Autonomous Museum Tour Guide Robot With Object Detection Using Tensorflow Learning Machine

This device focuses on making tools as a substitute for human resources in historical tourism by creating the Autonomous Museum Tour Guide Robot. This robot is made to help tourists in the Museum walk around in a coherent way throughout the museum area and explain comprehensive information and experiences related to the Museum and the existing collections. The tool’s implementation is controlled by the deep learning Convolutional Neural Network method using the Tensor flow framework to recognize and classify the detected objects. The robot that will be created is a 3-wheeled robot with one camera as a sensor to detect objects around the robot. The robot is equipped with an audio speaker to provide object detection information. The robot detected six objects at the Sumenep Palace Museum, which were integrated into a robot with a 100% success percentage in the 5th epoch with the 175th iteration. The time required was 117.11 error of 0.393. Weaknesses in this study are the need for control throughout the tour guide robot so that it runs more stable and the use of a camera with a higher resolution, but when run on the system, it does not affect system performance.