Hybrid neural-based control system for mobile robot

Abstract

problems of discrete-continuous structure optimization is possible by means of integration of artificial intelligence technologies with methods of mathematical programming and computingretrieval procedures. Further research will be made with the purpose of increasing efficiency of the designed set of algorithms and programs, decreasing search time at preset parameters of optimization and implementation on several parallel systems or processors.

New Paradigm of Context based Programming-Learning of Intelligent Agent

“…Associative memory satisfied to above requirements may be based on hybrid approach and similar to Long-Short Term Memory (LSTM) [15] or table based memory proposed by author in [16]. Concept in Fig.…”

Hybrid Neural Network Model Based on Multi-layer Perceptron and Adaptive Resonance Theory

“…The absence of adapting to transformations of input vector to same one in real time is a result to that for real applications the model ART-2 is almost never used. For example, for clustering and pattern recognition by the mobile robot [6] it is required to recognize the object in different foreshortenings and allocating in different parts of a field of vision, i.e. recognition should be invariant concerning conversions of the map, such as shifts, rotations and others.…”

An Approach for Invariant Clustering and Recognition in Dynamic Environment

“…It leads to the result that the model ART-2 is nearly never used for real applications. For example, considering the clustering and pattern recognition of a mobile robot, [6], [7] it is required to recognize the object in different foreshortenings which may allocated in different parts of a field of vision, i.e. recognition should be invariant concerning transformation of the input pattern, such as shifts, rotations and others.…”