Change search
ReferencesLink to record
Permanent link

Direct link
A Framework for Hierarchical Perception–Action Learning Utilizing Fuzzy Reasoning
University of Surrey, Guildford, U.K..
Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, The Institute of Technology.ORCID iD: 0000-0002-6096-3648
University of Surrey, Guildford, U.K..
2013 (English)In: IEEE transactions on systems, man and cybernetics. Part B. Cybernetics, ISSN 1083-4419, E-ISSN 1941-0492, Vol. 43, no 1, 155-169 p.Article in journal (Refereed) Published
Abstract [en]

Perception-action (P-A) learning is an approach to cognitive system building that seeks to reduce the complexity associated with conventional environment-representation/action-planning approaches. Instead, actions are directly mapped onto the perceptual transitions that they bring about, eliminating the need for intermediate representation and significantly reducing training requirements. We here set out a very general learning framework for cognitive systems in which online learning of the P-A mapping may be conducted within a symbolic processing context, so that complex contextual reasoning can influence the P-A mapping. In utilizing a variational calculus approach to define a suitable objective function, the P-A mapping can be treated as an online learning problem via gradient descent using partial derivatives. Our central theoretical result is to demonstrate top-down modulation of low-level perceptual confidences via the Jacobian of the higher levels of a subsumptive P-A hierarchy. Thus, the separation of the Jacobian as a multiplying factor between levels within the objective function naturally enables the integration of abstract symbolic manipulation in the form of fuzzy deductive logic into the P-A mapping learning. We experimentally demonstrate that the resulting framework achieves significantly better accuracy than using P-A learning without top-down modulation. We also demonstrate that it permits novel forms of context-dependent multilevel P-A mapping, applying the mechanism in the context of an intelligent driver assistance system.

Place, publisher, year, edition, pages
IEEE , 2013. Vol. 43, no 1, 155-169 p.
Keyword [en]
Autonomous agents, fuzzy logic (FL), hierarchical systems, machine learning, online learning, perception–action (P–A) learning, subsumption architectures, vehicle safety
National Category
Engineering and Technology
URN: urn:nbn:se:liu:diva-85688DOI: 10.1109/TSMCB.2012.2202109ISI: 000317643500013OAI: diva2:572598
Available from: 2012-11-28 Created: 2012-11-28 Last updated: 2016-05-04

Open Access in DiVA

fulltext(1565 kB)529 downloads
File information
File name FULLTEXT01.pdfFile size 1565 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Other links

Publisher's full text

Search in DiVA

By author/editor
Felsberg, Michael
By organisation
Computer VisionThe Institute of Technology
In the same journal
IEEE transactions on systems, man and cybernetics. Part B. Cybernetics
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
Total: 529 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Altmetric score

Total: 475 hits
ReferencesLink to record
Permanent link

Direct link