Artificial intelligence in robot control systems

2nd International Conference "Cognitive Robotics"

工业技术;无线电电子学;计算机科学

Korikov, A.^1,2

Tomsk State University of Control Systems and Radioelectronics, Tomsk, Russia^1

Tomsk Polytechnic University, Tomsk, Russia^2

关键词: Artificial intelligence systems;    Computing technology;    High dimensionality;    Intelligent robot controls;    Optimization method;    Priori uncertainty;    Robot control systems;    Robotic control system;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/363/1/012013/pdf
DOI  :  10.1088/1757-899X/363/1/012013

来源: IOP

【 摘 要 】

This paper analyzes modern concepts of artificial intelligence and known definitions of the term »level of intelligence». In robotics artificial intelligence system is defined as a system that works intelligently and optimally. The author proposes to use optimization methods for the design of intelligent robot control systems. The article provides the formalization of problems of robotic control system design, as a class of extremum problems with constraints. Solving these problems is rather complicated due to the high dimensionality, polymodality and a priori uncertainty. Decomposition of the extremum problems according to the method, suggested by the author, allows reducing them into a sequence of simpler problems, that can be successfully solved by modern computing technology. Several possible approaches to solving such problems are considered in the article.