Cognitive robotics: Difference between revisions

Revision as of 05:47, 2 August 2025

Latest revision as of 05:48, 2 August 2025

(One intermediate revision by the same user not shown)

Line 26:

A more complex learning approach is "autonomous [[knowledge acquisition]]": the robot is left to explore the environment on its own. A system of goals and beliefs is typically assumed.

A somewhat more directed mode of exploration can be achieved by "curiosity" algorithms, such as Intelligent Adaptive Curiosity<ref>http://www.pyoudeyer.com/ims.pdf {{Bare URL PDF|date=March 2022}}</ref><ref>http://www.pyoudeyer.com/oudeyer-kaplan-neurorobotics.pdf {{Bare URL PDF|date=March 2022}}</ref> or Category-Based Intrinsic Motivation.<ref>http://science.slc.edu/~jmarshall/papers/cbim-epirob09.pdf {{Bare URL PDF|date=March 2022}}</ref> These algorithms generally involve breaking sensory input into a finite number of categories and assigning some sort of prediction system (such as an [[Artificialartificial Neuralneural Networknetwork]]) to each. The prediction system keeps track of the error in its predictions over time. Reduction in prediction error is considered learning. The robot then preferentially explores categories in which it is learning (or reducing prediction error) the fastest.

==Other architectures==

Line 53:

== References ==

{{Reflist}}

== Sources ==

*[https://web.archive.org/web/20090314232056/http://www.ss-rics.org/ The Symbolic and Subsymbolic Robotic Intelligence Control System (SS-RICS)]

*[https://web.archive.org/web/20060617202124/http://www.cs.uu.nl/groups/IS/robotics/robotics.html Intelligent Systems Group - University of Utrecht]