ipmash@ipme.ru | +7 (812) 321-47-78
пн-пт 10.00-17.00
Институт Проблем Машиноведения РАН ( ИПМаш РАН ) Институт Проблем Машиноведения РАН ( ИПМаш РАН )

МИНОБРНАУКИ РОССИИ
Федеральное государственное бюджетное учреждение науки
Институт проблем машиноведения Российской академии наук

МИНОБРНАУКИ РОССИИ
Федеральное государственное бюджетное учреждение науки
Институт проблем машиноведения Российской академии наук

Adaptive exchange protocol for multi-agent communication in augmented reality system

Авторы:
Ekaterina Moseiko , Oleg Granichin , Victoria Erofeeva ,
Страницы:
275-281
Аннотация:

Augmented Reality (AR) is one of the modern ways of providing users with different types of information. In applications based on this technology, the main influence on the subjective assessment of product quality is the speed of interaction between the system and the user and between users of the system. The main purpose of the work series are creation and modification for algorithms to assessing and improving the quality of AR services. This paper describes and justifies an adaptive communication protocol for multi-agent interaction. We consider a general nonlinear dynamic system with introducing feedback control which is based on measurements under almost arbitrary noise. In practice, this control is realized as a superposition of neural networks with the estimation of the result of mutual additional learning based on the system identification method (M.K.Campi and E.Weyer’s LSCR is used). The prototype is built for a system with a distributed server architecture and multiagent behavior of both clients and servers in the system.

Файл (pdf):
17:26
321
Используя этот сайт, вы соглашаетесь с тем, что мы используем файлы cookie.