Task Scheduling and Path Planning of Hotel Service Robots Driven by Artificial Intelligence

Abstract

Many hotels embrace intelligent service robots as a novel approach to enhancing customer satisfaction and streamlining operations. However, the ever changing hotel environment makes scheduling and managing obligations difficult. This research introduces Scheduling + Navigation Robotic Executor (SchedNav RX) for real time route planning and job prioritization. The proposed SchedNav RX utilizes a deep Q network based adaptive task scheduling reinforcement learning model, enabling robots to reschedule and prioritize tasks based on urgency and context dynamically. A convolutional neural network (CNN) enhances the standard A* algorithm for navigation by predicting obstacles in real time. This makes dynamic interior navigation safer and more efficient. TurtleBot 3 units were used for physical validation to enhance performance evaluation. SchedNav RX outperforms standard planning systems by 27% in task completion time and 35% in navigation safety while dealing with unexpected vehicle traffic. These findings demonstrate that SchedNav RX is essential for intelligent, autonomous robots to perform hotel service tasks efficiently and easily. The concept allows complicated hospitality environments to accommodate dispersed robotic systems driven by artificial intelligence. This work will evolve to incorporate reinforcement learning based guest feedback and interaction modules, enhancing the system's capabilities.