Research on mechanism of intelligent manufacturing equipment based on eye tracking control

Shi-long XIAO，Rong MAO，Bo CHEN

（1 Shenyang university of technology，Liaoning 110159，China）

（2 School of Computer Science and Technology，Shandong University of Technology，Zhibo 255000，China）

Abstract：The intelligent manufacturing equipment industry has long placed emphasis on technology，which makes the manufacturing technology increasingly mature，but its external form is relatively backward.In response to this problem.A test study was carried out on different kinds of experimental subjects.First，the eye tracking control algorithm is combined with VR simulation technology，a local adjustment effect experiment was performed on the host equipment of the three microelectronics technology.Then，by analyzing the evaluation indexes such as the total fixation duration，average fixation duration and fixation frequency of three equipment plans to determine the core visual orientation，the pros and cons of the scheme can be analyzed through intuitive data.Finally，the best decision-making control algorithm is adopted to improve the design efficiency of industrial intelligent manufacturing construction.The experimental results verify the feasibility of the eye tracking control algorithm in intelligent manufacturing equipment，and can significantly improve the quality and efficiency of intelligent manufacturing equipment.

Key words：Eye tracking technology，Intelligent manufacturing equipment，Intelligent control，Chip manufacturing，Virtual reality technology，Automation industry

Intelligent manufacturing equipment refers to manufacturing equipment with the functions of perception，analysis，reasoning，decision-making，and control.It is the integration and deep fusion of advanced manufacturing technology，information technology and intelligent technology.Products with intelligent functions promote the upgrading of manufacturing industry to meet the needs of the intelligent era，and the intelligent manufacturing industry also affects the update of intelligent quality［1］.Intelligent manufacturing is in a critical period of large-scale promotion，transformation and upgrading.In order to enhance China’s overall industrial automation strength and the international competitiveness of the manufacturing industry，and change the situation that China’s automation and manufacturing industry have been in the middle and low end of the industrial chain for a long time，with unreasonable structure，low efficiency and lack of innovation ability［2-4］，which should optimize and upgrade the manufacturing of intelligent manufacturing equipment itself，and from the beginning of the preliminary program screening，deliberated to meet the social needs of intelligent manufacturing equipment.At present，there have been relevant studies on the cognitivecharacteristics and feasibility analysis of web page and interface layout using eye movement tracking technology［5-7］.However，there are few intelligent screening schemes for intelligent manufacturing equipment by eye tracking and virtual reality technology.In this paper，taking three machine renderings as an example.Are taken as examples，then extracted the subject’s eye movement data through VR simulation technology and eye tracking technology，conducted depth analysis of the heat map and trajectory map in terms of gaze point，gaze duration，gaze track，area of interest，etc.，and Combined with questionnaire method as data support，the intelligent screening system of intelligent manufacturing equipment scheme is formed to select the best scheme.This creative problem-solving process can objectively screen out the best solution and meet the market demand.

1 Eye movement tracking technology and control

Eye movement technology is also called eye movement recording technology or eye movement tracking technology.Take the visual sensory organ as the main body，use the eye movement recorder to record and collect various visual focus and visual dynamic indicators and data of the human eye movement in the visual scene，Obtain experimental data through visual laws to analyze people’s mental activities and behavioral motivations［8-13］.

The human eye’s visual system is divided into perceptual mechanisms such as color and shape，space and movement.An eye tracker is a device that can track and detect the movement of the eyeball and the position of the eyeball，through the analysis of eye movement data，deduce the correlation between eye movement rules and human mental activity.Eye movement not only reflects the trajectory of the eye，but also reflects that the characteristics of eye movement patterns are closely related to the process of brain information processing.Eye movement is divided into three categories， namely fixation， trajectory movement and twitching of the eyelid.The eye-tracking device can be used to visually observe the movement track of the subjects’eye fixation points during the visual response，Where and when the eyes stay，And you can get the time of each fixation point，and the percentage of watching time in a certain period of time.Since humans’high acuity is limited to the central concave，eye movements are needed to choose what to see clearly，so recording eye movements tells us what the brain thinks it needs under any circumstances.The way the eye moves is affected by what we do and what we see［14］.Therefore，we can integrate a series of data such as the average length of the subjects’gaze point，the area of interest in a specific area，the total length of staying in an area of interest，blink data，the distance between points of interest，and the average pupil diameter to derive Show the participants’psychological activities and visualize the research results.Through the eye movement experiment，the gaze trajectory of the subjects can be accurately captured，and the degree of attention of the subjects in each area can be analyzed.

2 Screening experiment of smart manufacturing equipment schemes under eye tracking technology

2.1 Experiment process

Arrange the appearance design drawings of the three KS-C300 chip manufacturing machines designed for Shenyang Xinyuan Microelectronics Equipment Co.，Ltd.in Fig.1 from left to right，Respectively，the plan of picture 1，the plan of picture 2，and the plan of picture 3，and use this as experimental material，Synchronize advanced virtual reality technology，eye movement technology and human-computer interaction，Put the three prepared designs in a virtual reality environment using unity software，Select 12 people as the subjects，6 general groups and 6 trained groups，to explain to the subjects in advance，but do not mention any words that interfere with the judgment of the subjects，let these 12 subjects wear the TOBII eye tracker adjusted in advance and Ergolab software equipment to observe the plan for 1 minute，and then extract the data of the heat map and trajectory map.After collecting the eye tracking data，each subject immediately conducted a questionnaire survey for subjective evaluation.Finally，the eye tracking data and the data of the questionnaire were sorted and analyzed，and the optimal plan was screened out to manufacture intelligent manufacturing equipment that best meets the market demand.

Fig.1 Machine renderings designed by Shenyang Xinyuan Microelectronics Equipment Co.，Ltd.

2.2 Laboratory equipment

This experiment uses TOBII eye tracker and Ergolab software equipment，which is a human-machine environment test cloud platform that combines 3D virtual reality eye tracking and recording and analysis modules，this platform is based on the principle of visual interaction and deeply integrates eye tracking technology and virtual reality technology to construct visual research scenes，form eye tracking records and generate modular analysis data.When the eye sees things alone，it sees nothing at all.But perception helps the eye to choose what the mind needs to see，to show its choice through the mind，to make what it is aware of what it needs to see［15］，Through the experiment of this device，experimental data can be proposed to more accurately analyze the visual concerns of subjects and their emotional and psychological reactions to visual stimuli，and better reflect that this device is the product of the combination of eyes and psychology.

2.3 The experimental materials

The experimental objects are three machine renderings designed for Shenyang Xinyuan Microelectronic Equipment Co.，Ltd.Renderings according to customer needs，industry characteristics，visual fatigue，and other factors to consider the overall color［16-17］，After comprehensive consideration of the existing modeling design research methods，hierarchical structure，formal beauty rules，multi-perspective image differences，charm engineering，product DNA inheritance and variability and product semantic analysis［18-20］，the following three schemes are finally obtained.These three schemes only design the appearance of the host，as shown in Fig.2.

2.4 Experimental group and control mechanism

The experimental group consisted of 12 people，divided into two groups and six groups，one of which was the general social group and the group with professional aesthetic training：

Data collection for groups with professional aesthetic training：those who have received professional aesthetic training have a comprehensive and systematic understanding of all kinds of products on the market，and evaluate the scheme from a professional design standard and high aesthetic perspective.And the use of positive thinking and reverse thinking two-way thinking to find the design of innovation points and shortcomings.

Fig.2 Shenyang Xinyuan microelectronic equipment Co.，Ltd.Design of the division of the machine renderings

Data collection from the general population：the audience of intelligent manufacturing equipment is ordinary people.By collecting data from them，we can better grasp their psychological demands and aesthetics，intuitively understand their psychological needs，and reflect their cognitive psychological status of the scheme，truly achieve the people-oriented design points.

3 The results of the study

3.1 Analysis of visual visualization results of virtual Reality line of sight interaction

First，filter the collected eye tracking graphs.There were problems with data collection of two subjects，excluding the data of these two subjects，the data of the remaining 10 subjects was valid at last，they are 5 ordinary social groups and 5 students with professional aesthetic training respectively.Secondly，the eye movement points of each subject were screened，and the eye movement points that did not correspond to the diagram were excluded，and only the effective eye movement points that fell on each design plan were extracted.Finally，calculate the duration of these eye movement points and make the following data figure and tables.From Fig.3 and Table 1，it can be seen intuitively that the design plan of picture 2 obtained the longest fixation duration.

Fig.3 Histogram comparison of fixation duration of each subject to each figure

It can be intuitively seen from the data in Table 1 that among the 10 valid data，only participant 1-1 has the longest fixation time on the plan of picture 1，and this subject was part of the general population.The remaining 9 data clearly showed that these 9 subjects gazed at picture 2 for the longest time during a oneminute period.From the perspective of fixation time，subjects showed more interest on plan 2，so their eyes unconsciously stayed in picture 2 region for a longer time.

Table 1 The fixation duration data scale of each subject to each figure

The total fixation time is the sum of the gaze time of human eyes on a certain area.This experiment is to screen out the best scheme，Therefore，the data in Table 1 are calculated according to the total fixation time obtained by each scheme，the fixation duration of 10 subjects on the plan of picture 1 was added to obtain a total fixation duration of 99.8 s in the plan of picture 1；The fixation duration of 10 subjects on the plan of picture 2 was added to obtain a total fixation duration of 195.1s in the plan of picture 2；The fixation duration of 10 subjects on the plan of picture 3 was added to obtain a total fixation duration of 64.1 s in the plan of picture 3.It can be seen that the total fixation time of the 10 subjects obtained is the longest in Fig.2.Judging from the total fixation time，the plan of picture 2 was more attractive to the participants.

3.2 Heat map analysis

From the 10 heat map samples，it can be clearly observed that in the visualization results of eye movement，the longer fixation time and the more fixation times，the deeper the appearance of red in the visual activation area.As shown in Fig.4，Through the visual results of the subjects’eye movements，the user’s area of interest can be intuitively judged，Red has the largest proportion in the second design which shows that the subjects’focus is mainly on this design，and most of them are concentrated in the middle of the machine，this is the strongest interest area of the subjects，and the biggest difference between these three solutions is also in the middle，In short，from a visual search perspective，Participants have a higher degree of visual search for the plan of picture 2，indicating that the subjects prefer the plan of picture 2.On the contrary，the plan of picture 3 has a lower degree of visual search，indicating that the plan of picture 3 is not popular and recognized by the public.

Fig.4 Gaze point trajectory diagram

3.3 Track control chart analysis

From the fixation point trajectory graph，you can see the subject’s first fixation point，gaze order and other information during the experiment，and show the subject’s browsing situation and the order of the subject’s gaze in numerical form.Through the analysis of the gaze point trajectory in Fig.5，we can see that except for the first fixation points of participant 1-4 and 2-3 on the plan of picture 1，the other 8 subjects are on the plan of picture 2.

Fig.5 Gaze point heat map

Table 2 Gaze point quantitative table

Trajectory map is the data that dynamically reflects the order of the gaze point of the human eye，the size of the track point represents the length of fixation time，The bigger the point，the longer the fixation time，By quantify the gaze point trajectory map，the gaze point quantitative table of Table 2 is formed.In this experiment，the number of eye movement points also indicates the attractiveness of the program to thesubjects，It can be seen from Table 3，that Each subject has the most eye movement points on the scheme in Fig.2，Secondly，it falls on the scheme of Fig.1，The least eye movement is the scheme in Fig.3，which means that the scheme in Fig.2 is the most attractive.

3.4 Questionnaire analysis

After the eye movement test，the subjects were asked to give the questionnaire scores，and the three schemes were given the full score of 5.By analyzing and comparing the questionnaire scores in Table 3，it can be concluded that the total score value of the plan of picture 1 is 33，that of the plan of picture 2 is 46，and that of the plan of picture 3 is 24.In addition，except for the highest score 5 given by participant 1-1 for the plan of picture 1，the same score 5 is given by participant 2-2 for the plan of picture 1 and the plan of picture 2，As for the scheme in Fig.3，the remaining participant all gave a higher score for the plan of picture 2.It can also be seen from this set of data that the 10 people were more satisfied with the plan of picture 2.

Table 3 Questionnaire

4 Conclusion

Through the overall analysis of eye movement data and control mechanism，we can see that compared with the other two schemes，the results in Fig.2 are of the highest concern，To a certain extent，it can more effectively help enterprises to better carry out eye tracking control and intelligent manufacturing equipment scheme mechanism screening，effectively control and decide on the best method and evaluate the selection method that best meets the needs and satisfaction of the public，And in the virtual environment of immersion to avoid category defects in advance，control the cost.The best way to analyze the solution from the perspective of more professional decision making can greatly reduce the cost of product research and development and control the advantages of products more effectively，To show the development of more effective intelligent manufacturing equipment technology by automatic selection method.