Machine Learning Approach for Smart Serving Robots in Healthcare and Retail Supply Chain Management

Abstract

Smart serving robots have emerged as intelligent autonomous systems capable of improving operational efficiency in healthcare and retail supply chain management through the integration of machine learning, computer vision, and autonomous navigation technologies. This study investigated machine learning approaches implemented in a smart serving robotic system designed for autonomous object detection, obstacle avoidance, path planning, and delivery operations within dynamic environments. The developed robotic platform integrated YOLOv4 object detection, convolutional neural networks, RRT* path planning, ultrasonic sensing, and embedded communication systems using NVIDIA Jetson TX2 and ATmega328 microcontroller architecture. Experimental evaluation demonstrated that the robotic system achieved reliable real-time object recognition, efficient collision-free navigation, stable communication performance, and adaptive obstacle avoidance capabilities. In healthcare environments, the robot showed strong potential for medicine transportation, contactless delivery, and workload reduction for healthcare staff, while in retail supply chain applications the system improved inventory transportation, warehouse automation, and logistics efficiency. The findings indicate that machine learning significantly enhances robotic intelligence, operational reliability, and autonomous decision-making within service-oriented environments. Despite challenges related to computational constraints and dynamic navigation complexity, the study concludes that AI-driven smart serving robots represent a transformative solution for intelligent logistics automation and Industry 4.0 applications.