이동로봇의 셀 분할 기법에 기반한 전체 커버리지 경로계획과 경로추종제어

Abstract

This thesis proposes exhaustive coverage path planning based on cellular decomposition and path-following control of a mobile robot. For this work, the following tasks are implemented. First, exhaustive coverage path planning including coverage path of each cell and moving path between cells is proposed. For coverage path, cellular decomposition method and back-and-forth motion are used. For generating moving path, tangent bug algorithm is used. Furthermore, direction of coverage path and generating graph path using proposed special spanning tree is considered to get minimum distance of moving path. Then, the exhaustive coverage path is obtained by combining coverage path and moving path. Second, for path-following, a mobile robot with two passive wheels in front and back of the mobile robot and two active wheels in left and right sides of the mobile robot is presented. Kinematic modeling of this system is presented. Based on the kinematic modeling, a backstepping controller is proposed to follow the proposed path from exhaustive coverage path planning. Third, to implement the proposed controller, a control system is developed based on microcontroller AVR ATmega128. A compass sensor and rotary encoder is used to measure the rotation angle and relative coordinate of a mobile robot system. Finally, simulation and experiment results are presented to show the effectiveness of the proposed exhaustive coverage path planning and path-following controller for AGV system with navigation device introduced in appendix A.