An approach for integrating multi-autonomous mobile robot (AMR) systems that work in flexible manufacturing environments while avoiding deadlocks and livelocks, as well as obstacles that are not registered on the map server with the Control Lyapunov Barrier Function (CLBF). Dijkstra's algorithm is utilized for path planning and CLBF as the control method for AMRs to ensure that it runs safely to a determined station without colliding with unregistered obstacles. The setup assigns the AMRs a mission retrieving goods from the station and transporting them to another station. Additionally, the map server does not have any previously registered obstacles. A communication and coordination algorithm among AMRs is designed to prevent deadlocks, livelocks, and collisions among them. The efficacy of the proposed control and algorithm is demonstrated in a ROS simulation. The simulation is divided into two scenarios: one with two AMRs in a 16 m × 9 m workspace with unregistered obstacles, and another without any obstacles. The simulation result shows that the latter scenario has completed more missions by 26.6% than the former.