An artificial intelligence fuzzy system for improvement of physical education teaching method

Abstract

There are certain disadvantages in the traditional physical education teaching model. In order to improve the advanced nature of physical education teaching methods, this paper builds a physical education evaluation system based on artificial intelligence fuzzy algorithm. The system uses fuzzy control instructions as the basis to combine human language and mechanical language, so that the machine can recognize human working language habits and execute commands according to the instructions. Moreover, in this study, the trapezoid function is selected as the membership function, and the improved particle optimization algorithm is used to capture the student’s motion process and the motion vector decomposition, and the system structure model is constructed based on the functional requirements analysis. In addition, this study conducts system performance analysis through experimental teaching methods. The research results show that this system can effectively promote the reform of teaching methods in physical education and has a certain practical effect.

Keywords

Artificial intelligence fuzzy system physical education teaching improvement

1 Introduction

Compared with traditional physical education, there is a big difference in sports teaching based on artificial intelligence. The artificial intelligence physical education system can not only evaluate the effect of physical education in real time, but also effectively improve the efficiency of teachers’ classroom teaching. Moreover, intelligent demonstrations can provide students with effective references and reduce students’ self-exploration practice. It can be seen from this that the artificial intelligence sports teaching system will surely become the future development trend of sports teaching [1].

In recent years, sports undertakings have flourished, the value of sports has been continuously explored, recognized by people, and its social status has also continued to improve. Moreover, the sports talent market has expanded accordingly, and the demand for sports talent in society continues to rise. At the same time, it is subject to higher requirements and standards. Therefore, there is a long way to go to train high-quality sports talents. Colleges and universities are one of the main ways to convey sports talents. The standards for the development of sports talents are gradually improved, and the corresponding education activities of sports talents should be constantly innovated, and the corresponding teaching evaluation is equally important. How to improve the scientific nature of evaluation based on development standards has become a hot issue in teaching research, which has attracted the attention of college physical education educators and teaching leaders [2].

At present, the main courses of physical education majors include sports humanities and sociology courses, sports human science courses, which are called “subject” in this study, and track and field, gymnastics, ball and martial arts, are called “technical subjects” in this article. At present, the technical subject is mainly based on technical assessment and compliance testing, supplemented by technical theoretical knowledge, and focuses on the evaluation of technical level. At present, western educational evaluation theories have been introduced into domestic research. Based on the national conditions, how to further combine theory and practice to implement the evaluation and promote the development of educational evaluation has become a core issue of concern to everyone [3].

2 Related works

The literature [4] developed the physical education measurement tool “NSSE-China”, including 5 first-level indicators and 28 second-level indicators, and measured the learning effect of college students in physical education courses from the perspective of academic challenges and cooperative learning levels. The literature [5] obtained the purpose and requirements of the physical education learning evaluation of colleges and universities in Henan Province through the survey results. Its content is basically consistent with the guiding ideology of the “Outline of Physical Education Curriculum Guidance for National Colleges and Universities”. However, in the operation process, the evaluation form is too simple and lacks a reasonable scientific evaluation system, so the students’ satisfaction with the evaluation results is low. The literature [6] believed that the following problems exist in the evaluation of physical education in colleges and universities: the error of sports skill evaluation; physical quality and sports ability evaluation ignore the individual differences; the content of extracurricular sports activities is neglected; the assessment of lifelong sports ability is neglected. The literature [7] pointed out that the teachers and students of ordinary colleges and universities do not have enough understanding of the concepts, basic theories and methods of physical education evaluation, and lack of understanding of the basic concepts of physical education curriculum reform with “health first” as the core. Moreover, it pointed out that the content of the evaluation is not comprehensive, and the design of the score ratio of the evaluation content lacks scientificity, which is not conducive to students’ awareness of “lifelong sports”. In addition, it pointed out that most colleges and universities only use the attendance rate of the above courses as a standard to measure students’ learning attitude, and the evaluation standard content is too simple, and colleges use quantitative evaluation instead of qualitative evaluation, so the evaluation method is single. The literature [8] learned from the survey that teachers and students of Zhejiang colleges and universities hold a supportive attitude towards the reform of physical education evaluation. However, there are few specific reform proposals, and teachers still use the traditional evaluation model. The evaluation standards are formulated by the school and are different from the ideal evaluation methods and standards of students. At present, teachers in some schools have begun to change the concept of evaluation, and strive to highlight students’ dominant position and individual differences. Literature [9] evaluates the content of physical education learning evaluation in colleges and universities, and there is a big gap between it and the “Guidelines for the National Physical Education Curriculums for Colleges and Universities”. The content of physical education evaluation is mainly based on summative evaluation, and it ignores the differences in physical fitness of different students. This is likely to cause frustration for students with relatively poor athletic ability, which is not conducive to the healthy development of their body and mind and reduces their enthusiasm for learning physical education courses. Moreover, it is not conducive to students’ profound understanding of the importance of sports, and to a large extent weakens students’ interest in sports. With the development of society, in recent years, the vigorous development of sports undertakings and physical education, the function of sports has been gradually explored, and its core function is not only to educate the body, but also to educate people [10].At present, education scholars have fully recognized the shortcomings and limitations of the traditional evaluation system. Based on the study of foreign advanced theories, my country’s academic evaluation system is rapidly reforming and developing, and the academic evaluation theoretical system has been relatively perfect [11].From the research results of different provinces and cities, it is not difficult to see that although there are different regions and different levels of sports development in colleges and universities, the problems in their development are in parallel. The main problem lies mainly in the lack of a specific organizational implementation system in the theoretical academic evaluation system [12], which still has great resistance in practical exploration. At present, the theory and practice of academic evaluation are not well integrated, and there is a phenomenon of disconnection. In the actual evaluation process, due to various limitations, the evaluation of the learning process is still ignored [13-14]. Moreover, the personality differences [15-18], personal progress [19], and non-intelligence factors [20] of students in the learning process are not well reflected in the evaluation.

3 Basic theory of conventional control

At present, the selection of the three parameters of PID control can only use the empirical method, and there is still no accurate method to find the value of the PID parameter [21].

The conventional closed-loop speed regulation control and its controller are dominated by PID control. The basis of PID control is proportional control; integral control can eliminate steady-state errors, but may increase overshoot; differential control can speed up the response speed of large inertial systems and reduce the tendency of overshoot. The control law can be summarized as [22]: $u (t) = K_{p} [e (t) + \frac{1}{T_{i}} \int_{0}^{t} e (t) dt + Td \frac{de (t)}{dt}]$ (1)

Among them, e (t) = r (t) - y (t) is the deviation between the controller design initial value r (t) and the measured value y (t). KP is the proportional coefficient, TI is the integral time coefficient, and TD is the differential time coefficient. Its transfer function form can be expressed as: $G (s) = \frac{U (s)}{E (s)} = K_{p} [1 + \frac{1}{T_{i} s} Tds]$ (2)

The Ziegler-Nichols empirical formula is usually used to calculate and select the specific parameters of PID. ${\begin{matrix} K_{p} = \frac{1.2 T}{KL} \\ T_{i} = 2 L \\ T_{d} = \frac{L}{2} \end{matrix}$ (3)

Among them, L is the capacity lag time. The capacity lag is generally caused by the resistance encountered during the transfer of materials or energy. In other word, it is the time required for the output signal to become stable again after the object is subjected to a certain effect. After the initial tuning of the parameters, the PID controller with initial tuning must be brought to the site for on-site commissioning. During the debugging process of the control system, the output value oscillates left and right, and the output is unstable. First, the integral time T_i is increased. If the output value is still fluctuating, the proportional coefficient K_p is selected to be adjusted. If the output value is found to fluctuate greatly, then the proportional coefficient K_p is increased first. If the output response time is longer, the output of increasing the derivative term T_d by a certain amount or decreasing the integral term T_i is selected.

Conventional PID parameter adjustment is only applicable to some known mathematical models and parameter control systems, and is artificially adjusted. At the same time, empirical factors account for a large proportion. After the PID parameter tuning is completed, the control system can only perform the following work in a fixed working mode until the PID control parameters are changed again.

4 Fuzzy PID control design

Fuzzy control is a combination of human language and mechanical language, so that the machine can recognize human working language habits and execute commands according to instructions. It imitates the thinking process of people and completes the control of the equipment by logically judging the data and operating trends collected at the time, which not only ensures the correctness of the operation process, but also has humanized self-regulation to prevent singularization operation [23].

The fuzzy controller obtains data and transmission signals through its I/O interface. Specifically, the air gap, speed, and torque signals collected by the sensor are transmitted to the fuzzy controller. Digital-to-analog conversion is performed on these data to convert specific data into analog signals that can be understood by the control components, so as to facilitate the subsequent fuzzification and deblurring, and then transfer the analog signals to the implementing components.

The range of the input and output data of the I/O port in the fuzzy set is usually divided into two categories. One is the domain of real-valued domains, which is called the basic domain, and the other is the quantitative domain, which refers to the range of the fuzzy sets. It is usually expressed as [- E, E]. In the process of fuzzification and defuzzification, the two domains need to be interacted with each other. Among them, the basic domain of the input error e is [- xe, xe], and its quantitative conversion formula is $Ke = \frac{n}{xe}$ ,where n is the maximum range of the domain of the fuzzy set.

The membership function is used to describe the converted fuzzy set, which is mainly used to indicate which part of the fuzzy set the loaded data prefers in the fuzzy set. Although in theory, any function curve can be used to express the numbers in the fuzzy set, the Gaussian function (Gaussmf), trapezoidal function (trapmf), and triangle function (Tdmf) are the main expressions in practical engineering applications. The form and function images are as follows [24]:

(1) Gaussian function $gaussian (x, c, σ) = e^{- \frac{1}{2} {(\frac{x - c}{σ})}^{2}}$ (4)

(2) Trapezoidal function $trapmf (x, a, b, c, d) = {\begin{matrix} 0 & x ⩽ a \\ \frac{x - a}{b - a} & a ⩽ x ⩽ b \\ 1 & b ⩽ x ⩽ c \\ \frac{d - x}{d - c} & c ⩽ x ⩽ d \\ 0 & x > d \end{matrix}}$ (5)

(3) Triangle function $triangle (x, a, b, c) = {\begin{matrix} 0 & x ⩽ a \\ \frac{x - a}{b - a} & a ⩽ x ⩽ b \\ \frac{c - a}{c - a} & b ⩽ x ⩽ c \\ 0 & c ⩽ x \end{matrix}}$ (6)

The Gaussian function is relatively sharp, and it is suitable for situations where the error input value is small, but it is almost impossible to handle high error input; the trapezoid function is gentle and the output is stable, and it is suitable for occasions with large error input value, but it is not sensitive to small errors and may not be recognized at all. In summary, the trapezoidal function that can handle both low-error input and high-error input is used as the membership function of fuzzy control.

Fuzzy variables are generally expressed as “negative big”, “negative middle”, “negative small”, “zero”, “positive small”, “positive middle”, “positive big”. These seven linguistic variables are generally expressed in the form of English abbreviations in fuzzy rules, that is [24-28]: ${NB, NM, NS, Z, PS, PM, PB}$

When writing the control program, we can either write it off the PC side or implement real-time on-site modification of internal rules through the PLC port. Intelligent control under various working conditions is realized by changing parameter values or increasing or decreasing control sentences.

The writing of fuzzy rules usually adopts the following generalized positive reasoning.

Input: If X is A′ and y is B′

Premise: If X is A and y is B, then Z is c

Conclusion: z is C′

Among them, A, B and C are control sets. A′, B′ and C′ are corresponding fuzzy language control sets, and x, y and z are elements in the set.

The maximum membership method is to select the maximum membership value as the output value of defuzzification, that is: $v_{0} = max μ_{v} (v), v \in V$ (7)

When the output value corresponding to the maximum membership degree is greater than one, then all the values corresponding to the maximum membership degree need to be averaged, and the final defuzzification result of the obtained amount is: $\begin{matrix} V_{0} = \frac{1}{N} \sum_{i = 1}^{N} v_{i} \\ v_{i} = max_{v \in V} (μ_{v} (v)) \end{matrix}$ (8)

Among them: N is the total number of the same maximum membership.

The center of gravity method

The center of gravity of the graph area enclosed by the membership function curve and the X axis of the abscissa is taken as the final deblurring result, that is $V_{0} = \frac{\int_{V} v μ_{v} (v) dv}{\int_{V} μ_{v} (v) dv}$ (9)

When there are multiple discrete thresholds of output quantization series, namely $V_{0} = \frac{\sum_{k = 1}^{m} v_{k} μ_{v} (v_{k})}{\sum_{k = 1}^{m} μ_{v} (v_{k})}$ (10)

Median method

The median method is also called the area bisection method, and it takes the median of all the deblurring results as the output value. It satisfies: $\int_{a}^{v_{0}} μ (v) dv = \int_{v_{0}}^{b} μ (v) dv$ (11)

That is to say, v₀ is the dividing line, a is the lower boundary, and b is the upper boundary. Meanwhile, the area between the μc′ (v) and z axes is equal to the left and right sides. Figure 4 shows the median diagram.

Fig. 1

PID control logic diagram.

Fig. 2

Fuzzy control logic diagram.

Fig. 3

Comparison of membership functions.

Fig. 4

Schematic diagram of the median method.

Compared with the maximum membership method and the median method, the output of the center of gravity method is smoother. Due to the solution method, the referenced solution information is also more comprehensive, which reduces the loss of data and changes in the input signal in a timely manner. Meanwhile, the response is in the output. Therefore, the center of gravity method is selected as the method of deblurring.

5 Particle swarm optimization

The basic definition of particle swarm optimization

The particle group consisting of n particles is set in D to find the optimal solution in space, where the position of the i-th particle is set to A and the speed is set to B. At the same time, the direction of the particle movement should consider the historical best position of the particle itself PBest and the best position for the group GBest. The actual meaning of the particle position is a feasible solution in the optimization problem.

The value of x is brought into the target problem, and the results are compared to evaluate the optimization degree of the feasible solution. In most cases, particle suitability can be used to refer to the pros and cons of feasible solutions. For example, if the value of the objective function is required to be the smallest, the feasible solution with the smallest particle suitability should be selected as the solution of the objective function.

Based on the three principles of motion, particles use the iterative formula to continuously update their position and velocity until the particles find the optimal solution under this condition. $v_{id}^{t + 1} = {wv}_{id}^{t} + c_{1} r_{1} (p_{id}^{t} - x_{id}^{t}) + c_{2} r_{2} (p_{gd}^{t} - x_{gd}^{t})$ (12) $x_{id}^{t + 1} = x_{id}^{t} + v_{id}^{t + 1}$ (13)

Among them, i = 1, 2, ⋯ , n represents the number of particles in the particle swarm; d = 1, 2, ⋯ , D represents the spatial dimension; w is the inertial weight; t is the number of iterations; c₁ and c₂ are learning factors; r₁ and r₂ are random numbers between [0, 1].

Formulas (12) and (13) are called standard particle swarm optimization algorithm iteration formulas.

It can be seen that the particle velocity is mainly composed of three parts. The first part is the inheritance of the previous iteration speed V_t, which represents the historical record of the particle movement; the second part is the particle’s optimal position V_p, which is the reference of the particle’s self-experience and is the guiding role of self-experience for the optimization process; the third part is the group’s best position V_g. The information sharing mechanism is the embodiment of the population advantage, which greatly helps the individual particle to find the optimal solution. Figure 5 shows the particle Speed vector synthesis

Fig. 5

Speed vector synthesis diagram.

Compared with other population algorithms and intelligent optimization algorithms, the most prominent advantage of particle swarm optimization is that under the premise of ensuring high accuracy of the results, fewer parameters need to be selected. Due to the information sharing mechanism between particles, the running speed is faster than other The intelligent control responds faster.

The parameters of particle swarm optimization algorithm are: particle swarm size n, inertia weight w, learning factors c₁, c₂, maximum speed V_max, minimum speed V_min, iteration number t, particle suitability ɛ.

The decreasing formula of inertia weight is as follows: $w = w_{start} - \frac{w_{start} - w_{end}}{t_{max}} \times t$ (14)

Among them, t is the current number of iterations, t_max is the maximum number of iterations, w_start is the initial inertia weight, and w_end is the final inertia weight.

(3) Learning factors c₁, c₂

The learning factors c₁ and c₂ are also called acceleration constant speed, and represent the proportional weight of the particle’s own experience and group experience in the process of motion exploration respectively.

When c₁ = 0, it means that the particle’s self-experience does not play a role in the process of exploring the optimal solution. At this time, the particle swarm only has the concept of group, and its convergence speed is faster than the standard particle swarm optimization algorithm, but when dealing with complex problems Because there is no empirical support of individual particles, it is easier to fall into the concentration of all particles around one or a few extreme values. The optimal solution obtained may be only a local extreme value but not an optimal solution.

When c₂ = 0, it means that there is no group information sharing between the particles, and the experience of the population does not play a role in the movement of the particle. At this time, the advantages of the population algorithm are not reflected, which means that all particles are in independent motion, not only the running speed is slow and It is difficult to find the optimal solution of the objective function.

(4) Maximum speed V_max

The maximum velocity V_max represents the maximum distance that the particle can fly in the D-dimensional space in each iteration. If the value of V_max is too large, the particle may miss the optimal solution area; if the value of V_max is too small, it will not only to slow down the solution speed, it is also possible to find only the local extrema instead of the global optimal solution. Generally, V_max is set to one-fifth of the difference between the upper and lower boundaries in the D-dimensional space. The mathematical expression is expressed as $V_{max} = (x_{max} - x_{min}) / 5$ (15)

(5) Minimum speed V_min

The minimum speed is the minimum distance of each particle movement, which is the guarantee of the operation speed. Usually, it is set to the opposite of the maximum speed, which is V_min = - V_max.

(6) Number of iterations t

The particle swarm optimization algorithm continuously seeks the optimal solution through the iterative updating of particles, and generally terminates the operation by setting the number of iterations.

(7) Particle suitability ɛ

The applicability can be simply summarized as the absolute value of the difference between the actual optimal solution and the theoretical optimal solution. It is judged whether the optimization goal has been reached by calculating the suitability.

Professor Carlisle and Professor Dozier summarized previous work experience, and then summed up a set of standardized formulas and standardized parameters for calculating particle swarm parameters. $v_{ij} = {\begin{matrix} K (\begin{matrix} v_{ij} + c_{1} r_{1 j} (p_{ij} - x_{ij}) + \\ c_{2} r_{2 j} (p_{gj} - x_{ij}) \end{matrix}) & x_{min} < x_{ij} < x_{max} \\ 0 & otherwise \end{matrix}}$ (15) $v_{ij} = {\begin{matrix} x_{ij} + v_{ij} & x_{min} < x_{ij} < x_{max} \\ x_{max} & x_{max} < x_{ij} < v_{ij} \\ x_{min} & x_{min} > x_{ij} + x_{max} \end{matrix}}$ (16) $K = \frac{2}{| 2 - C - \sqrt{C^{2} - 4 C} |}$ (17)

Among them, K is the shrinkage coefficient, and the optimal solution of the individual parameters is c₁ = 2.8, c₂ = 1.3, C = c₁ + c₂. Meanwhile, the total number of particle swarms must ensure the number, but also to ensure that the computing power can be satisfied. Generally, the value is g n = 30.

6 Physical education teaching evaluation system construction and performance analysis

With the support of the above algorithms and models, a physical education teaching evaluation system based on artificial intelligence fuzzy algorithm is constructed in this study. The system structure diagram is as follows:

Before teaching, we need to train students on how to use them. Through training, students can understand and master the methods and steps of the entire online course learning, including platform download and installation, registration and login, course selection, video viewing, reading and downloading of text materials, discussion of the use of message boards, completion of online assignments and quizzes. Through these trainings, students can better adapt to teaching and lay a solid foundation for the smooth development of subsequent teaching work.

After doing the above preparations, we can start experimental teaching. This study set up a control group and a test group, and the number of people of the control group and the test group is 60. Before the start of the experiment, the sports performance of the test group and the control group was evaluated. In this paper, 1000 meters was used as an example for comparison. The results obtained are shown in Table 1 and Fig. 7.

Table 1

Comparison table of 1000m grades between test group and control group before experimental teaching (s)

	Control group (s)	Test group (s)		Control group (s)	Test group (s)
1	251.1	250.0	31	281.0	283.5
2	251.4	252.2	32	283.4	285.3
3	253.1	256.4	33	284.2	286.4
4	255.5	258.4	34	284.6	286.6
5	255.6	260.2	35	286.6	288.6
6	256.8	261.7	36	287.7	289.7
7	257.3	262.9	37	287.9	290.0
8	258.7	263.4	38	288.4	295.1
9	259.7	263.7	39	289.3	295.3
10	261.0	263.7	40	290.0	295.5
11	261.0	264.4	41	290.3	297.0
12	261.0	265.0	42	292.1	301.2
13	262.4	265.2	43	295.1	301.9
14	263.8	265.5	44	295.1	302.7
15	265.9	265.7	45	296.4	302.9
16	266.5	265.7	46	298.3	304.9
17	267.1	265.9	47	298.5	308.2
18	267.4	267.5	48	298.6	309.6
19	269.4	268.9	49	300.1	309.7
20	269.7	270.1	50	302.3	310.7
21	270.1	271.9	51	305.3	311.5
22	271.8	273.4	52	306.6	312.3
23	272.5	275.5	53	307.7	312.9
24	273.7	275.9	54	309.4	313.7
25	274.1	276.8	55	311.1	315.6
26	275.2	276.8	56	315.2	316.6
27	275.6	278.1	57	316.6	318.0
28	276.3	280.5	58	318.0	318.6
29	277.3	280.7	59	318.6	319.6
30	279.3	282.3	60	318.6	319.8

Fig. 6

Physical teaching evaluation system based on artificial intelligence fuzzy algorithm.

Fig. 7

Comparison diagram of 1000-meter grades between the experimental group and the control group before experimental teaching (s).

As shown in Fig. 7, the results of the students in the test group and the control group are basically the same before the test, so it can be considered that the initial results of the two groups of students are the same. On this basis, experimental teaching can be carried out, and sports performance evaluation can be conducted after the end of the semester. The research results are shown in Table 2 and Fig. 8.

Table 2

Comparison table of 1000m grades between test group and control group after experimental teaching (s)

	Control group (s)	Test group (s)		Control group (s)	Test group (s)
1	212.5	210.8	31	248.6	229.5
2	212.8	211.2	32	251.6	229.7
3	213.0	212.0	33	252.6	229.9
4	217.1	212.7	34	252.7	230.3
5	217.2	212.9	35	257.0	233.2
6	217.9	213.4	36	257.5	233.6
7	220.3	214.1	37	257.7	235.1
8	222.1	214.5	38	258.4	235.5
9	222.4	214.9	39	258.8	236.6
10	223.9	215.2	40	258.8	237.6
11	224.2	215.9	41	261.0	239.0
12	224.8	215.9	42	261.6	239.2
13	228.8	218.2	43	262.0	240.6
14	230.0	219.1	44	264.2	240.8
15	231.3	219.4	45	264.8	241.7
16	232.6	219.5	46	269.4	242.5
17	232.9	219.6	47	274.9	243.7
18	234.1	220.6	48	274.9	243.9
19	234.9	221.1	49	275.1	244.0
20	235.1	223.0	50	277.5	245.0
21	236.0	224.0	51	277.5	245.1
22	236.4	224.1	52	278.8	246.1
23	239.6	227.0	53	279.5	246.3
24	240.1	227.0	54	279.7	246.8
25	240.6	228.2	55	279.8	247.0
26	240.7	228.5	56	280.4	248.3
27	243.4	228.7	57	281.4	248.5
28	243.5	228.8	58	281.5	249.2
29	246.3	228.9	59	282.9	249.6
30	247.9	229.0	60	284.9	249.6

Fig. 8

Comparison diagram of 1000m grades between test group and control group after experimental teaching (s).

As shown in Fig. 8, with the aid of the system model built by this study, the students in the test group have a grade higher than the students in the control group at 1000 meters. It can be seen that the physical education evaluation system based on artificial intelligence fuzzy algorithm constructed in this paper has certain practical effects.

6 Conclusion

Before the application of artificial intelligence models, in traditional classroom teaching, students only passively accepted the knowledge and skills transmitted by teachers. The knowledge and skills in textbooks are the only things students can learn. Moreover, tools such as multimedia projectors are just auxiliary tools to help teachers teach. The artificial intelligence sports teaching system is not near, it can evaluate the effect of physical education in real time and can effectively improve the classroom teaching efficiency of teachers. Meanwhile, combined with intelligent demonstrations, it can provide effective reference for students. Based on this, this paper analyzes the shortcomings of the traditional physical education teaching method with the support of the fuzzy system and combines with the actual needs of physical education to construct a physical education evaluation system based on artificial intelligence fuzzy algorithm. In addition, after constructing the overall structure of the model, experiment teaching is carried out through practical teaching methods, comparison is made by setting up a control combination test group, and the results of the experiment teaching are statistically analyzed. It can be seen that the physical teaching evaluation system based on artificial intelligence fuzzy algorithm constructed in this paper has certain practical effects.

Footnotes

Acknowledgments

This work was supported by National key research and development project. Project Name: “Application Verification and Demonstration of Ice and Snow Sports Equipment” Project No.: 2019YFF0302005, Project Leader: Yang Jintian.

References

Ahilan

, Manogaran

, Raja

, Kadry

, Kumar

S.N.

, Kumar

C.A.

and Murugan

N.S.

, Segmentation by fractional order darwinian particle swarm optimization based multilevel thresholding and improved lossless prediction based compression algorithm for medical images [J], IEEE Access 7 (2019), 89570–89580.

Appelbaum

L.G.

and Erickson

, Sports vision training: A review of the state-of-the-art in digital training techniques[J], International Review of Sport and Exercise Psychology 11(1) (2018), 160–189.

Bennour

, Teaching Practices and Student Action in Physical Education Classes: Perspectives for Teacher Education[J], Creative Education 6(10) (2015), 934–944.

Bulat

, Can

N.K.

, Arslan

Y.Z.

, et al., Musculoskeletal Simulation Tools for Understanding Mechanisms of Lower-Limb Sports Injuries[J], Current Sports Medicine Reports 18(6) (2019), 210–216.

Cutforth

and Belansky

E.S.

, A Community-Engaged Approach to Translating Research into Practice: A Physical Education Story[J], Progress in Community Health Partnerships Research Education & Action 9(4) (2015), 571–582.

Ezhilmaran

and Adhiyaman

, Fuzzy Approaches and Analysis in Image Processing [J], In Advanced Image Processing Techniques and Applications, 2017, pp. 1–31. IGI Global, 2017.

Hwang

E.R.

and Kim

T.Y.

, Intensification of the education of public health, hygiene, and martial arts during the Japanese colonial period (1937–1945)[J], Journal of Exercise Rehabilitation 14(2) (2018), 160–167.

Ivashchenko

O.V.

and Kapkan

O.O.

, Simulation of process of 14–15 years old girls’ training of light athletic and gymnastic exercises[J], Pedagogics, Psychology, Medical-Biological Problems of Physical Training and Sports (8) (2015), 32–40.

Khari

, Garg

A.K.

, Gonzalez-Crespo

and Verdú

, Gesture Recognition of RGB and RGB-D Static Images Using Convolutional Neural Networks [J], International Journal of Interactive Multimedia and Artificial Intelligence 5(Regular Issue) (2019), 22–27.

10.

Khudolii

O.M.

, Ivashchenko

O.V.

, Iermakov

S.S.

, et al., Computer simulation of junior gymnasts’ training process[J], Science of Gymnastics Journal 8(3) (2016), 215–228.

11.

, Hu

, Xue

, et al., Preliminary Teaching Reform of Experiment of Sports Human Science Based on Virtual Simulation Technology[J], Journal of Huaibei Normal University (Natural Science Edition) (2) (2016), 17.

12.

Lopatiev

, Ivashchenko

, Khudolii

, et al., Systemic approach and mathematical modeling in physical education and sports[J], Journal of Physical Education and Sport (JPES) 17(1) (2017), 146–155.

13.

Mullen

, Highton

and Twist

, The internal and external responses to a forward-specific rugby league simulation protocol performed with and without physical contact[J], International Journal of Sports Physiology and Performance 10(6) (2015), 746–753.

14.

Mulligan

, Lohse

K.R.

and Hodges

N.J.

, An action-incongruent secondary task modulates prediction accuracy in experienced performers: evidence for motor simulation[J], Psychological Research 80(4) (2016), 496–509.

15.

Murase

, Yokoyama

, Fujii

, et al., Baseball Catching Patterns Differ According to Task Constraints[J], Advances in Physical Education 06(3) (2016), 151–157.

16.

Owen

P.D.

and King

, Competitive balance measures in sports leagues: The effects of variation in season length[J], Economic Inquiry 53(1) (2015), 731–744.

17.

Payne

, Mitchell

, Bibb

, et al., The evaluation of new multi-material human soft tissue simulants for sports impact surrogates[J], Journal of the Mechanical Behavior of Biomedical Materials 41 (2015), 336–356.

18.

Prathik

, Anuradha

and Uma

, Survey on Spatial Data Mining, Challenges and Its Applications [J], Journal of Computational and Theoretical Nanoscience 15(9-10) (2018), 2769–2776.

19.

Rein

and Memmert

, Big data and tactical analysis in elite soccer: future challenges and opportunities for sports science[J], SpringerPlus 5(1) (2016), 1–13.

20.

Ridderinkhof

K.R.

and Brass

, How kinesthetic motor imagery works: a predictive-processing theory of visualization in sports and motor expertise[J], Journal of Physiology-Paris 109(1-3) (2015), 53–63.

21.

Ngan

R.T.

, Ali

, Fujita

, Giang

N.L.

, Manogaran

and Priyan

M.K.

, A New Representation of Intuitionistic Fuzzy Systems and Their Applications in Critical Decision Making [J], IEEE Intelligent Systems.

22.

Saleem

and Khattak

M.I.

, Deep Neural Networks for Speech Enhancement in Complex-Noisy Environments [J], International Journal of Interactive Multimedia and Artificial Intelligence (2020), 84–90.

23.

Seo

S.W.

, Yun

, Kim

M.G.

, et al., Screen-Based Sports Simulation Using Acoustic Source Localization[J], Applied Sciences 9(15) (2019), 2970.

24.

Sghaier

, Elandoulsi

, Mami

, et al., Physical Education Teacher’s Training in Swimming under the Joint Didactic Action[J], Creative Education 6(22) (2015), 2433–2437.

25.

Wang

, Zhou

, Yang

, Sun

and Hsu

C.-H.

, Service Composition in Cyber-Physical-Social Systems [J], IEEE Transactions on Emerging Topics in Computing.

26.

de Souza

Í.M.M.

and Silva

C.L.D.

, Body Practices and Brazilian Culture: Pedagogical Contributions to Physical Education Professionals[J], Procedia Social & Behavioral Sciences 174 (2015), 434–440.

27.

Sudin

M.N.

, Abdullah

S.S.

and Nasudin

M.F.

, Humanoid Localization on Robocup Field using Corner Intersection and Geometric Distance Estimation [J], International Journal of Interactive Multimedia and Artificial Intelligence 2019; 50–56

28.

Namasudra

, Devi

, Kadry

, Sundarasekar

and Shanthini

, Towards DNA based data security in the cloud computing environment, Computer Communications 151 (2020), 539–547.

29.

Tivener

K.A.

and Gloe

D.S.

, The effect of high-fidelity cardiopulmonary resuscitation (CPR) simulation on athletic training student knowledge, confidence, emotions, and experiences[J], Athletic Training Education Journal 10(2) (2015), 103–112.

30.

Yang

, The Simulation of Table Tennis during the Course of Sports[J], Caribbean Journal of Science 52(4) (2019), 1561–1564.