}</ref>

<ref>{{cite magazine |last=Powers |first=William T. |date=1978 |title=The Nature of Robots: Part II: Simulated Control Systems |url=http://www.livingcontrolsystems.com/appendixes/byte_july_1979.pdf |url-status=dead |magazine=Byte: The Small Systems Journal |___location=Peterborough, NH |publisher=McGraw-Hill |issue=4 |pages=134-152134–152 |archive-url=https://web.archive.org/web/20070604100935/http://www.livingcontrolsystems.com/appendixes/byte_july_1979.pdf |archive-date=June 4, 2007 |access-date=}}</ref>

<ref>{{cite magazine |last=Powers |first=William T. |date=1978 |title=The Nature of Robots: Part III: A closer look at human behavior |url=http://www.livingcontrolsystems.com/appendixes/byte_aug_1979.pdf |url-status=dead |magazine=Byte: The Small Systems Journal |___location=Peterborough, NH |publisher=McGraw-Hill |issue=4 |pages=94-11694–116 |archive-url=https://web.archive.org/web/20070604100935/http://www.livingcontrolsystems.com/appendixes/byte_aug_1979.pdf" |archive-date=June 4, 2007 |access-date=}}</ref>

<ref>{{cite magazine |last=Powers |first=William T. |date=1978 |title=The Nature of Robots: Part IV: Looking for controlled variables |url=http://www.livingcontrolsystems.com/appendixes/byte_sep_1979.pdf |url-status=dead |magazine=Byte: The Small Systems Journal |___location=Peterborough, NH |publisher=McGraw-Hill |issue=4 |pages=96-11296–112 |archive-url=https://web.archive.org/web/20070604100935/http://www.livingcontrolsystems.com/appendixes/byte_sep_1979.pdf |archive-date=June 4, 2007 |access-date=}}</ref> The comparatively simple architecture,<ref name=Young2017>{{cite journal | last =Young| first =Rupert | author-link = | title =A General Architecture for Robotics Systems: A Perception-Based Approach to Artificial Life. | journal =Artificial Life| volume =23 | issue =2 | pages =236–286 | date =Jun 2017 | access-date = | doi =10.1162/ARTL_a_00229 | pmid =28513206 }}</ref> a hierarchy of perceptual controllers, has no need for complex models of the external world, inverse kinematics, or computation from input-output mappings. Traditional approaches to robotics generally depend upon the computation of actions in a constrained environment. Robots designed this way are inflexible and clumsy, unable to cope with the dynamic nature of the real world. PCT robots inherently resist and counter the chaotic, unpredictable disturbances to their controlled inputs which occur in an unconstrained environment. The PCT robotics architecture has recently been applied to a number of real-world robotic systems including robotic rovers,<ref>

{{cite AV media

| people =Young, Rupert

Line 167:

| publisher =Perceptual Robots

}}

</ref> Some commercially available robots which demonstrate good control in a naturalistic environment use a control-theoretic architecture which requires much more intensive computation. For example, Boston Dynamics has said<ref>{{Cite web |url=https://bostondynamics.com/blog/starting-on-the-right-foot-with-reinforcement-learning/#:~:text=The%20legged%20robots%20we've,to%20take%20in%20the%20moment |title=Starting on the Right Foot with Reinforcement Learning |author=<!--Not stated--> |date=March 19, 2024 |website=https://bostondynamics.com/ |publisher=Boston Dynamics |access-date=November 1, 2024 |quote=The legged robots we’ve built at Boston Dynamics have historically leveraged Model Predictive Control (MPC), a control strategy that predicts and optimizes the future states of the robot in order to decide what action to take in the moment.}}</ref> that its robots use historically leveraged [[Model_predictive_control |model predictive control]].

==Current situation and prospects==

Line 177:

While many computer demonstrations of principles have been developed, the proposed higher levels are difficult to model because too little is known about how the brain works at these levels. Isolated higher-level control processes can be investigated, but models of an extensive hierarchy of control are still only conceptual, or at best rudimentary.

Perceptual control theory has not been widely accepted in mainstream psychology, but has been effectively used in a considerable range of domains<ref>{{Cite web | url=https://thepsychologist.bps.org.uk/volume-28/november-2015/perceptual-control-revolution |title = A perceptual control revolution? |website=The Psychologist}}</ref><ref name="IJHCS">The June 1999 Issue of ''The International Journal of Human-Computer Studies'' contained papers ranging from tracking through cockpit layout to self-image and crowd dynamics.</ref> in human factors,<ref name="CBUT">PCT lies at the foundation of [[Component-Based Usability Testing]].</ref> clinical psychology, and psychotherapy (the "[[Method of Levels]]"), it is the basis for a considerable body of research in sociology,<ref>For example: McClelland, Kent A. and Thomas J. Fararo, eds. 2006, ''Purpose, Meaning and Action: Control Systems Theories in Sociology'', New York: Palgrave Macmillan. (McClelland is co-author of Chapter 1, "Control Systems Thinking in Sociological Theory," and author of Chapter 2, "Understanding Collective Control Processes."). McClelland, Kent, 2004, "Collective Control of Perception: Constructing Order from Conflict", ''International Journal of Human-Computer Studies'' 60:65-99. McPhail, Clark. 1991, ''The myth of the madding crowd'' New York: Aldine de Gruyter.</ref> and it has formed the conceptual foundation for the reference model used by a succession of [[NATO]] research study groups.<ref name="IST">volume-28november-2015 Reports of these groups are available from the [[NATO Research and Technology Administration]] publications page: {{cite web |url=http://www.rta.nato.int/Abstracts.aspx |title=NATO Research & Technology Organisation Scientific Publications |access-dateurl=2010-05-15http://www.rta.nato.int/Abstracts.aspx |url-status=dead |archive-url=https://web.archive.org/web/20100623055236/http://www.rta.nato.int/abstracts.aspx |archive-date=2010-06-23 |access-date=2010-05-15}}> under the titles RTO-TR-030, RTO-TR-IST-021, and RTO-TR-IST-059.</ref>

Recent approaches use principles of perceptual control theory to provide new algorithmic foundations for [[artificial intelligence]] and [[machine learning]].<ref>{{cite journal |last1=Monaco |first1=Joseph D. |last2=Hwang |first2=Grace M. |title=Neurodynamical Computing at the Information Boundaries of Intelligent Systems |journal=Cognitive Computation |date=27 December 2022 |volume=16 |issue=5 |pages=1–13 |doi=10.1007/s12559-022-10081-9|s2cid=255222711 |doi-access=free |pmid=39129840 |pmc=11306504 }}</ref>

Line 244:

[[Category:Cybernetics]]

[[Category:Formal sciences]]

[[Category:Robotics engineering]]