An object localization and reaching method for wheeled mobile robots using laser rangefinder

Abstract

This paper proposes an efficient scheme to solve the problem of identifying an object from range data and navigating to it by on line calculating and performing highly continuous paths for a precise target reaching. Our approach combines a laser range-based object localization method with the PC-sliding, a technique based upon polynomial curvature paths that are suitable for nonholonomic mobile robots and robust to measurement and model uncertainties. First the object is on-line localized using a laser rangefinder taking into account sensor and robot pose uncertainties and then a PC-path is both planned and real-time controlled with a unique strategy.