Hybrid Visual Servo Control for Point-to-Point Localization of an Autonomous Wheeled Mobile Robot

Abstract

Position-based visual servo (PBVS) control and image-based visual servo (IBVS) control have been the viable approach to control of autonomous mobile robots and manipulators. However, PBVS suffers from the sensitivity to model for position calculation, and IBVS is limited to the field of view covered. In this paper, a hybrid visual servo control (HVSC) integrating PBVS and IBVS was proposed for tradeoffs of above performance. To achieve the visual servo control, a fixed depth camera is mounted to an autonomous wheeled mobile robot (WMR) for visual feedback information. Point-to-point localization tasks were executed for the WMR. Experimental results demonstrate the feasibility and effectiveness of the proposed hybrid visual feedback schemes.