|
News of the Kabardin-Balkar scientific center of RAS, 2017, Issue 6-2, Pages 43–53
(Mi izkab221)
|
|
|
|
This article is cited in 3 scientific papers (total in 3 papers)
COMPUTER SCIENCE. CALCULATION EQUIPMENT. MANAGEMENT
On one approach to measurement
of robotic complexes autonomy and intelligence
O. O. Varlamovabc a Bauman Moscow State Technical University (National research university of technology), 105005, Moscow, ul. Baumanskaya 2-ya, 5
b Moscow Automobile and Road Construction State Technical University (MADI), 125319, Moscow, 64, Leningradsky prospect
c Scientific Research Institute MIVAR,127521, Moscow, Oktyabrskaya street, 72
Abstract:
With the development of artificial intelligence (AI), the problem of measuring the degree of robotic
complexes autonomy has been arisen. Robotic complexes autonomy is determined by their control systems
intelligence, i.e. the ability of robotic complexes to react and make decisions in various situations. The
problem of measuring robotic complexes intelligence is related to the assessment of decision-making
algorithms complexity. This problem is studied in the sphere of «expert systems» using the metric «variables, rule». This approach is based on estimating the maximum possible combinations numbers in a
situation determined by the factorial of decision-making rules numbers. So the evaluation criterion is «the
worst case of complexity» with a complete search of all possible rules describing the situation. These rules
describe possible actions of robotic complexes in a certain situation, and all possible solutions are
determined by the factorial depending from the number of rules ($N!$).
Keywords:
artificial intelligence, robots, mivar, autonomy measurement, intelligence measurement,
decision making systems, behaviour, algorithms computational complexity, аctive reflection, logical
reasoning systems, expert systems.
Received: 25.10.2017
Citation:
O. O. Varlamov, “On one approach to measurement
of robotic complexes autonomy and intelligence”, News of the Kabardin-Balkar scientific center of RAS, 2017, no. 6-2, 43–53
Linking options:
https://www.mathnet.ru/eng/izkab221 https://www.mathnet.ru/eng/izkab/y2017/i62/p43
|
Statistics & downloads: |
Abstract page: | 68 | Full-text PDF : | 44 | References: | 7 |
|