Trajectory Tracking of End Effector on Mobile Robot with Multiple Onboard Cameras

Abstract

This paper presents technologies for precise trajectory tracking of end effector with two core principles; optimizing relative position of end effector and compensation of position by visual feedback. First of all, based on holonomic robot control using kinematics, this paper suggests a redundant system by adding additional Degrees Of Freedom(DOF). This system, unlike in kinematics, can consider the inertia effect in modeling and reduce slippage and skid errors of the robot. Also, this system could reduce the accumulated error by comparing the real-time image of the trace of end effector with the pre-modeled trajectory. The entire process is controlled by the combination of image processing unit, navigation planner and multiple embedded systems. This system ensures 1) stable real-time communication between multiple processors, 2) self-feedback of localization using its trajectory captured by onboard cameras, which allows 3) accurate position control of end effector during navigation.