ICP Algorithm in Mobile Robot Navigation: Analysis of Computational Demands in Embedded Solutions**This work was supported by the project SP2016/162, Development of algorithms and systems for control, measurement and safety applications II of Student Grant System, VSB-TU Ostrava.

Abstract

An accurate robot localization, particularly in indoor environments, represents a challenging aspect in mobile robotics due to lack of computational power of commonly used embedded platforms. To reduce the required computational demands it is necessary to apply an efficient algorithm capable of parallelization. This contribution elaborates on implementation of Iterative Closest Point (ICP) laser scan matching algorithm. A computer program developed and optimized in C# language is compared to a mathematical model designed in MATLAB. The characteristics of both approaches are assessed based on computational demands (e.g. time) and nearest neighbor determination.