Article
The purpose was to develop an optimal model of a straight punch in boxing based on an artificial neural network in the form of a multilayer perceptron.
Methods and organization of the research. The architecture of the neural network optimal punch model included an input layer of 600 nodes – the values of absolute accelerations and angular velocities, three inner layers (256 -128 - 64 nodes), as well as a binary output layer (the best and the worst punches). The model used the activation function in the form of a sigmoid on each layer, and the loss function was in the form of a binary cross-entropy. The Adam algorithm was used as the optimization algorithm. To measure accelerations and angular velocities, inertial measurement units (IMUs) were attached to the boxers ' wrists. Highly qualified boxers (more than 5 years of training experience) participated in the data set for the development of the optimal model. The best punches were chosen according to the criteria of strength and speed. The punch force was determined using a boxing pad with the function of measuring the punch force. In order to be able to compare punches, a unified parameter was developed, called the quality of the punch, which is equal to the product of the effective force (a characteristic proportional to the power of the punch) and the speed of the punch. To study the effects of biofeedback, the boxing pads were equipped with five LEDs. When punching the pad, the more LEDs were turned on, the more the punch corresponded to the optimal model.
The results of the research. In the course of research, an almost linear relationship was found between the punch quality of entry-level boxers and the optimal model. The use of feedback allowed for an increase in the quality of punches from 11 to 25%, which is on average twice as high as in the group where the feedback method was not used.
Conclusion. The research has shown that it is possible to develop an optimal punch model. According to the degree of compliance with this model, you can evaluate and teach the technique of punches of boxers.
- Главная страница Keras, https://keras.io.
- Главная страница Tensorflow, https://www.tensorflow.org
- Chadli, S., Ababou, N., Ababou, A.: A New Instrument for Punch Analysis in Boxing. Procedia Engineering. 72. 411. 10.1016/j.proeng.2014.06.073 (2014).
- Kimm, D., Thiel, D.V.: Hand Speed Measurements in Boxing. Procedia Engineering, Volume 112, 502-506 (2015).
- Guo, J., Yang, L., Umek, A., Bie, R., Tomažič, S., Kos, A.: A Random Forest-Based Accuracy Prediction Model for Augmented Biofeedback in a Precision Shooting Training System. Sensors, 20, 4512 (2020).
