This paper presents a framework for telepresence robot navigation in dynamic environments with network-induced time delays. The proposed system introduces a predictive control model that processes sensor data, implements real-time control algorithms, and transmits commands to enable robust remote navigation. To address visual and control discrepancies caused by latency, a state estimation model is employed to minimise the visual disparity between the robot’s actual and perceived positions. Additionally, a simulation-based predictive controller anticipates operator commands to improve teleoperation accuracy. A key contribution of this work is the development of a low-cost, simulation-based telepresence platform that enables controlled experiments without relying on expensive physical infrastructure. The system is designed for flexibility, allowing parameter adjustments to suit a range of experimental conditions. By integrating predictive technologies and addressing latency-related challenges, this research advances the state-of-the-art in telepresence robotics and provides a practical, reproducible foundation for future studies in remote robot navigation.