Time prediction modeling method under dynamic target selection

Abstract

The invention discloses a time prediction modeling method under dynamic target selection. The method comprises the following steps that a dynamic target selection experiment platform is built, a dynamic target selection task is formalized, and core variables such as the initial interval, width and speed of a target and experiment parameters are determined; key influence factors of a dynamic target selection task are analyzed, and a time prediction hypothesis is provided to cover specific influences of target parameters and interaction of the target parameters; combining the characteristics of static and motion stages, expanding an existing theoretical model, and establishing a comprehensive time prediction model; different target conditions are set through an experimental platform, a testee is recruited to complete a dynamic target selection task, task completion time data are collected, experimental data are analyzed, a proposed hypothesis is verified, a time prediction model is fitted, and a scientific basis is provided for setting and design optimization of target parameters in a dynamic interaction system. The method is suitable for a dynamic interaction scene, and is beneficial to improving the system design rationality and the user operation efficiency.

Description

Technical Field

[0001] The present invention relates to the field of human-computer interaction technology, and in particular to a time prediction modeling method under dynamic target selection, which is suitable for user behavior modeling and time prediction tasks in a dynamic interaction system. Background Art

[0002] With the popularity of dynamic interactive systems in various scenarios, such as computer games, drone control, traffic management, and video surveillance, the task performance of users when interacting with dynamic targets has become a design key. Among them, dynamic target selection tasks require users to make rapid selections and precise operations among moving targets, a process that usually involves complex feedback-decision closed-loop control. Although existing static target selection models (such as Fitts's law) can well predict the selection time of static or uniform targets, they show obvious limitations when dealing with complex dynamic scenarios (such as non-uniform motion or target acceleration).

[0003] At present, time prediction for dynamic target selection tasks mainly relies on experience-driven design methods, which adjust system performance through a large number of user tests. However, this method is not only time-consuming and costly, but may also be limited by the experimental environment and sample conditions, resulting in a lack of generalization of the results. In addition, existing models fail to fully consider the complex interaction between target motion characteristics and user behavior patterns, especially when it comes to non-uniform motion or multi-target selection tasks, and it is difficult to provide accurate time prediction. Therefore, there is an urgent need for a theoretical time prediction modeling method that can combine the motion characteristics of the target and the user's operation behavior to provide more scientific and efficient guidance for the design of dynamic interactive systems. Summary of the invention

[0004] The present invention provides a time prediction modeling method under dynamic target selection, which solves the limitations of the prior art in processing dynamic target tasks, especially for complex scenarios such as non-uniform motion and target acceleration. The existing methods rely on experience-driven and a large number of user tests, which are time-consuming and costly, and lack generalization. The present invention combines the target motion characteristics with the user operation behavior to construct a time prediction model, which provides a scientific basis for the target parameter setting and interaction design of the dynamic interactive system, and improves the system performance and user operation efficiency.

[0005] To achieve the above technical objectives, the present invention provides the following technical solutions: a time prediction modeling method under dynamic target selection, comprising the following steps:

[0006] S1. Build a dynamic target selection experimental platform, formalize the dynamic target selection task, and define the core variables and parameter settings of the experimental scenario;

[0007] S2. Based on relevant research, key influencing factors and hypotheses for dynamic target selection time prediction are proposed, including the specific manifestations of target initial spacing, width, speed and their interactions;

[0008] S3. Construct a dynamic target selection time prediction model by extending the existing theoretical model, and form a comprehensive prediction model by combining the characteristics of the static stage and the moving stage;

[0009] S4. Based on the dynamic target, select the experimental platform, set the experimental conditions and recruit subjects to perform the task, collect task completion time data, verify the hypothesis, and finally fit the prediction model and obtain the prediction result of the task completion time, so as to provide a basis for the design optimization of the target parameters in the dynamic interaction system design;

[0010] Preferably, the S1 specifically includes the following steps:

[0011] S11 builds a dynamic target selection experimental platform and sets up experimental scenarios to simulate user interaction tasks in a dynamic target environment;

[0012] The experimental platform includes a dynamic target generation module, a data acquisition module and an input control module, which are used to provide target generation, data recording and user operation input functions required for the experiment;

[0013] The dynamic target generation module S12 is used to generate the static targets and moving targets required for the experiment, and set the initial spacing, target width and moving target speed of the targets according to the experimental requirements;

[0014] The initial spacing A is the horizontal distance between the stationary target and the edge of the moving target, which is used to control the moving distance requirement of the cursor; the target width W is the effective area size of the dynamic target, which is used to affect the difficulty of target selection; the target speed V is the speed of the dynamic target moving at a uniform speed in the experiment, which is used to simulate the moving characteristics of the target in a dynamic scene;

[0015] S13 In the experimental task, the user first clicks the stationary target with the mouse to trigger the moving target to start moving in the specified direction at a preset speed;

[0016] The moving target is the effective target of the task, and the user needs to track and click the effective area of ​​the moving target in real time to complete the task interaction operation;

[0017] The task completion time T is defined as the total time from when the user clicks the stationary target to when the user successfully clicks the moving target, which is an important measurement indicator of the experimental data;

[0018] The S14 data acquisition module records key data during user operation in real time, including initial spacing, target width, target speed, and task completion time;

[0019] The S15 input control module is used to receive the user's mouse operation signal to ensure the synchronous operation of the target generation and data acquisition modules to ensure the accuracy and integrity of the experimental data;

[0020] Through the above-mentioned experimental platform and task settings, the formal modeling of the dynamic target selection task is completed, the core variables required for the experiment and the parameter settings of the experimental scenario are clarified, and data support and technical basis are provided for the construction of the subsequent time prediction model.

[0021] Preferably, S2 specifically includes the following steps:

[0022] S21 is based on the characteristics of dynamic target selection tasks and assumes that the initial target spacing, target width, target speed, and the interaction between them have an impact on the task completion time;

[0023] Based on the above influence relationships, S22 proposes key assumptions for dynamic target selection time prediction, and clarifies the specific effects of target initial spacing, target width, target speed and their interactions on task completion time.

[0024] Preferably, S3 specifically includes the following steps:

[0025] S31 Based on the assumptions in S2, the selection phase can be divided into the target static phase and the movement phase;

[0026] In the static stage, the task completion time is affected by the initial target spacing and target width. Based on the existing theoretical model of static target selection, a time prediction formula for the static stage is proposed.

[0027] In the motion phase, the task completion time is related to the target's moving speed and target width. Based on the characteristics of dynamic targets, a time prediction formula for the motion phase is proposed.

[0028] S32 proposes a comprehensive time prediction model to simultaneously consider the impact of stationary targets and moving targets on time prediction. By introducing a trade-off factor u, the model is used to adjust the relative effects of the stationary stage and the moving stage, uniformly express the time prediction relationship of the dynamic target selection task, and obtain a comprehensive time prediction model.

[0029] Preferably, the S4 specifically includes the following steps:

[0030] S41 recruits a certain number of subjects for the experiment. The subjects will perform tasks on the experimental platform according to pre-set conditions. The start of each task is triggered by the user clicking on a stationary target, and then the target starts to move at the set speed and direction. The user needs to track and click on the effective area of ​​the target in real time to complete the task. The system records the completion time and related data of each task in real time.

[0031] Preferably, the subject has normal vision or corrected to normal vision, possesses good operational skills, and can accurately complete the target selection task.

[0032] S42 performs strict preprocessing on the collected data; outliers are cleaned to ensure data quality, extreme values, i.e. values ​​that deviate from the mean by more than three standard deviations, are eliminated, and then repeated measures ANOVA is performed on the data at a confidence level of α = 0.05;

[0033] Preferably, when the data violate the sphericity assumption, the data are subjected to Greenhouse-Geisser correction during analysis;

[0034] S43 conducted a test of inter-subject effects, analyzing the differences in task completion time between target initial spacing, target width, and target speed, and evaluating the impact of the different factors on the experimental results;

[0035] S44 conducted an interaction effect test to analyze the comprehensive impact of the interaction between the initial target spacing, target width, and target speed on the task completion time, and observed the joint effect of each variable on the task completion time;

[0036] S45 verifies the correctness of the hypothesis by testing and analyzing the data, and adjusts the parameters of the prediction model based on the results to ensure that the model can accurately predict the task completion time;

[0037] Based on the test results and experimental data, a fitting method is applied to obtain a dynamic target selection time prediction model.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] The present invention provides a time prediction modeling method under dynamic target selection, which solves the problem of low operation efficiency caused by unreasonable target parameter setting in dynamic target selection tasks.

[0040] The present invention accurately predicts the user's task completion time based on the characteristics of the targets such as the initial spacing, width and speed, thereby guiding the optimal setting of parameters such as the position layout and movement speed of the targets in the dynamic interactive system, and improving the fluency and accuracy of the user's selection operation.

[0041] The present invention is applicable to a variety of dynamic interaction scenarios, can realize the scientific layout of target presentation, reduce user operation delay, and improve the response speed of the system and the overall interaction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 The overall flow chart of the time prediction modeling method under dynamic target selection;

[0043] Figure 2Schematic diagram of the experimental platform for constructing dynamic targets;

[0044] Figure 3 Schematic diagram of the dynamic target selection task experiment interface;

[0045] Figure 4 Schematic diagram of experimental data collection and model hypothesis verification. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] The present invention proposes a time prediction modeling method under dynamic target selection, comprising the steps of:

[0048] S1. Build a dynamic target selection experimental platform, formalize the dynamic target selection task, and define the core variables and parameter settings of the experimental scenario;

[0049] S2. Based on relevant research, key influencing factors and hypotheses for dynamic target selection time prediction are proposed, including the specific manifestations of target initial spacing, width, speed and their interactions;

[0050] S3. Construct a dynamic target selection time prediction model by extending the existing theoretical model, and form a comprehensive prediction model by combining the characteristics of the static stage and the moving stage;

[0051] S4. Based on the dynamic target, the experimental platform is selected, the experimental conditions are set and subjects are recruited to perform the task, the task completion time data is collected, the hypothesis is verified, and finally the prediction model is fitted and the prediction result of the task completion time is obtained, which provides a basis for the design optimization of the target parameters in the dynamic interaction system design.

[0052] The S1 construction of a dynamic target selection experimental platform specifically includes the following steps:

[0053] S11 builds a dynamic target selection experimental platform and sets up experimental scenarios to simulate user interaction tasks in a dynamic target environment;

[0054] The experimental platform includes a dynamic target generation module, a data acquisition module and an input control module, which are used to provide target generation, data recording and user operation input functions required for the experiment;

[0055] The dynamic target generation module S12 is used to generate the static targets and moving targets required for the experiment, and set the initial spacing, target width and moving target speed of the targets according to the experimental requirements;

[0056] The initial spacing A is the horizontal distance between the stationary target and the edge of the moving target, the target width W is the effective area size of the dynamic target, and the target speed V is the speed of the dynamic target moving at a uniform speed in the experiment;

[0057] S13 In the experimental task, the user controls the mouse to click on the stationary target on the left side that meets the target width W, which will trigger the moving target on the right side with the same width to move quickly in the specified direction. The moving target is the effective target of the task. The user needs to track and click the effective area of ​​the moving target in real time to complete the task interaction operation;

[0058] The task completion time T is defined as the total time from when the user clicks the stationary target to when the user successfully clicks the moving target, which is an important measurement indicator of the experimental data;

[0059] The S14 data acquisition module records key data during user operation in real time, including initial spacing, target width, target speed, and task completion time;

[0060] The S15 input control module is used to receive the user's mouse operation signal to ensure the synchronous operation of the target generation and data acquisition modules to ensure the accuracy and integrity of the experimental data;

[0061] Through the above-mentioned experimental platform and task settings, the formal modeling of the dynamic target selection task is completed, the core variables required for the experiment and the parameter settings of the experimental scenario are clarified, and data support and technical basis are provided for the construction of the subsequent time prediction model.

[0062] Among them, the key influencing factors and assumptions proposed in S2 for dynamic target selection time prediction specifically include the following steps:

[0063] S21 is based on the characteristics of dynamic target selection tasks and assumes that the initial target spacing, target width, target speed, and the interaction between them have an impact on the task completion time;

[0064] Based on the above influence relationships, S22 proposes key assumptions for dynamic target selection time prediction, and clarifies the specific effects of target initial spacing, target width, target speed and their interactions on task completion time.

[0065] The key assumptions are as follows:

[0066] The task completion time increases as the initial target spacing increases, and the cursor needs to move a longer distance, resulting in an increase in the time required for the operation;

[0067] The task completion time is shortened as the target width increases, and users are more likely to complete the selection operation within a larger target area;

[0068] The task completion time increases with the increase of the target speed. The faster the target moves, the faster the user needs to operate, and the completion time is correspondingly extended.

[0069] There is an interaction between the initial target spacing and the target width. When the target spacing is larger and the target width is smaller, the task completion time is significantly prolonged.

[0070] There is also an interaction between target speed and target width. When the target speed is faster and the target width is smaller, the user's operation difficulty increases and the task completion time also increases.

[0071] The step S3 of constructing a dynamic target selection time prediction model specifically includes the following steps:

[0072] S31 Based on the assumptions in S2, the selection phase can be divided into the target static phase and the movement phase;

[0073] In the static stage, the task completion time is proportional to the initial target spacing and inversely proportional to the target width. Based on the existing theoretical model of static target selection, a time prediction formula for the static stage is proposed.

[0074] In the motion phase, the task completion time is proportional to the target's moving speed and inversely proportional to the target's width. Based on the characteristics of dynamic targets, a time prediction formula for the motion phase is proposed.

[0075] S32 proposes a comprehensive time prediction model to consider the influence of both static and moving targets on time prediction. By introducing a trade-off factor u, the model is used to adjust the relative effects of the static phase and the moving phase, and uniformly express the time prediction relationship of the dynamic target selection task to obtain a comprehensive time prediction model.

[0076] The S4 experimental data collection and model hypothesis verification specifically include the following steps:

[0077] S41 recruited 32 subjects for the experiment, with an age range of 18-34 years old. Among them, there were 16 males and 16 females; 10 undergraduates, 12 postgraduates, and 10 employed; 11 were between 18 and 22 years old, 9 were between 22 and 26 years old, and 12 were over 26 years old. The subjects had normal vision or corrected to normal vision, good operational skills, and could accurately complete the target selection task.

[0078] The subjects will perform tasks on the experimental platform according to pre-set conditions. The start of each task is triggered by the user clicking on a stationary target, and then the target starts to move at the set speed and direction. The user needs to track and click on the effective area of ​​the target in real time to complete the task. The system records the completion time and related data of each task in real time.

[0079] S42 performs strict preprocessing on the collected data; outliers are cleaned to ensure data quality, extreme values, i.e. values ​​that deviate from the mean by more than three standard deviations, are eliminated, and then repeated measures ANOVA is performed on the data at a confidence level of α = 0.05;

[0080] Preferably, when the data violate the sphericity assumption, the data are subjected to Greenhouse-Geisser correction during analysis;

[0081] S43 conducted a test of inter-subject effects, analyzing the differences in task completion time between target initial spacing, target width, and target speed, and evaluating the impact of the different factors on the experimental results;

[0082] Pairwise comparisons showed significant differences between all pairs; the target initial spacing A(F 3,36 =16.295, P<0.001) had a significant effect on the task completion time. As the initial target spacing increased, the task time increased significantly. Pairwise comparisons showed significant differences between all pairs; target width W (F 3,36 =1251.797, P<0.001) had a significant effect on the task completion time. As the target width increased, the task time decreased significantly. Pairwise comparisons showed significant differences between all pairs; target speed V (F 4,36 =39.528, P<0.001) had a significant impact on the task completion time. As the target speed continued to increase, the task time increased significantly.

[0083] S44 conducted an interaction effect test to analyze the comprehensive impact of the interaction between the initial target spacing, target width, and target speed on the task completion time, and observed the joint effect of each variable on the task completion time;

[0084] A×W(F 9，36 =4.086, P<0.001) had a significant interaction effect on task completion time. This indicates that their effects on the endpoint distribution are not independent. 12,36 =8.129, P<0.001) had a significant interaction effect on task completion time, indicating that their effects on task completion time were not independent.

[0085] S45 verifies the correctness of the hypothesis by testing and analyzing the data, and adjusts the parameters of the prediction model based on the results to ensure that the model can accurately predict the task completion time;

[0086] Based on the test results and experimental data, the dynamic target selection time prediction model was obtained by applying the fitting method. The model was fitted with the data, and the results showed that R 2 The value is 0.936, and the model fits well.

[0087] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the core idea of ​​the present invention. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present invention, the present invention can also be improved and modified, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. A time prediction modeling method under dynamic target selection, characterized in that , including the following steps: S1. Build a dynamic target selection experimental platform, formalize the dynamic target selection task, and define the core variables and parameter settings of the experimental scenario; S2. Based on relevant research, key influencing factors and hypotheses for dynamic target selection time prediction are proposed, including the specific manifestations of target initial spacing, width, speed and their interactions; S3. Construct a dynamic target selection time prediction model by extending the existing theoretical model, and form a comprehensive prediction model by combining the characteristics of the static stage and the moving stage; S4. Based on the dynamic target, the experimental platform is selected, the experimental conditions are set and subjects are recruited to perform the task, the task completion time data is collected, the hypothesis is verified, and finally the prediction model is fitted and the prediction result of the task completion time is obtained, which provides a basis for the design optimization of the target parameters in the dynamic interaction system design.

2. The dynamic target selection experimental platform according to claim 1, characterized in that: (2-1) Build a dynamic target selection experimental platform and set up experimental scenarios to simulate users' interactive tasks in a dynamic target environment; The experimental platform includes a dynamic target generation module, a data acquisition module and an input control module, which are used to provide target generation, data recording and user operation input functions required for the experiment; (2-2) The dynamic target generation module is used to generate the static targets and moving targets required for the experiment, and set the initial spacing, target width and moving target speed of the targets according to the experimental requirements; The initial spacing A is the horizontal distance between the stationary target and the edge of the moving target, which is used to control the moving distance requirement of the cursor; the target width W is the effective area size of the dynamic target, which is used to affect the difficulty of target selection; the target speed V is the speed of the dynamic target moving at a uniform speed in the experiment, which is used to simulate the moving characteristics of the target in a dynamic scene; (2-3) In the experimental task, the user first clicks the stationary target with the mouse to trigger the moving target to start moving in the specified direction at a preset speed; The moving target is the effective target of the task, and the user needs to track and click the effective area of ​​the moving target in real time to complete the task interaction operation; The task completion time T is defined as the total time from when the user clicks the stationary target to when the user successfully clicks the moving target, which is an important measurement indicator of the experimental data; (2-4) The data acquisition module records key data during the user operation in real time, including initial spacing, target width, target speed, and task completion time; (2-5) The input control module is used to receive the user's mouse operation signal to ensure the synchronous operation of the target generation and data acquisition modules to ensure the accuracy and integrity of the experimental data.

3. The key influencing factors and assumptions for dynamic target selection time prediction according to claim 1 are characterized by: (3-1) Based on the characteristics of the dynamic target selection task, it is assumed that the initial target spacing, target width and target speed, as well as the interaction between them, have an impact on the task completion time; (3-2) Based on the above influence relationships, the key assumptions for dynamic target selection time prediction are proposed to clarify the specific effects of the initial target spacing, target width, target speed and their interactions on task completion time.

4. The dynamic target selection time prediction model according to claim 1, characterized in that: (4-1) Based on the assumptions in S2, the selection phase can be divided into the target static phase and the movement phase; In the static stage, the task completion time is affected by the initial target spacing and target width. Based on the existing theoretical model of static target selection, a time prediction formula for the static stage is proposed. In the motion phase, the task completion time is related to the target's moving speed and target width. Based on the characteristics of dynamic targets, a time prediction formula for the motion phase is proposed. (4-2) In order to simultaneously consider the impact of stationary targets and moving targets on time prediction, a comprehensive time prediction model is proposed. By introducing the trade-off factor u, the model is used to adjust the relative effects of the stationary phase and the moving phase, uniformly express the time prediction relationship of the dynamic target selection task, and obtain a comprehensive time prediction model.

5. The experimental data collection and model hypothesis verification according to claim 1, characterized in that: (5-1) A certain number of subjects are recruited for the experiment. The subjects will perform tasks on the experimental platform according to pre-set conditions. The start of each task is triggered by the user clicking on a stationary target, and then the target starts to move at a set speed and direction. The user needs to track and click on the effective area of ​​the target in real time to complete the task. The system records the completion time and related data of each task in real time. Preferably, the subject has normal vision or corrected to normal vision, possesses good operational skills, and can accurately complete the target selection task. (5-2) The collected data were strictly preprocessed; outliers were cleaned to ensure data quality, extreme values ​​(i.e. values ​​that deviated from the mean by more than three standard deviations) were eliminated, and then repeated measures ANOVA was performed on the data at a confidence level of α = 0.05; Preferably, when the data violate the sphericity assumption, the data are subjected to Greenhouse-Geisser correction during analysis; (5-3) Conducting a between-subjects effect test to analyze the differences in task completion time between target initial spacing, target width, and target speed, and to evaluate the impact of these different factors on the experimental results; (5-4) Conduct an interaction effect test to analyze the comprehensive impact of the interaction between the initial target spacing, target width and target speed on the task completion time, and observe the joint effect of each variable on the task completion time; (5-5) Verify the correctness of the hypothesis by testing and analyzing the data, and adjust the parameters of the prediction model based on the results to ensure that the model can accurately predict the task completion time; Based on the test results and experimental data, a dynamic target selection time prediction model was obtained by applying the fitting method.

Images