In the context of multi-agent systems, movement behaviors of robots playing Nine Robot's Morris (NRM) will be discussed. The robots simulate autonomously the movement of stones in the original game Nine Men's Morris by maneuvering a 2m x 2m game board. In this paper, we deal with the movement of the robots with respect to their limited computational capabilities and inadequate number of sensors. Firstly, a line following algorithm is developed based on two Reflective Optical Light Sensors (ROLS), providing stable movement and avoiding the zigzag phenomena. Secondly, an intersection maneuvering algorithm was developed to detect and maneuver intersections on the game board with integration into the line following algorithm. The low resolutions of the sensors and actuators were taken into consideration while maintaining robust intersection handling. Experimental data is provided, comparing several approaches to accomplish these tasks.