THE DESIGN OF TABLE TENNIS ROBOT COMBINED WITH DEEP LEARNING

This paper analyses the equipment based on a table tennis opponent made by combining machine vision and embedded control of deep learning .It is mainly divided into two parts: hardware and software. The hardware part includes the sensor part, the power transmission part and the shape structure part; the software part includes the machine vision part and the embedded control algorithm part. Machine will be through the combination of sensor and machine vision monitoring to the position of the table tennis, and the machine will by monitoring the opponent posture can be used to predict the flight path of the ball, and then to forecast the movement trend, and the parameter data transmit to MCU, MCU will receive the data and compare the sensor received data integration, and then control mechanical arm swing to hit the ball.