<![CDATA[This study discusses the design and implementation of a delivery robot system based on object detection using a camera with the YOLOv5 algorithm. The system is designed to operate autonomously in a limited-space MSME warehouse environment, equipped with a gripper and forklift mechanism to pick, lift, and deliver box-shaped goods. The dataset used includes ±300 images at various distances, positions, and lighting conditions. The test results show that the detection accuracy reaches 96.67% in low-light conditions and 83.33% in high-light conditions. The robot is able to operate for 1 hour 15 minutes with a maximum lifting capacity of 1 kg and optimal performance at loads up to 700 g. The implementation of this system has the potential to increase the logistics efficiency of MSMEs by reducing reliance on manual labor, accelerating distribution, and improving the accuracy of goods handling.]]>
Copyrights © 2025
