DOCUMENT DELIVERY ROBOT BASED ON IMAGE PROCESSING AND FUZZY CONTROL

Abstract

The objective of this study is to integrate image processing, pattern recognition, RFID, and fuzzy theory into an omnidirectional wheeled mobile robot for receiving and delivering documents between rooms. In image pre-processing, the Hue-Saturation-Lightness color space is applied to avoid light interference, and then grayscale image threshold is used to obtain binary image. The median filter is utilized to filter the noises of speckle and salt-and-pepper, so color segmentation is then applied to capture desired color for tracking control. Pattern recognition is performed by the Adaptive Resonance Theory. RFID reader and room tag is used to verify the room number of the destination so that the recognition error from image processing can be avoided. Fuzzy theory is implemented into an omnidirectional wheeled mobile robot control design for driving the wheels of the robot. Experimental results show that the proposed control scheme can make the omnidirectional mobile robot move to destination, receive and deliver documents between offices. © 2016, Canadian Society for Mechanical Engineering. All rights reserved.