Optimal Trajectory Planning for Autonomous Robots Under Dynamic Network Connectivity Constraints: A GraphSAGE Approach

Abstract

In this paper, we introduce an innovative trajectory optimization framework for autonomous robots, designed to effectively navigate the complexities of maintaining network connectivity, avoiding collisions, and ensuring comprehensive area coverage. This framework employs GraphSAGE, a cutting-edge graph-based deep learning algorithm, notable for its adaptability to real-time environmental changes. Our approach enables autonomous robots to make strategic path-planning decisions that uphold continuous network connectivity, minimize the risk of collisions, ensuring optimal area coverage through strategic spacing and coordinated exploration. Through rigorous simulations, we have benchmarked the performance of our GraphSAGE-based method against traditional trajectory planning strategies. This research not only underscores the viability and scalability of integrating GraphSAGE into autonomous robotic systems but also marks a significant progression in autonomous navigation technologies, highlighting the capacity of graph-based deep learning algorithms to substantially improve the adaptability, performance, and operational efficiency of autonomous robots in complex network scenarios.