The development of technology in automation and robotics is overgrowing because of its high-efficiency level in terms of labor and time. In the warehousing system, one of the robots used is Automated Guided Vehicle (AGV). AGV is a transportation device in the form of a robot that can be controlled automatically, which functions as a carrier of goods using a navigation system to move in a predetermined direction. One of the existing AGV navigation systems is by following a line pattern on the floor. The system is inefficient because, gradually, the line pattern will wear out and can not be detected again due to the friction force of the AGV wheels itself. Therefore, it is necessary to develop an AGV navigation system to minimize these obstacles. This pattern recognition system uses a pattern placed on the building ceiling and camera as a sensor facing upwards, so that AGV can freely detect patterns. Then, the detected pattern was processed through a programmed Raspberry Pi 4 Model B. The test results show that this system can detect the position and successfully displays the coordinate point (x, y) of the AGV and will continue to run at any time until the program is changed as ordered.