Development of a badminton teaching system with wearable technology for improving students’ badminton doubles skills

Abstract

The aim of this study was to develop a badminton teaching system using the wearable technology for improving badminton teaching and learning. This system can provide the similarity scores automatically by comparing a student’s movement and strength while playing badminton with a well-trained expert model. A quasi-experiment was conducted for eight weeks to evaluate the effectiveness of the developed system. A paired-sample t-test was analyzed to evaluate whether there is a significant difference between the pretest and posttest for the experimental group (EG) and the control group (CG) respectively. Also, an independent sample t test was utilized to compare the posttests between EG and CG. The results show that the badminton teaching system developed in this study could enhance the learning effectiveness of novice badminton learners. The results not only show that applying wearable technology, the Myo armband, can benefit learners’ badminton training, but also can demonstrate the importance of providing instant and adaptive feedback in motor skill learning.

Appendix

The Myo armband is equipped with a 3-Axis gyroscope which was used to track and model players’ movements for a specific badminton skill. To provide technical information about the developed system, smashing will be used as an example to describe the technical details. There were seven steps for the whole development procedure.

Step 1: Collecting the 3-Axis gyroscope data from four national Badminton players. Each expert Badminton player was asked to wear the Myo armband and execute smashing motions 15 times.

Step 2: Storing the angular velocity of Axis X, Y, and Z for each data point collected from the Myo armband.

Step 3: Dividing a whole motion of smashing data into four sub-motions based on the sub-motion definition provided by the badminton teacher (domain expert). These four sub-motions can be visualized as shown in Fig. 

Fig. 11

The illustrations of the four sub-motions of smashing

11.

The division based on the angular velocity timing diagram from the collected data is shown in Fig. 

Fig. 12

The decomposition of smashing and algorithm of movement division

12 and Table 8.

Table 8 The description of each division point of smashing

The justifications for the sub-motion divisions are described below:

An apparent and unique valley exists in the angular velocity timing diagram of Axis Y as indicated at (1). A deep valley represents that the arm was moved downward swiftly, and it corresponds to the time period of “turning the elbow and swinging the racket swiftly” of sub-motion 3. To divide a smash into four sub-motions, the design algorithm picks this minimum value as the first judging point to search for other dividing points. Furthermore, the system drew out the dividing time scope by means of (1) the minimum value because of learners’ unfamiliarity with the standard movement which caused them to fail to perform similar movements to those of the expert players, and the difference in the angular velocity timing diagram could lead the algorithm to falsely locate the first judging point. To correctly find the dividing points of every sub-motion of smashing learners, this study set (2) the scope of sub-motion 3 as the additional judging basis to limit the searching in this scope.

After the minimum value was found at (1), the starting point of sub-motion 3 could also be found by going back to the earlier time point to locate (3) the pre-maximum value. Moving back to the earlier timing from (3) of the pre-maximum value, another value which was closest to 0 on Axis Y could be marked. When the value of angular velocity of Axis Y was rather close to 0, it means there was no change of angle. Such closeness could be regarded as the transitional time point between sub-motions. Accordingly, the value closest to 0 was (4), which should be the starting point of sub-motion 3.

To find out the ending point of sub-motion 3, it should start from the minimum value and move forward so that the post-maximum value at (5) could be identified. Just continue moving forward to find the point which was the closest to 0 on Axis Y; this point at (6) should be the ending point of sub-motion 3.

After the time period of sub-motion 3 was determined, the next one is to find the time period for sub-motion 2. In II, since the starting point of sub-motion 3 was already identified, the point at (4) should also be the ending of sub-motion 2; thus, all we had to do was find the starting point of sub-motion 2. First, let’s begin with the ending point of sub-motion 2 and move backward to search for (7), the maximum value of sub-motion 2, then continue moving backward again to find the point closest to 0 on Axis Y. The point at (8) should be the starting point of sub-motion 2.

The starting point of sub-motion 1 at (9) and the ending point of sub-motion 4 at (10) were the beginning point and the terminal point for the whole timing diagram of Axis Y. The ending point of sub-motion 1 should be the starting point of sub-motion 2. Moreover, the starting point of sub-motion 4 should be the ending point of sub-motion 3.

Step 4: Calculating the movement angles of smashing using the Riemann sum integral for each sub- motion using the values of angular velocity of Axis X, Y and Z. The algorithm formula is presented as follows:

$${\text{Riemann}}\;{\text{sum(S) = }}\int_{a}^{b} {f(angular\;Vector\;(x))} dx$$

Step 5: Normalizing the calculated movement angles and training the smashing movement expert model.

Step 6: The ten-fold cross-validation was used to evaluate the accuracy and suitability of the trained smashing movement expert model.

Step 7: Developing the application and user-interface for the badminton teaching system.