Obstacle avoidance path design and simulation of in-space capsule service robotic arm based on YOLOv8

• Main Author: Jiang, Zhouhao
• Other Authors: Hu Guoqiang
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering; Obstacle avoidance; ROS; YOLOv8
• Online Access: https://hdl.handle.net/10356/176329
• Institution: Nanyang Technological University

With the development of modern industry and computer technology, robotic arms are gradually used in various fields, industry, logistics, medical care, service and so on. Whether the robotic arm successfully avoids obstacles is an important part of the robotic arm path planning needs to be studied. At present, the robotic arm used in the field of spaceflight is mainly to complete the sample collection, repair instruments and other tasks, but for the service of the astronauts in the daily life of the robotic arm is seldom involved. In this paper, the path design of robotic arm for space capsule service is combined with RRT* path planning algorithm and FCL collision test for obstacle avoidance. In addition, considering the zero-gravity environment in space, the obstacles may be floating and uncertain, and the dynamic obstacles are specially considered. Obstacles were detected in real time by YOLOv8 image recognition technology. The feasibility and reasonableness of the involved path planning methods are verified through the path trajectories of Rviz, the visualisation interface of ROS, and the simulation results of Gazebo. Finally, the motion state images of each joint also demonstrate the coordination ability and coherence of the robotic arm’s motion process. In the future, more irregular obstacles will be considered and combined with the grasping object and machine obstacle avoidance path planning to comprehensively consider the obstacle avoidance ability of the robotic arm under the working state, and complete the obstacle avoidance experiments using the solid robotic arm.