Relative directions such as left, right, forward, and backward are fundamental to spatial navigation. This paper presents a novel computational technique for tracking location by learning relative directions between two intelligent agents. In this model, agents communicate via radio signals, where one agent assists the other in locating itself through spatial learning. S. Rayhan Kabir and the research team propose the Relative Direction Based Location Tracking (RDBLT) model as a decentralized alternative to GSM (Global System for Mobile Communications). This approach is particularly designed for AI and Multi-agent systems operating in environments where standard network infrastructure is unavailable. The study introduces three proficient algorithms developed to construct the RDBLT model, enabling agents to identify and learn relative directions autonomously. By reducing reliance on centralized networks, this research advances the field of Artificial General Intelligence (AGI) and provides a robust framework for autonomous navigation in remote or network-denied locations.

Keywords: Artificial General Intelligence (AGI), Multi-agent System, RDBLT, Location Tracking, Relative Directions, Autonomous Agents, AUB.