Wrist MEMS Sensor for Movements Recognition in Ball Games

In ball games, the arm movements are diverse and complex. The accurate recognition of the arm movements is of great significance to the quality of training and sports guidance. Wrist sensor is one of the important means. Traditional wrist sensors recognize batting action by measuring the acceleration and angular velocity of the arm and have a good performance. But they measure few features, and it has a slightly larger size that has some effect on player performance. In this paper, we develop a motion recognition platform for Basketball, Ping-Pong, and badminton. It is accurate enough for recognizing these motions whether hits or miss the Ping-Pong ball. The system integrates a low-power nine-axis inertial sensor and a low-power environment sensor that can output eleven different characteristics such as triaxial acceleration, triaxial angular velocity, triaxial magnetic field strength, air pressure, and temperature. This system includes the power management part, attitude detection part, Bluetooth wireless communication part and mobile terminal monitoring part. The inertial sensor can identify the arm action, and then transmits posture data to the Microcontroller Unit (MCU) for analysis and processing. Bluetooth 5.0 is used between the MCU and the mobile monitor for information exchange, which makes data transmission faster and more stable. Experiments of the 3 athletes have demonstrated that the movements can be accurately identified. And the entire system has very low power consumption and very small size (23.05mm*9.25mm*5.5mm). It provides a multi-feature measurement platform for body movement posture research and can recognize a variety of ball games movement, which is of great significance for intelligent sports training and research.