A precise scan matching based localization method for an autonomously guided vehicle in smart factories

Abstract

Precise localization and autonomous docking at industrial standards are major issues to be handled in order to establish smart factories having end-user oriented production focus and to operate fully autonomous vehicles in the logistics of such factories. Therefore, in this paper, a precise localization algorithm utilizing the affine iterative closest point (ICP) method is proposed for logistic applications in manufacturing systems. In classical ICP, the least-squares criterion and the point-to-point metric are employed. In this study, however, correlation entropy (correntropy) criterion is used in order to provide robustness against noise and/or outliers. In addition, affine transformation is exploited to increase the flexibility of the developed algorithm and the point-to-line metric is put into use to perform pose estimation faster. The developed algorithm is suitable for logistic applications in smart factories, which involves reaching a target at industrial standards. On the other hand, this method can also be employed in some outdoor applications, for example, parking problems of autonomous vehicles. The improvements and performance gains achieved using this method are demonstrated in nine separate real-world cases. In the field tests, ITU-AGV, a robot operating system (ROS)-enabled differential drive mobile robotic platform, is deployed. The achieved results prove that it is possible to dock a vehicle used in logistics to a target with sub-centimeter precision.