Detection of 3D perceptual changes in robot navigation using cloud computing

摘要

This paper presents the implementation of 3D perceptual parameter estimation in robot texture images using multi-core Graphic Processing Units (GPUs) connected into a cloud computing environment. The paper discusses the architecture of the system and its benefits in providing real time estimation of parameter estimation to get an approximation of the change in position of robot from a certain position for application of robot navigation. The implementation of this scheme on conventional computing resources is much time consuming and resource intensive. The advances in cloud computing provide cost effective solution for this task of 3D change estimation.