Real-Time Object Detection and Grasping with an Autonomous Mobile Robot: A Case Study

Abstract

This study presents an autonomous mobile robot with advanced grasping capabilities designed to perform two missions: Pick and Place and Hold and Run. The robot’s four-wheeled chassis ensures stability and maneuverability, while an articulated arm with a specialized gripper manipulates 8-cm-diameter balls. Utilizing YOLO algorithms for accurate object detection, the vision system offloads image processing to a laptop via Wi-Fi for real-time monitoring and decision. Extensive tests validated the robot’s consistent task execution, demonstrating dexterity and intelligent navigation. Challenges such as hardware integration and vision system optimization were innovatively addressed. The modular design allows for easy upgrades and adaptability, highlighting significant future potential. The proposed robot fulfilled the preset goals of picking any defined colored ball and placing them into a box autonomously. Besides, achieving the athletic runner task around the classroom interior. This project showcases the interdisciplinary nature of robotics, combining mechanical engineering, computer vision, and software development with potential applications in service robotics.