A practical training teaching method for digital intelligence engineers based on robot learning

Abstract

The present invention discloses a digital engineer project training teaching method based on robot learning, which relates to the field of teaching technology. The present invention adjusts the task difficulty and content in real time according to the student's personal performance and ability status through dynamic task allocation and personalized learning path optimization. The adaptive teaching mode enables each student to learn at a pace and difficulty that suits him or her. It integrates project management tools, version control systems and real-time feedback mechanisms, so that students can obtain instant guidance and optimization suggestions in actual operations, provide real-time operation suggestions when submitting codes and merging versions, and automatically generate version snapshots and associate them with task progress, so that students can continuously optimize their operations in practice. It automatically generates learning reports, evaluates students' code quality, task completion efficiency and version control rationality, provides students with comprehensive learning feedback, and provides data support for teachers to improve teaching strategies.

Description

Technical Field

[0001] The present invention relates to the field of teaching technology, and in particular to a method for teaching digital intelligence engineers to practice their projects based on robot learning. Background Art

[0002] In the current practical training teaching of digital engineer projects, the teaching platform only provides basic task modules and lacks complete project management functions, such as task allocation, progress tracking, resource management and task priority control. Students and teams need to use manual methods to manage projects during the project process. This method is inefficient and prone to errors, and the project progress is difficult to control. Some platforms have introduced version control tools such as Git, but the integration is not high, and there is a lack of guidance and application support for students' actual needs. Students often encounter code conflicts, incorrect submissions, and version confusion during project development. They cannot effectively manage the iteration and change history of the project, which affects the quality of students' code. In addition, the feedback in the existing teaching platforms is mostly preset static content, such as error prompts and task descriptions, and cannot provide personalized suggestions based on students' real-time operation dynamics. When students encounter problems, they still need to find solutions through external materials such as csdn forums and Zhihu. They cannot obtain timely and targeted guidance, which greatly reduces students' learning efficiency and experience.

[0003] In terms of project management, traditional solutions use independent project management software such as Trello and Jira instead of integrating them into the teaching platform. This separate tool chain increases the difficulty for students to learn and use, and cannot achieve automated project status updates and integrated feedback, affecting the efficiency of project management. In terms of platform feedback, to make up for the lack of targeted teaching, teachers mainly manually correct homework and provide face-to-face Q&A feedback. However, this method cannot guarantee the timeliness and personalization of feedback in large-scale teaching.

[0004] It can be seen that in the current digital engineer project training teaching method, there is still a problem of separation between version control and project management tools in project management. Students need to frequently switch between multiple software, which increases the complexity of operation and is prone to errors. There is a lack of guided version management and project organization. Students lack a clear understanding of how to reasonably plan projects and control versions. The feedback from the teaching platform lags behind the students' operations and cannot correct students' wrong behaviors in time. The feedback content is mostly general information, and it is impossible to provide customized guidance for students' actual problems and learning paths, and they cannot get effective help. Therefore, there is an urgent need for a digital engineer project training teaching method based on robot learning to solve such problems. Summary of the Invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a digital engineer project training teaching method based on robot learning to solve the problem that the current digital engineer project training teaching method still has significant deficiencies in integration, personalized guidance and real-time feedback, which directly affects students' learning effects and skill mastery in complex projects.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] The present invention provides a method for teaching digital engineer project training based on robot learning, which includes:

[0009] Step S1, project management and version control embedded,

[0010] The integrated project management feedback system provides a unified task panel display, including task progress and version control status. When students submit code and merge versions, the project management feedback system will pop up feedback suggestions in real time, and each code submission will automatically generate a version snapshot, which is associated with the task progress.

[0011] Step S2: Dynamic task allocation and intelligent workflow management.

[0012] Dynamically assign tasks based on students' project execution, track task progress, associate version snapshots, analyze code changes when code is submitted, detect and prompt conflict risks,

[0013] Step S3, personalized feedback,

[0014] Combined with students' debugging logs and project execution status, repeated errors in students' debugging logs are automatically marked and learning suggestions are output, and subsequent tasks are adjusted according to the learning suggestions;

[0015] Step S4, project summary,

[0016] A learning report is generated based on the students' conflict risks, debugging logs, and project execution status throughout the project. The learning report includes code quality assessment, version control rationality analysis, and task completion efficiency.

[0017] Furthermore, the project management feedback system is integrated in step S1 as follows:

[0018] Assume that each task T i With task status S i (t), where t is time, and the task status update is expressed as:

[0019] S i (t+1)=S i (t)+ΔS i , where S i (t) represents task Ti The current state at time t, ΔS i It is the status change of the task, reflecting the progress of the current task. Every time the code is submitted, the system generates a version snapshot V j , and associate it with the current task progress, task progress P i The update formula is:

[0020] Among them, P i Represents task T i The current progress of V j Indicates the status of the version snapshot generated by the j-th code submission, W j is the weight, reflecting the version snapshot V j The degree of impact on task progress;

[0021] When the code is submitted and merged, the system will i (t) Generate feedback suggestions A i :

[0022] A i =f(S i ,C i ,R i ), where A i For task T i Feedback suggestions, S i is the current task status, C i Calculate the impact of the change, R, for the complexity of the current code change. i The task panel integrates the current status, version control status and feedback suggestions for the risk assessment value of the current code submission.

[0023] Generate visual display data Among them, D is the data set displayed on the panel, αi,β i ,γ i is a weight parameter that adjusts the display weight of the current task status, current progress, and feedback suggestions;

[0024] The project management feedback system updates task status in real time, generates feedback suggestions, and dynamically displays them on the task panel.

[0025] Furthermore, in step S1, the real-time feedback method during code submission and version merging is as follows:

[0026] When the code is submitted, the system updates the current task status. Task status S i (t) The update formula is:

[0027] S i (t+1)=S i (t)+ΔC i+ΔM i , where S i (t) represents task T i The state at time t, ΔC i Indicates the incremental impact of code changes on task status, ΔM i Indicates the incremental impact of the merge operation on the task status;

[0028] After each code submission, the system automatically generates a version snapshot V j , and is associated with the task progress P i , the update formula of task progress is:

[0029] Among them, P i For the current task T i Progress, V j is the impact weight of the j-th version snapshot, T j The time interval weight of submission is used. When the code is merged, the system detects the version conflict risk and generates feedback suggestions in real time. i , calculate feedback suggestions:

[0030] Among them, A i For task T i Feedback suggestions, R i is the risk factor of the current task, which evaluates the possibility of conflict. k Score the error of the kth code change to reflect the severity of the error, W k is the weight of the k-th change,

[0031] The system associates each version snapshot with the corresponding task and generates a version association matrix M:

[0032] M ij =α·V j +β·P i +γ·A i , where M ij For task T i With version snapshot V j , α, β, and γ are adjustment parameters that control the weights of version snapshots, the progress of the current task, and feedback suggestions in task association. They provide intelligent feedback on code submissions and version merges, and dynamically update task progress and version associations.

[0033] Furthermore, in step S2, dynamic task allocation, task progress tracking, version snapshot association, and conflict detection are performed. Specifically:

[0034] When dynamically assigning task content, the task fitness A is calculated based on the student's current task execution status and historical performance. i :

[0035] Among them, A i For task T i The fitness score, Q j S is the student's performance score in task j, j is the complexity coefficient of the jth task, reflecting the difficulty of the task, and is calculated based on the fitness score A i ,Dynamically adjust the allocation of tasks, and the allocation strategy is determined by the fitness score;

[0036] Task Progress P i Updates are made based on the number of student submissions and the quality of the work completed:

[0037] Where: P i For task T i The current progress of C k is the impact of the code change submitted for the kth time, Q k Score the quality of the kth submission. After each code submission, generate a version snapshot V j and compare it with the task progress P i Related: Where: R ij For task T i With version snapshot V j The degree of correlation, V j is the weight value of the version snapshot, D i is the delay coefficient of the current task, reflecting the task completion time deviation.

[0038] Furthermore, in step S2, code change analysis and conflict detection are performed:

[0039] When the code is submitted, the code change ΔC is analyzed to assess the conflict risk Among them, R c is the conflict risk value of the current submission, E l Score the severity of the conflict in the first change, F l is the impact factor of the change, reflecting the impact of the change on the overall structure of the code; if R c If the set threshold is exceeded, the system will prompt the conflict risk and provide merge suggestions.

[0040] This enables dynamic adjustment of task allocation, tracking of task progress, association of version snapshots, and real-time analysis of conflict risks in code changes.

[0041] Furthermore, in step S3, repeated errors are automatically marked based on the student debugging log and project execution status:

[0042] Extract the error information from the student debugging log and count the frequency of each error type, error frequency F e for: Among them, F e is the total frequency of error r, E i Score the severity of the error that occurs for the i-th time, I i is the weight of the i-th error. Based on the error log data, the pattern recognition algorithm is used to identify repeated errors and calculate the repetition rate R e : Among them, R e is the repetition rate of error e, T e The number of times the error is allowed to occur is used to compare the actual frequency of occurrence;

[0043] When R e If the error exceeds the set threshold, it will be automatically marked as a high-frequency error.

[0044] Assess the impact of each error on project execution and generate an impact factor I e : Among them, I e is the average impact factor of error e on the project, L j Score the items affected by the jth error, S j A rating of the project status when the error occurred.

[0045] Furthermore, in step S3, the method of generating learning suggestions is:

[0046] Combine the severity, repetition rate and impact factor of the error to generate learning suggestions G e , the calculation formula is:

[0047] Among them, G e C is the recommended weight for learning error e. e is the error category coefficient, and K is the smoothing factor, which prevents excessive emphasis on a certain type of error, thereby effectively analyzing repeated errors in the debug log and providing personalized learning suggestions based on the project execution status to help students improve their skills in a targeted manner.

[0048] Furthermore, in step S3, the content of subsequent tasks is adjusted according to the learning suggestions as follows:

[0049] Will learn suggestion G e Adjustment factor A e Expressed as:

[0050] Among them: A e Suggestions for learning G e The adjustment factor is used to adjust the intensity of the task content, R eis the error repetition rate, which indicates the frequency of errors. e is the error impact factor, which evaluates the overall impact of the error on the project, W e To learn the recommended priority weights, C e is the smoothing factor;

[0051] Comprehensively analyze the debugging log performance, learning speed and task completion quality to evaluate the student's current ability status S c , the calculation formula is: Among them, S c Score the student's current ability status to reflect the overall learning effect. k Score the completion quality of the kth task, L k is the learning intensity of the kth task, indicating the difficulty and complexity of the task, T k is the task completion time coefficient, reflecting the efficiency of task completion,

[0052] Combined adjustment factor A e and student ability status S c , calculate the subsequent task adjustment weight W t :

[0053] Among them, W t is the adjustment weight of subsequent tasks, S c Score the student's current ability status, A e The suggested adjustment factor for each study, D t is the task difficulty adjustment factor;

[0054] According to the adjustment weight W t , optimize the distribution of subsequent task content, and adjust the strategy optimization formula to:

[0055] T adj =T base ×(1+λ·W t ), where T adj For the adjusted task content, T base is the basic task content and the standard task before adjustment. λ is the adjustment coefficient, which controls the sensitivity of the adjustment range, thereby dynamically adjusting the subsequent task content to make it more in line with the students' current learning status and recommended requirements, thereby improving students' learning effects and project execution capabilities.

[0056] Furthermore, in step S4, a learning report is generated based on the student's conflict risk, debugging log, and project execution status as follows:

[0057] Calculate the conflict risk score R c Assess students' conflict risk in their projects. This score reflects potential problems when merging code. The calculation formula is:

[0058] Among them, R c To evaluate the conflict possibility during student code submission, C i Score the severity of the i-th conflict detection, F i is the conflict frequency, which indicates the number of times the conflict problem occurs repeatedly, W i is the weight of the conflict, reflecting the impact of the conflict on the overall project;

[0059] Extract and evaluate error patterns in debug logs to generate an error complexity score E c , which measures the complexity of the problems encountered by students during debugging:

[0060] Among them, E c Score the error complexity, E j is the severity of the jth error, S j is the duration of the error in the project, D j is the detection frequency of this error.

[0061] Furthermore, the method of generating the learning report in step S4 also includes:

[0062] Carry out multi-dimensional data integration of project execution to generate execution quality score Q p ,This score reflects the comprehensive performance of students in task execution, efficiency and code quality. The calculation formula is,

[0063] Among them, Q p Score the project execution quality, T k Score the completion of the kth task, Q k is the code quality assessment score, P k The priority of the task;

[0064] The conflict risk score, error complexity score and execution quality score are combined to generate the core score S of the learning report r :

[0065] S r =α·R c +β·E c +γ·Q p , where S r Score the comprehensive learning report, α, β, γ are weight coefficients, according to S r The learning report is automatically generated based on the score.

[0066] The beneficial effects of the present invention are:

[0067] The present invention adjusts the difficulty and content of tasks in real time according to students' individual performance and ability status through dynamic task allocation and personalized learning path optimization. The adaptive teaching model enables each student to learn at a pace and difficulty that suits him or her.

[0068] The present invention integrates project management tools, version control systems and real-time feedback mechanisms. Students can obtain instant guidance and optimization suggestions in actual operations, provide feedback suggestions when submitting codes and merging versions, and automatically generate version snapshots associated with task progress, so that students can continuously optimize their operations in practice.

[0069] The present invention automatically generates learning reports, evaluates students' code quality, task completion efficiency, and version control rationality, provides students with comprehensive learning feedback, and provides data support for teachers to improve teaching strategies.

[0070] The present invention provides real-time guidance and suggestions during the student operation process, simulating the instant interaction between teachers and students. When students encounter problems during the project, they can get immediate feedback, making up for the shortcomings of face-to-face interaction between teachers and students and improving learning enthusiasm and participation.

[0071] The present invention provides students with path suggestions for continuous learning improvement through personalized learning reports generated by data analysis. Students can adjust their learning strategies according to the targeted improvement suggestions in the report, helping students to form a clear learning direction. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0073] Figure 1 This is a flow chart of the digital engineer project training teaching method of the present invention. DETAILED DESCRIPTION

[0074] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0075] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0076] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0077] Example 1, with reference to Figure 1 This embodiment provides a method for teaching a digital engineer project training based on robot learning, including the following steps:

[0078] Step S1, project management and version control embedded,

[0079] Integrate the project management feedback system to display a unified task panel, including task progress and version control status.

[0080] When students submit codes and merge versions, the project management feedback system will pop up feedback suggestions in real time, and each code submission will automatically generate a version snapshot, which is associated with the task progress.

[0081] The integrated project management feedback system approach is:

[0082] Assume that each task T i With task status S i (t), where t is time, and the task status update is expressed as:

[0083] S i (t+1)=S i (t)+ΔS i , where S i (t) represents task T i The current state at time t, ΔS i It is the status change of the task, reflecting the progress of the current task. Every time the code is submitted, the system generates a version snapshot V j , and associate it with the current task progress, task progress P i The update formula is:

[0084] Among them, P i Represents task T i The current progress of V j Indicates the status of the version snapshot generated by the j-th code submission, W j is the weight, reflecting the version snapshot V j The degree of impact on task progress;

[0085] When the code is submitted and merged, the system will i (t) Generate feedback suggestions A i :

[0086] A i =f(S i ,C i ,R i ), where A i For task T i Feedback suggestions, S i is the current task status, C i Calculate the impact of the change, R, for the complexity of the current code change. i The task panel integrates the current status, version control status and feedback suggestions for the risk assessment value of the current code submission.

[0087] Generate visual display data Among them, D is the data set displayed in the panel, α i ,β i ,γ i is a weight parameter that adjusts the display weight of the current task status, current progress, and feedback suggestions;

[0088] The project management feedback system updates task status in real time, generates feedback suggestions, and displays them dynamically on the task panel;

[0089] The real-time feedback method during code submission and version merging is:

[0090] When the code is submitted, the system updates the current task status. Task status S i (t) The update formula is:

[0091] S i (t+1)=S i (t)+ΔC i +ΔM i , where S i (t) represents task T i The state at time t, ΔC i Indicates the incremental impact of code changes on task status, ΔM i Indicates the incremental impact of the merge operation on the task status;

[0092] After each code submission, the system automatically generates a version snapshot V j , and is associated with the task progress P i , the update formula of task progress is:

[0093] Among them, P i For the current task T i Progress, V j is the impact weight of the j-th version snapshot, T jThe time interval weight of submission is used. When the code is merged, the system detects the version conflict risk and generates feedback suggestions in real time. i , calculate feedback suggestions:

[0094] Among them, A i For task T i Feedback suggestions, R i is the risk factor of the current task, which evaluates the possibility of conflict. k Score the error of the kth code change to reflect the severity of the error, W k is the weight of the k-th change,

[0095] The system associates each version snapshot with the corresponding task and generates a version association matrix M:

[0096] M ij =α·V j +β·P i +γ·A i , where M ij For task T i With version snapshot V j The association value of α, β, and γ are adjustment parameters that control the weight of version snapshots, the progress of the current task, and feedback suggestions in task association. It provides intelligent feedback on code submissions and version merges, and dynamically updates task progress and version associations.

[0097] Specifically, the execution of student projects is analyzed in real time, task content is dynamically assigned, and task difficulty and intensity are adjusted to make the tasks more compatible with students' current abilities and learning status. Dynamic adjustments are based on the evaluation of students' historical performance, and personalized task allocation strategies are implemented by calculating the correlation between task fitness and task progress.

[0098] Step S2: Dynamic task allocation and intelligent workflow management.

[0099] Dynamically assign task content based on student project execution, track task progress, and associate version snapshots.

[0100] When the code is submitted, the code changes are analyzed, and conflict risks are detected and prompted.

[0101] Perform dynamic task allocation, task progress tracking, version snapshot association, and conflict detection. Specifically:

[0102] When dynamically assigning task content, the task fitness A is calculated based on the student's current task execution status and historical performance. i :

[0103] Among them, A iFor task T i The fitness score, Q j S is the student's performance score in task j, j is the complexity coefficient of the jth task, reflecting the difficulty of the task, and is calculated based on the fitness score A i ,Dynamically adjust the allocation of tasks, and the allocation strategy is determined by the fitness score;

[0104] Task Progress P i Updates are made based on the number of student submissions and the quality of the work completed:

[0105] Where: P i For task T i The current progress of C k is the impact of the code change submitted for the kth time, Q k Score the quality of the kth submission. After each code submission, generate a version snapshot V j and compare it with the task progress P i Related: Where: R ij For task T i With version snapshot V j The degree of correlation, V j is the weight value of the version snapshot, D i is the delay coefficient of the current task, reflecting the task completion time deviation,

[0106] Perform code change analysis and conflict detection:

[0107] When the code is submitted, the code change ΔC is analyzed to assess the conflict risk Among them, R c is the conflict risk value of the current submission, E l Score the severity of the conflict in the first change, F l is the impact factor of the change, reflecting the impact of the change on the overall structure of the code; if R c When the set threshold is exceeded, conflict risks are prompted and merge suggestions are provided, thereby dynamically adjusting task allocation, tracking task progress, linking version snapshots, and analyzing conflict risks in code changes in real time;

[0108] Specifically, in the project management process, an intelligent feedback system is integrated, and students can get feedback suggestions when submitting and merging code. Each submission will automatically generate a version snapshot and associate it with the task progress. Students can get instant guidance at each key node, which improves their understanding of the task and reduces the probability of task errors and version conflicts.

[0109] Step S3, personalized feedback,

[0110] Combined with the student debugging log and project execution status, the repeated errors in the student debugging log are automatically marked and learning suggestions are output.

[0111] Adjust the content of subsequent tasks based on learning suggestions;

[0112] Combine student debugging logs and project execution to automatically mark repeated errors:

[0113] Extract the error information from the student debugging log and count the frequency of each error type, error frequency F e for: Among them, F e is the total frequency of error e, E i For the i The severity score of the error that occurs, I i is the weight of the i-th error. Based on the error log data, the pattern recognition algorithm is used to identify repeated errors and calculate the repetition rate R e : Among them, R e is the repetition rate of error e, T e The number of times the error is allowed to occur is used to compare the actual frequency of occurrence;

[0114] When R e If the error exceeds the set threshold, it will be automatically marked as a high-frequency error.

[0115] Assess the impact of each error on project execution and generate an impact factor I e : Among them, I e is the average impact factor of error e on the project, L j Score the items affected by the jth error, S j Rating the status of the project when the error occurred;

[0116] Specifically, by analyzing the frequency, influencing factors and severity of errors, we provide students with precise improvement paths, help them focus on solving their own weaknesses, and improve their learning efficiency. In-depth analysis of repetitive errors helps teachers identify common learning difficulties and provide data support for overall teaching optimization.

[0117] The way to generate learning suggestions is:

[0118] Combine the severity, repetition rate and impact factor of the error to generate learning suggestions G e , the calculation formula is:

[0119] Among them, G e C is the recommended weight for learning error e. eis the error category coefficient, and K is the smoothing factor, which prevents over-emphasis on a certain type of error. This effectively analyzes repeated errors in debug logs and provides personalized learning suggestions based on project execution, helping students improve their skills in a targeted manner.

[0120] Adjust the content of subsequent tasks according to the learning suggestions as follows:

[0121] Will learn suggestion G e Adjustment factor A e Expressed as:

[0122] Among them: A e Suggestions for learning G e The adjustment factor is used to adjust the intensity of the task content, G e is the error repetition rate, which indicates the frequency of errors. e is the error impact factor, which evaluates the overall impact of the error on the project, W e To learn the recommended priority weights, C e is the smoothing factor;

[0123] Comprehensively analyze the debugging log performance, learning speed and task completion quality to evaluate the student's current ability status S c , the calculation formula is: Among them, S c Score the student's current ability status to reflect the overall learning effect. k Score the completion quality of the kth task, L k is the learning intensity of the kth task, indicating the difficulty and complexity of the task, T k is the task completion time coefficient, reflecting the efficiency of task completion,

[0124] Combined adjustment factor A e and student ability status S c , calculate the subsequent task adjustment weight W t :

[0125] Among them, W t is the adjustment weight of subsequent tasks, S c Score the student's current ability status, A e The suggested adjustment factor for each study, D t is the task difficulty adjustment factor;

[0126] According to the adjustment weight W t , optimize the distribution of subsequent task content, and adjust the strategy optimization formula to:

[0127] T adj =T base ×(1+λ·Wt ), where T adj For the adjusted task content, T base is the basic task content and the standard task before adjustment. λ is the adjustment coefficient, which controls the sensitivity of the adjustment range, thereby dynamically adjusting the content of subsequent tasks to make it more consistent with the students' current learning status and recommended requirements, thereby improving students' learning effects and project execution capabilities.

[0128] Step S4, project summary,

[0129] Generate a learning report based on the student's conflict risks, debugging logs, and project execution status throughout the project. The learning report includes code quality assessment, version control rationality analysis, and task completion efficiency.

[0130] The method of generating learning reports based on students' conflict risks, debugging logs and project execution is as follows:

[0131] Calculate the conflict risk score R c Assess students' conflict risk in their projects. This score reflects potential problems when merging code. The calculation formula is:

[0132] Among them, R c To evaluate the conflict possibility during student code submission, C i Score the severity of the i-th conflict detection, F i is the conflict frequency, which indicates the number of times the conflict problem occurs repeatedly, W i is the weight of the conflict, reflecting the impact of the conflict on the overall project;

[0133] Extract and evaluate error patterns in debug logs to generate an error complexity score E c , which measures the complexity of the problems encountered by students during debugging:

[0134] Among them, E c Score the error complexity, E j is the severity of the jth error, S j is the duration of the error in the project, D j is the detection frequency of the error;

[0135] Carry out multi-dimensional data integration of project execution to generate execution quality score Q p ,This score reflects the comprehensive performance of students in task execution, efficiency and code quality. The calculation formula is,

[0136] Among them, Q p Score the project execution quality, T kScore the completion of the kth task, Q k is the code quality assessment score, P k The priority of the task;

[0137] The conflict risk score, error complexity score and execution quality score are combined to generate the core score S of the learning report r :

[0138] S r =α·R c +β·E c +γ·Q p , where S r Score the comprehensive learning report, α, β, γ are weight coefficients, according to S r Scoring, automatically generate learning reports;

[0139] Specifically, the learning report integrates students' multi-dimensional performance in the project, including conflict risk score, error complexity analysis and project execution quality, to provide students with clear improvement suggestions.

[0140] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for practical training of digital intelligence engineers based on robot learning, characterized by: include, Step S1: Project management and version control are embedded, and the project management feedback system is integrated to display a unified task panel. The display content includes task progress and version control status. When students submit code and merge versions, the project management feedback system pops up feedback suggestions in real time, and each code submission automatically generates a version snapshot, which is associated with the task progress. The method of integrating the project management feedback system in step S1 is: Assume that each task T i With task status S i (t), where t is time, and the task status update is expressed as: S i (t+1)=S i (t)+ΔS i , where S i (t) represents task T i The current state at time t, ΔS i It is the status change of the task, reflecting the progress of the current task. Every time the code is submitted, the system generates a version snapshot V j , and associate it with the current task progress, task progress P i The update formula is: Among them, P i Represents task T i The current progress of V j Indicates the status of the version snapshot generated by the j-th code submission, W j is the weight, reflecting the version snapshot V j The degree of impact on task progress; When the code is submitted and merged, the system will i (t) Generate feedback suggestions A i : A i =f(S i ,C i ,R i ), where A i For task T i Feedback suggestions, S i is the current task status, C i Calculate the impact of the change, R, for the complexity of the current code change. i The task panel integrates the current status, version control status and feedback suggestions for the risk assessment value of the current code submission. Generate visual display data Among them, D is the data set displayed in the panel, α i ,β i ,γ i is a weight parameter that adjusts the display weight of the current task status, current progress, and feedback suggestions; In step S1, the real-time feedback method during code submission and version merging is as follows: When the code is submitted, the system updates the current task status. Task status S i (t) The update formula is: S i (t+1)=S i (t)+ΔC i +ΔM i , where S i (t) represents task T i The state at time t, ΔC i Indicates the incremental impact of code changes on task status, ΔM i Indicates the incremental impact of the merge operation on the task status; After each code submission, the system automatically generates a version snapshot V j , and is associated with the task progress P i , the update formula of task progress is: Among them, P i For the current task T i Progress, V j is the impact weight of the j-th version snapshot, T j The time interval weight of submission is used. When the code is merged, the system detects the version conflict risk and generates feedback suggestions in real time. i , calculate feedback suggestions: Among them, A i For task T i Feedback suggestions, R i is the risk factor of the current task, which evaluates the possibility of conflict. k Score the error of the kth code change to reflect the severity of the error, W k is the weight of the k-th change, The system associates each version snapshot with the corresponding task and generates a version association matrix M: M ij =α·V j +β·P i +γ·A i , where M ij For task T i With version snapshot V j The association value of α, β, and γ are adjustment parameters that control the weights of version snapshots, the progress of the current task, and feedback suggestions in task association; Step S2: Dynamic task allocation and intelligent workflow management. Task content is dynamically assigned based on the student's project execution status, task progress is tracked, version snapshots are associated, and when code is submitted, code changes are analyzed to detect and prompt conflict risks. In step S2, dynamic task allocation, task progress tracking, version snapshot association, and conflict detection are performed. Specifically: When dynamically assigning task content, the task fitness F is calculated based on the student’s current task execution status and historical performance. i : Among them, F i For task T i The fitness score, Q j S is the student's performance score in task j, j is the complexity coefficient of the j-th task, reflecting the difficulty of the task; Task Progress P i Updates are made based on the number of student submissions and the quality of the work completed: Where: P i For task T i The current progress of C k is the impact of the code change submitted at k, Q k Score the quality of the kth submission. After each code submission, generate a version snapshot V j and compare it with the task progress P i Related: Where: R ij For task T i With version snapshot V j The degree of correlation, V j is the weight value of the version snapshot, D i is the delay coefficient of the current task, reflecting the deviation of task completion time; Step S3: Personalized feedback. Combining the student's debugging log and project execution status, the system automatically marks the repeated errors in the student's debugging log and outputs learning suggestions. The subsequent task content is adjusted according to the learning suggestions. Step S4, project summary, generates a learning report based on the student's conflict risks, debugging logs, and project execution status throughout the project. The learning report includes code quality assessment, version control rationality analysis, and task completion efficiency.

2. A method for teaching digital engineer project training based on robot learning according to claim 1, characterized in that: In step S2, code change analysis and conflict detection are performed: When the code is submitted, the code change ΔC is analyzed to assess the conflict risk Among them, R c is the conflict risk value of the current submission, E l Score the severity of the conflict in the first change, F l is the impact factor of the change, reflecting the impact of the change on the overall structure of the code; if R c If the set threshold is exceeded, the system will prompt the conflict risk and provide merge suggestions.

3. The method for teaching digital engineer project training based on robot learning according to claim 2 is characterized in that: In step S3, repeated errors are automatically marked based on the student debugging log and project execution status: Extract the error information from the student debugging log and count the frequency of each error type, error frequency F e for: Among them, F e is the total frequency of error e, E i Score the severity of the error that occurs for the i-th time, I i is the weight of the i-th error. Based on the error log data, the pattern recognition algorithm is used to identify repeated errors and calculate the repetition rate R e : Among them, R e is the repetition rate of error e, T e The number of times the error is allowed to occur is used to compare the actual frequency of occurrence; When R e If the error exceeds the set threshold, it will be automatically marked as a high-frequency error. Evaluate the impact of each error on project execution and generate an impact factor I e : Among them, I e is the average impact factor of error e on the project, L j Score the items affected by the jth error, S j A rating of the project status when the error occurred.

4. The method for teaching digital engineer project training based on robot learning according to claim 3 is characterized in that: In step S3, the method of generating learning suggestions is as follows: Combine the severity, repetition rate and impact factor of the error to generate learning suggestions G e , the calculation formula is: Among them, G e C is the recommended weight for learning error e. e is the error category coefficient, and K is the smoothing factor to prevent excessive emphasis on a certain type of error.

5. The method for teaching digital engineer project training based on robot learning according to claim 4 is characterized in that: In step S3, the content of subsequent tasks is adjusted according to the learning suggestions as follows: Will learn suggestion G e Adjustment factor A e Expressed as: Among them: A e Suggestions for learning G e The adjustment factor is used to adjust the intensity of the task content, R e is the error repetition rate, which indicates the frequency of errors. e is the error impact factor, which evaluates the overall impact of the error on the project, W e To learn the recommended priority weights, C e is the smoothing factor; Comprehensively analyze the debugging log performance, learning speed and task completion quality to evaluate the student's current ability status S c , the calculation formula is: Among them, S c Score the student's current ability status to reflect the overall learning effect. k Score the completion quality of the kth task, L k is the learning intensity of the kth task, indicating the difficulty and complexity of the task, T k is the task completion time coefficient, reflecting the efficiency of task completion, Combined adjustment factor A e and student ability status S c , calculate the subsequent task adjustment weight W t : Among them, W t is the adjustment weight of subsequent tasks, S c Score the student's current ability status, A e The suggested adjustment factor for each study, D t is the task difficulty adjustment factor; According to the adjustment weight W t , optimize the distribution of subsequent task content, and adjust the strategy optimization formula to: T adj =T base ×(1+λ·W t ), where T adj For the adjusted task content, T base is the basic task content, the standard task before adjustment, and λ is the adjustment coefficient.

6. The method for teaching digital engineer project training based on robot learning according to claim 5 is characterized in that: In step S4, the method for generating a learning report based on the student's conflict risk, debugging log, and project execution status is as follows: Calculate the conflict risk score R c Assess students' conflict risk in their projects. This score reflects potential problems when merging code. The calculation formula is: Among them, R c To evaluate the conflict possibility during student code submission, C i Score the severity of the i-th conflict detection, F i is the conflict frequency, which indicates the number of times the conflict problem occurs repeatedly, W i is the weight of the conflict, reflecting the impact of the conflict on the overall project; Extract and evaluate error patterns in debug logs to generate an error complexity score E c , which measures the complexity of the problems encountered by students during debugging: Among them, E c Score the error complexity, E j is the severity of the jth error, S j is the duration of the error in the project, D j is the detection frequency of this error.

7. The method for teaching digital engineer project training based on robot learning according to claim 6 is characterized in that: The method of generating the learning report in step S4 also includes: Carry out multi-dimensional data integration of project execution to generate execution quality score Q p ,This score reflects the comprehensive performance of students in task execution, efficiency and code quality. The calculation formula is, Among them, Q p Score the project execution quality, T k Score the completion of the kth task, Q k is the code quality assessment score, P k The priority of the task; The conflict risk score, error complexity score and execution quality score are combined to generate the core score S of the learning report r : S r =α·R c +β·E c +γ·Q p , where S r Score the comprehensive learning report, α, β, γ are weight coefficients, according to S r The learning report is automatically generated based on the score.

Images