Development of a Vision Based18-DOF Hexapod Robot Controlled by ROS2

This paper presents the design and implementation of an 18-degree-of-freedom (DOF) hexapod robot, utilizing 3D printing techniques for its construction and controlled by Robot Operating System 2 (ROS2). The hexapod features six legs, each with three DOF, enabling movement across various surfaces, particularly those challenging for wheeled vehicles. Equipped with a camera system, the robot incorporates a custom-trained You Only Look Once version 8 (YOLOv8) model for object detection, specifically targeting toy vehicles, mechanical design process, including challenges encountered during 3D printing, implementation of ROS2 for coordinating locomotion and vision processing and the incorporation of a basic gait algorithm for the robot's movement. Additionally, The paper explores the robot's vision capabilities, including the training process and performance of the YOLOv8 model. This project demonstrates the integration of legged locomotion with basic machine vision in a customizable platform, showcasing potential applications in educational robotics and exploration of uneven surfaces.