Research on teaching based on tennis-assisted robot image recognition

摘要

Design and development of an autonomous robot will collect the scattered tennis balls on the tennis court. To make the camera active, the robot uses a camera placed above from these recently collected the ball to roll. And, given the machine design, because it falls to prevent the ball, catch the ball. Image processing, the use of color filters, the detected edge detection algorithm, corresponds to the Raspberry Pi 3, can be achieved by the distance to the library implementation of robot OpenCV, it is sorted. After the desired robot angle is generated, it is translated to the appropriate motor speed value is sent to the motor control unit. To realize this behavior, it has four DC motors to be used. For the movement platform, one of which is a scanning operation of the ball, which is used to control the camera angle, the last one will be used for the ball detent mechanism. The realized by the discrete PI controller with two DC motors position and speed control motor controller boards of the rotary encoder. Experimental tests of the real world validate the design of the system.