Abstract
In recent years, artificial intelligence has been widely used in the field of robotics. However, these robot-related tasks are difficult to migrate from the cloud to edge nodes due to the large computing and storage resource requirements. In this project, we develop a platform for a heterogeneous robotic system. The platform is built to facilitate the development of advanced robotic applications with minimal human interactions, where search and find system traversal is the main application. To support robots performing tasks in near field in real time, in this paper we introduce traversal and task division algorithms which are introduced to perform a cooperative search mission by a group of robotic agents to achieve intelligent search and find as an edge service. We evaluate the performance against the algorithm's parameters using data obtained in controlled field experiments. The aim was to identify and study some key performance parameters impacting the traversal function in the application.