A Convenient Pose Measurement Method of Mobile Robot Using Scan Matching and Eye-in-Hand Vision System

This paper presents a convenient and practical pose measurement method with satisfactory accuracy to obtain the absolute pose of the mobile robot near the destination in the wide-area and multi-destination scenario. The proposed method consists of the following distinguished features: first, an improved Monte-Carlo localization with the scan matching method is proposed to accurately calculate the absolute pose (target pose) of the mobile robot at the target destination; then, an eye-in-hand vision system is designed to attain the pose error between the actual and target pose of the robot, and the actual absolute pose can be obtained. At this point, the pose measurement near the destination is realized; moreover, these accurate measurement data are in application to parking error compensation. Compared to traditional measurement methods, our method does not require many additional devices to implement in the scene and meet the measurement requirements near the destination with guaranteed accuracy. The feasibility of the proposed method is verified through several experiments using the self-developed mobile robot.