Development of an AI-enabled AGV with robot manipulator

Abstract

The presented work describes a ROS-based automated guided vehicle (AGV) applied in usage of an assembly production line. By reaching the purpose of automatic navigation, two sets of LIDAR sensors are installed to approach simultaneous localization and mapping (SLAM). A RGBD camera is used to perceive the working environment and identify workers captured in sight, through a trained AI deep learning using YOLO 3. A robot manipulator is attached on AGV to perform pick-and-place task. The developed AGV has shown to perform well in SLAM, automatic navigation, obstacle avoidance in typical indoor environment.