Personnel work assisting method and system based on AI assistant

Abstract

The invention provides a personnel work assistance method and system based on an AI assistant, and the method comprises the steps: carrying out the work assistance of a person based on the AI assistant when the person works; wherein the AI assistant is an artificial intelligence model trained by using a large amount of working experience related to the working portrait of the personnel; the method for assisting the work of the personnel based on the AI assistant comprises the following steps: acquiring a multi-modal work record of the personnel in real time; based on an AI assistant, generating first auxiliary content according to the multi-modal working record; pushing the first auxiliary content to the person; and / or receiving a work assistance demand input by the personnel; based on the AI assistant, according to the work assistance demand, generating a second assistance content; and pushing the second auxiliary content to the person. According to the invention, intelligent work flow support and personalized auxiliary functions are realized, the working efficiency and precision of personnel are improved to a great extent, and the personnel are helped to better meet increasingly complex working requirements.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a personnel work assistance method based on an AI assistant and a system thereof. Background Art

[0002] At present, some staff members have heavy and complex work tasks, covering case handling, intelligence analysis, patrol inspections, emergency response, etc. Each task requires staff to process a large amount of information in a short period of time and make accurate judgments and decisions, so work assistance is needed.

[0003] Traditional work assistance tools mainly focus on information recording, storage and query functions, and are usually just simple task management systems that lack intelligent workflow support and personalized assistance functions.

[0004] Therefore, there is an urgent need for a more intelligent and efficient solution that can improve the efficiency and accuracy of personnel's work and help staff better cope with increasingly complex work requirements. Summary of the invention

[0005] One of the purposes of the present invention is to provide a personnel work assistance method based on an AI assistant. When the AI ​​assistant is used to assist the staff, it is divided into two modes: active assistance and passive assistance. In active assistance, based on the AI ​​assistant, a first auxiliary content is generated and pushed according to the multimodal work records of the staff. In passive assistance, the staff inputs the work assistance requirements, and the AI ​​assistant generates and pushes the corresponding second auxiliary content, thereby realizing intelligent workflow support and personalized auxiliary functions. Specifically, the staff can be given emergency response plans or optimal disposal suggestions. Secondly, it can help the staff to analyze the case, predict and warn risks, greatly improve the efficiency and accuracy of the staff's work, and help the staff better cope with increasingly complex work requirements.

[0006] An embodiment of the present invention provides a personnel work assistance method based on an AI assistant, comprising:

[0007] When people are working, AI assistants are used to assist them. AI assistants are artificial intelligence models trained using a large amount of work experience related to people's work profiles.

[0008] The AI ​​assistant-based work assistance for personnel includes:

[0009] Obtain multimodal work records of personnel in real time;

[0010] Based on AI assistant, the first auxiliary content is generated according to multimodal work records;

[0011] Pushing first auxiliary content to personnel;

[0012] and / or,

[0013] Receives personnel input on job assistance needs;

[0014] Based on AI assistant, generate second auxiliary content according to work assistance needs;

[0015] Push secondary auxiliary content to the person.

[0016] Optionally, the generating of the second auxiliary content based on the AI ​​assistant according to the work assistance requirement includes:

[0017] Perform vector transformation on work assistance requirements to obtain the requirement content vector;

[0018] Calculate the similarity between the demand content vector and the document vector in the vector database;

[0019] Generate second auxiliary content based on the document content corresponding to the document vector with the greatest similarity;

[0020] The steps for obtaining the document vector in the vector database are as follows:

[0021] Receive multi-format documents uploaded by personnel; the formats of multi-format documents include at least: TXT, DOCX, PDF;

[0022] Perform vector conversion on the document content of multi-format documents to obtain document vectors;

[0023] The step of converting the document content of the multi-format document into a vector to obtain a document vector includes:

[0024] The document content of multi-format documents is separated into text and converted into document vectors using an embedding model.

[0025] Optionally, the embedding model includes at least: bge-large-zh-v1.5 Chinese embedding model.

[0026] Optionally, the vector database uses the PGVector vector library for vector storage.

[0027] Optionally, after generating the second auxiliary content based on the document content corresponding to the document vector with the greatest similarity, the generated second auxiliary content is further optimized through the chatGLM API.

[0028] Optionally, when the document content of the multi-format document is converted into a vector, key information in the document content is automatically identified and extracted according to the element direction specified by the user.

[0029] Optional, AI assistant-based personnel work assistance methods also include:

[0030] When personnel use the work simulation sandbox, whether to enter the standard assistance timing is determined based on the real-time simulated content of the work simulation performed by the personnel in the work simulation sandbox;

[0031] When entering, the auxiliary basis is extracted from the real-time simulated content based on the auxiliary basis extraction template of the standard auxiliary timing;

[0032] Based on the AI ​​assistant, determine the auxiliary simulation content according to the auxiliary basis;

[0033] Generate interactive control timelines for auxiliary simulation content;

[0034] Based on the auxiliary simulation content, the corresponding work simulation is carried out in the work simulation sandbox, and during the simulation process, the corresponding interactive control is carried out with the personnel based on the interactive control timeline.

[0035] Optionally, the generating of the interactive control timeline of the auxiliary simulation content includes:

[0036] Get the simulation progress timeline of auxiliary simulation content;

[0037] Based on the time interval division constraint, a plurality of first target time intervals are divided on the simulation progress time axis;

[0038] Traverse each first target time interval in sequence;

[0039] Each time the traversal is completed, multiple immersive perspectives are determined based on the action changes and position changes of multiple simulated subjects within the first target time interval traversed on the simulation progress timeline;

[0040] Generate immersion trigger constraints for each immersion perspective;

[0041] Determine a second target time interval corresponding to the traversed first target time interval from the blank time axis;

[0042] Setting each immersive perspective and its respective immersive trigger constraint within a second target time interval;

[0043] After traversing each first target time interval, a blank timeline in which all immersive perspectives and respective immersive trigger constraints are fully set is used as an interactive control timeline.

[0044] Optionally, the time interval division constraint includes:

[0045] Different first target time intervals each completely and independently include continuous simulation time periods of at least N simulation situations; wherein N is a positive integer greater than or equal to 2;

[0046] as well as,

[0047] There is at least one attribute association relationship between any two simulation situations in consecutive simulation time periods included in the same first target time interval.

[0048] Optionally, determining multiple immersive perspectives and respective immersive trigger constraints based on the action changes and position changes of multiple simulated subjects within the first target time interval traversed on the simulation progress timeline includes:

[0049] The key action changes of each simulated subject are screened respectively to obtain the key action changes of each simulated subject;

[0050] Time-align the key action changes and position changes of each simulated subject to obtain an aligned sequence;

[0051] Traverse multiple action change nodes in the alignment sequence in sequence order;

[0052] Each time when traversing, the simulation subject of the traversed action change node is taken as the first target, and the remaining simulation subjects are taken as the second target;

[0053] Determine the first position change node and the second position change node of the first target and the second target respectively from the aligned position change nodes in the aligned sequence from the traversed action change nodes;

[0054] In the working simulation sandbox, when the first target is at the first position change node, the simulation perspective when the second target is at the second position change node is determined to be visible to the traversed action change node; wherein the simulation perspective includes: first-person perspective and third-person perspective;

[0055] When it is visible, the simulated perspective when the second target is at the second position change node is used as the immersive perspective to be selected; otherwise, the third-person perspective when the first target is at the first position change node is used as the immersive perspective to be selected;

[0056] Associating the selected immersive perspective with the change time of the traversed action change node;

[0057] After traversing each action change node, all candidate immersive perspectives are integrated to obtain a perspective set;

[0058] Based on the set optimization constraints, the view set is optimized;

[0059] Based on the optimized perspective set, multiple immersive perspectives are determined.

[0060] Optionally, the set optimization constraints include:

[0061] The change times of the associations of different candidate immersive perspectives in the optimized perspective set are different from each other;

[0062] as well as,

[0063] The optimized perspective set has the least number of categories of simulated subjects belonging to different candidate immersive perspectives.

[0064] Optionally, generating the immersion trigger constraint for each immersion perspective includes:

[0065] The personnel adjusts the viewing angle of the work simulation sandbox and the time difference between the adjustment time and the change time associated with the immersive perspective does not exceed the time difference threshold;

[0066] and / or,

[0067] The visual content overlap between the adjusted viewing perspective of the user and the immersive perspective exceeds the overlap threshold.

[0068] An embodiment of the present invention provides an AI assistant-based personnel work assistance system, including:

[0069] The first auxiliary module is used to assist personnel with their work based on an AI assistant when they are working; wherein the AI ​​assistant is an artificial intelligence model trained using a large amount of work experience related to the work profile of the personnel;

[0070] The first auxiliary module provides work assistance to personnel based on an AI assistant, including:

[0071] Obtain multimodal work records of personnel in real time;

[0072] Based on AI assistant, the first auxiliary content is generated according to multimodal work records;

[0073] Pushing first auxiliary content to personnel;

[0074] and / or,

[0075] Receives personnel input on job assistance needs;

[0076] Based on AI assistant, generate second auxiliary content according to work assistance needs;

[0077] Push secondary auxiliary content to the person.

[0078] Optional, AI-based personnel work assistance system also includes:

[0079] The second auxiliary module is used to:

[0080] When personnel use the work simulation sandbox, whether to enter the standard assistance timing is determined based on the real-time simulated content of the work simulation performed by the personnel in the work simulation sandbox;

[0081] When entering, the auxiliary basis is extracted from the real-time simulated content based on the auxiliary basis extraction template of the standard auxiliary timing;

[0082] Based on the AI ​​assistant, determine the auxiliary simulation content according to the auxiliary basis;

[0083] Generate interactive control timelines for auxiliary simulation content;

[0084] Based on the auxiliary simulation content, the corresponding work simulation is carried out in the work simulation sandbox, and during the simulation process, the corresponding interactive control is carried out with the personnel based on the interactive control timeline.

[0085] Optionally, the second auxiliary module generates an interactive control timeline of the auxiliary simulation content, including:

[0086] Get the simulation progress timeline of auxiliary simulation content;

[0087] Based on the time interval division constraint, a plurality of first target time intervals are divided on the simulation progress time axis;

[0088] Traverse each first target time interval in sequence;

[0089] Each time the traversal is completed, multiple immersive perspectives are determined based on the action changes and position changes of multiple simulated subjects within the first target time interval traversed on the simulation progress timeline;

[0090] Generate immersion trigger constraints for each immersion perspective;

[0091] Determine a second target time interval corresponding to the traversed first target time interval from the blank time axis;

[0092] Setting each immersive perspective and its respective immersive trigger constraint within a second target time interval;

[0093] After traversing each first target time interval, a blank timeline in which all immersive perspectives and respective immersive trigger constraints are fully set is used as an interactive control timeline.

[0094] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0095] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0097] Figure 1 Schematic diagram of a personnel work assistance method based on an AI assistant in an embodiment of the present invention;

[0098] Figure 2 This is a schematic diagram of a personnel work assistance system based on an AI assistant in an embodiment of the present invention. DETAILED DESCRIPTION

[0099] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0100] The embodiment of the present invention provides a personnel work assistance method based on an AI assistant, such as Figure 1 As shown, including:

[0101] S1. When personnel are working, AI assistants are used to assist them in their work; AI assistants are artificial intelligence models trained using a large amount of work experience related to personnel's work profiles;

[0102] In S1, personnel can be staff members; the work profile of personnel includes: position, work experience, gender, etc.; work experience is the historical case handling process, typical case handling regulations, historical case records, etc. related to the work profile of personnel; a large amount of such work experience is used as training samples to train the artificial intelligence model, so that it can be combined with the training work experience to assist personnel in adaptation;

[0103] Among them, the S1, based on the AI ​​assistant, provides work assistance to personnel, including:

[0104] S111. Real-time acquisition of multimodal work records of personnel;

[0105] In S111, the AI ​​assistant obtains the multimodal work records of personnel in real time through the work recorder or sensor worn by the personnel. The record content includes: text input, voice communication, image upload and other forms of data. For example, when the staff is performing patrol tasks, the AI ​​assistant can obtain the voice communication records and on-site images of the patrol process in real time through voice recognition and image recognition technology;

[0106] S112. Based on the AI ​​assistant, generate first auxiliary content according to the multimodal work record;

[0107] In S112, the AI ​​assistant analyzes the work situation based on the multimodal work records and combines the work experience used for training to generate targeted auxiliary content. For example, if a person is working on a case and collecting information about the case, the AI ​​assistant can automatically provide case-related suggestions and warnings based on the information based on the existing case data and historical experience;

[0108] S113, pushing the first auxiliary content to the personnel;

[0109] In S113, the AI ​​assistant pushes the generated first auxiliary content to the personnel in real time so that they can make reference decisions. For example, in an emergency, the AI ​​assistant can push emergency response plans and early warning information predicting risks to help personnel respond quickly.

[0110] and / or,

[0111] S121, receiving work assistance requirements input by personnel;

[0112] In S121, when a person encounters difficulties at work, he or she can send specific work assistance requests to the AI ​​assistant. For example, when a person is handling a complex task, he or she may request the AI ​​assistant to provide information support or operation suggestions in a certain related field.

[0113] S122. Generate second auxiliary content based on the AI ​​assistant and according to work assistance needs;

[0114] In S122, the AI ​​assistant generates second assistance content based on the work assistance needs and the work experience used for training. For example, when a staff member is handling a case, he asks the AI ​​assistant how to analyze the communication records of relevant personnel. The AI ​​assistant generates targeted analysis suggestions based on historical case experience.

[0115] S123: Push the second auxiliary content to the personnel.

[0116] In S123, finally, the AI ​​assistant pushes the generated second auxiliary content to the personnel to help them complete their work more efficiently.

[0117] When this application uses AI assistants to assist staff, it is implemented in two ways: active assistance and passive assistance. During active assistance, based on the AI ​​assistant, the first auxiliary content is generated and pushed according to the multimodal work records of the staff. During passive assistance, the staff inputs the work assistance needs, and the AI ​​assistant generates and pushes the corresponding second auxiliary content, thereby realizing intelligent workflow support and personalized auxiliary functions. Specifically, it can provide staff with emergency response plans or optimal disposal suggestions. Secondly, it can help staff conduct case analysis, risk prediction and early warning, greatly improving the efficiency and accuracy of staff work, and helping staff better cope with increasingly complex work requirements.

[0118] In one embodiment, the personnel work assistance method based on the AI ​​assistant further includes:

[0119] S2. When a person uses a work simulation sandbox, based on the real-time simulated content of the work simulation performed by the person in the work simulation sandbox, determine whether to enter the standard assistance timing;

[0120] In S2, the work simulation sandbox is a virtual environment used to simulate a certain personnel workflow or task, which is specifically implemented in the form of a computer-simulated 3D environment; the real-time simulated content is the content that the personnel have completed the simulation in the work simulation sandbox in real time, such as: simulated character settings, simulated plot settings, simulated location settings, etc.; the standard assistance timing refers to the standard timing when the personnel need to perform work simulation assistance, for example: the simulated content indicates that the personnel have completed the settings of all simulated characters, simulated plots, simulated locations, etc., and are starting to prepare for the simulation, so work simulation assistance is needed. For example: the simulated content indicates that the personnel have paused the simulation for a long time and may not know how to continue the simulation, so work simulation assistance is needed. Therefore, whether to enter the standard assistance timing can be determined based on the real-time simulated content;

[0121] S3, when entering, extracting auxiliary evidence from the real-time simulated content based on the auxiliary evidence extraction template of the standard auxiliary opportunity;

[0122] In S3, the standard assistance timing for entry also reflects how to provide work simulation assistance to personnel. The auxiliary basis extraction template of the standard assistance timing is to extract auxiliary basis that can determine the auxiliary simulation content for work simulation assistance to personnel from the real-time simulated content. For example, if the standard assistance timing for entry is when the personnel have completed the settings of all simulated characters, simulated plots, simulated locations, etc. and start to prepare for simulation, then the auxiliary basis extraction template is to extract the setting parameters of all simulated characters, simulated plots, simulated locations, etc. from the real-time simulated content as auxiliary basis.

[0123] S4. Based on the AI ​​assistant, determine the auxiliary simulation content according to the auxiliary basis;

[0124] In S4, the AI ​​assistant can determine the auxiliary simulation content based on the auxiliary basis combined with the work experience used for training (focusing on the development process of historical cases). The auxiliary simulation content is the plot development that can be simulated in the work simulation sandbox;

[0125] S5, generating an interactive control timeline of auxiliary simulation content;

[0126] In S5, the interactive control timeline is used to assist the simulation content in instructing the system and personnel to perform corresponding interactive control during the corresponding work simulation in the work simulation sandbox;

[0127] S6. Based on the auxiliary simulation content, corresponding work simulation is carried out in the work simulation sandbox, and during the simulation process, corresponding interactive control is carried out with personnel based on the interactive control timeline.

[0128] The embodiments of the present invention achieve the following beneficial effects:

[0129] Based on the real-time simulated content of personnel in the work simulation sandbox, the system can accurately identify standard assistance opportunities to ensure that personnel can obtain necessary auxiliary support in a timely manner when conducting work simulation; after the system automatically enters the standard assistance opportunity, it provides personnel with personalized assistance basis, and uses the AI ​​assistant to combine historical cases and work experience to intelligently generate auxiliary simulation content; it not only improves the authenticity and practicality of work simulation, but also effectively guides personnel to maintain efficient decision-making and response capabilities in complex work simulation scenarios; during the simulation process of the work simulation sandbox, the interactive control timeline serves as an interactive control guide, making the interaction between the system and personnel smoother and more efficient; it not only improves the work simulation efficiency of personnel, but also improves the user experience.

[0130] In one embodiment, the step S5, generating an interactive control timeline of auxiliary simulation content, includes:

[0131] S51, obtaining a simulation progress timeline of the auxiliary simulation content;

[0132] In S51, the simulation progress timeline is a timeline for real-time control of the simulation progress of the auxiliary simulation content in the work simulation sandbox;

[0133] S52, dividing a plurality of first target time intervals on the simulation progress time axis based on the time interval division constraint;

[0134] S53, traversing each first target time interval in sequence;

[0135] In S53, the time sequence refers to the time sequence of the first target time interval on the simulation progress time axis;

[0136] S54, each time the traversal is completed, based on the action changes and position changes of multiple simulated subjects within the first target time interval traversed on the simulation progress timeline, determine multiple immersive perspectives;

[0137] In S54, the first target time interval has local simulation content simulated in the first target time interval, the local simulation content includes action changes and position changes of multiple simulation subjects, the simulation subject can be a simulated character set by a person, the action change is the sequence change of the simulation subject generating the simulation action, and the position change is the sequence change of the position movement of the simulation subject; when using the work simulation sandbox to perform work simulation, a suitable future work plan is mainly formulated by observing the simulation situation of the simulation subject, and based on the action changes and position changes of the multiple simulation subjects, multiple immersive perspectives can be determined, and each immersive perspective can help the person immersively observe the simulation situation of the simulation subject in the traversed first target time interval;

[0138] S55, generating an immersion trigger constraint for each immersion perspective;

[0139] In S55, the immersion trigger constraint is a constraint condition that triggers the need to use the immersion perspective to lead the personnel to immerse and view;

[0140] S56, determining a second target time interval corresponding to the traversed first target time interval from the blank time axis;

[0141] In S56, the blank time axis is a time axis without any content set, and the corresponding second target time interval has the same time start time as the traversed first target time interval, and is distinguished only because they are on different time axes;

[0142] S57, setting each immersion perspective and its respective immersion trigger constraint within a second target time interval;

[0143] In S57, when each immersive perspective and each immersive trigger constraint are set within the second target time interval, when interactive control is performed with a person based on the interactive control timeline, if the time enters the second target time interval, and if the user viewing the work simulation sandbox meets the immersive trigger constraint, the person is led to immersively view the work simulation sandbox through the immersive perspective;

[0144] S58, after traversing each first target time interval, a blank time axis in which all immersive perspectives and respective immersive trigger constraints are fully set is used as an interactive control time axis;

[0145] In S58, after all traversals are completed, that is, after all immersive perspectives and respective immersive trigger constraints are set, the blank timeline is used as the interactive control timeline;

[0146] The time interval division constraints include:

[0147] Constraint A1: Different first target time intervals are continuous simulation time periods that do not overlap with each other and completely contain at least N simulation situations; wherein N is a positive integer greater than or equal to 2;

[0148] as well as,

[0149] Constraint A2: There is at least one attribute association relationship between any two simulation situations in consecutive simulation time periods included in the same first target time interval.

[0150] In constraint A1, when performing work simulation using auxiliary simulation content, different simulation situations are simulated in different simulation time periods. Setting constraint A1 can make interactive control occur in a complete continuous simulation time period in which at least N simulation situations are simulated, thereby avoiding inappropriate interactive control time. It can also make the content of interactive control at different times different, thereby avoiding duplication of interactive control.

[0151] In constraint A2, the attribute association relationship includes: the locations where the simulation situations occur are the same, there is a causal relationship between the simulation situations, the simulation situations involve the same simulated characters, etc. Setting constraint A2 can make the interactive control targeted at the related simulation situations, making the interactive control more appropriate.

[0152] The embodiments of the present invention achieve the following beneficial effects:

[0153] The simulation progress is organically combined with the interactive timeline. Through the precise division of simulation time periods and the intelligent allocation of immersive perspectives, users can observe and analyze a variety of simulation scenarios in a time-oriented and targeted manner. Specifically, by dividing the simulation progress timeline into multiple first target time intervals, the action changes and position changes of the simulated subject are analyzed in each first time interval, and then multiple immersive perspectives are determined, so that users can not only track the evolution of the simulation subject in real time, but also flexibly enter the immersive perspective according to the immersion trigger constraints to obtain specific situational information. Secondly, through the guidance of the interactive control timeline, the auxiliary personnel can deeply participate in the simulation process, which improves the accuracy of the personnel's work simulation.

[0154] In addition, by constraining A1, the system can avoid duplication and discontinuity of interactive control, ensuring that each simulation time period can provide personnel with sufficient and effective interactive experience; by constraining A2, the internal logical rigor of the simulation situation is ensured, so that when making decisions and analyses, users can not only pay attention to the development of individual situations, but also grasp the relationship between various situations from a holistic level; the combination of constraints A1 and A2 ensures that the simulation environment is not only continuous in the time dimension, but also enriched in the content dimension, thereby further enhancing the suitability of interactive control and the user's sense of immersion.

[0155] In one embodiment, the step S54, determining multiple immersive perspectives and respective immersive trigger constraints based on the action changes and position changes of multiple simulated subjects within the first target time interval traversed on the simulation progress timeline, includes:

[0156] S5401, performing key action screening on the action changes of each simulated subject respectively to obtain the key action changes of each simulated subject;

[0157] In S5401, in the key action screening, key action changes that are valuable to the work simulation of personnel are screened out from the action changes, that is, key action changes; the value may be that viewing the key action can enable personnel to further formulate future work plans, or viewing the key action can enable personnel to quickly understand the key points of the simulation situation, for example, the key action is the suspect simulating escape, etc.;

[0158] S5402, time-aligning the key action changes and position changes of each simulated subject to obtain an aligned sequence;

[0159] In S5402, when performing time alignment, the key action changes are first converted into action change sequences, and the position changes are converted into position change sequences. Then, according to the change time of each sequence item in the two change sequences, the sequence items with the same change time in the two change sequences are aligned to obtain an aligned sequence. In the aligned sequence, there is an action change node, which is a change action, and the position change node aligned with it is a change position that occurs simultaneously with it.

[0160] S5403, traversing multiple action change nodes in the alignment sequence in sequence order;

[0161] In S5403, the sequence order refers to the sequence order of the action change nodes in the alignment sequence;

[0162] S5404, each time the traversal is completed, the simulation subject of the traversed action change node is used as the first target, and the remaining simulation subjects are used as the second targets;

[0163] In S5404, the simulation subject of the action change node is the simulation subject that generates the action change node, and the remaining simulation subjects refer to the simulation subjects other than the first target;

[0164] S5405, determining the first position change node and the second position change node of the first target and the second target respectively from the aligned position change nodes in the aligned sequence from the traversed action change nodes;

[0165] In S5405, the first position change node is generated by the first target, and the second position change node is generated by the second target;

[0166] S5406, determining in the working simulation sandbox whether the simulation perspective when the first target is at the first position change node and the second target is at the second position change node is visible to the traversed action change node; wherein the simulation perspective includes: first-person perspective and third-person perspective;

[0167] In S5406, when determining whether it is visible, a judgment is made based on an actual simulation environment when the first target is at a first position change node and the second target is at a second position change node in the working simulation sandbox;

[0168] S5407: When it is visible, the simulated perspective when the second target is at the second position change node is used as the immersive perspective to be selected; otherwise, the third-person perspective when the first target is at the first position change node is used as the immersive perspective to be selected;

[0169] In S5406, when visible, the explanation personnel can view the traversed action change node through the simulated perspective when the second target is at the second position change node, and use it as the immersive perspective to be selected; otherwise, the third-person perspective when the first target is at the first position change node is used as the immersive perspective to be selected; the action change node can be clearly viewed through the simulated perspective of others, and the surrounding environment of the first target can be fully viewed, which is more suitable. Therefore, the visible simulated perspective is preferred as the immersive perspective to be selected; if it is not visible, the third-person perspective of the first target itself can more fully view the action change node generated by itself, and the third-person perspective of itself is used as the immersive perspective to be selected;

[0170] S5408, associating the to-be-selected immersive perspective with the change time of the traversed action change node;

[0171] In S5408, the action change node has a change time, which refers to the occurrence time of the Dong Zong change node, and the selected immersive perspective is associated with the change time;

[0172] S5409, after traversing all action change nodes, integrate all immersive perspectives to be selected to obtain a perspective set;

[0173] In S5409, the perspective set includes all the immersive perspectives to be selected;

[0174] S5410, optimizing the perspective set based on the set optimization constraint;

[0175] S5411. Determine multiple immersive perspectives based on the optimized perspective set;

[0176] In S5411, all candidate immersive perspectives in the optimized perspective set are used as immersive perspectives;

[0177] The set optimization constraints include:

[0178] Constraint B1: The change time of the associations of different candidate immersive perspectives in the optimized perspective set is different from each other;

[0179] as well as,

[0180] Constraint B2: The number of categories of simulation subjects belonging to different candidate immersive perspectives in the optimized perspective set is the least.

[0181] In constraint B1, it is ensured that the change time associated with different candidate immersive perspectives is different from each other, so that the final immersive perspective can repeatedly lead personnel to immerse and view the work simulation sandbox at the same time;

[0182] In constraint B2, the simulation subjects to which the candidate immersive perspectives associated with the same change time in the perspective set belong may be different. When different candidate immersive perspectives in the perspective set are used to lead personnel to immersively view the work simulation sandbox, there will be multiple choices of candidate immersive perspectives belonging to the simulation subjects at the same time. In the optimization process, to ensure that there is only one associated candidate immersive perspective for the same change time, there will be multiple candidate perspective sets. From these candidate perspective sets, the one with the least categories of simulation subjects belonging to different candidate immersive perspectives is selected as the optimized perspective set, which can ensure that when different immersive perspectives are used to lead personnel to immersively view the work simulation sandbox, the number of perspective switching is minimized.

[0183] The embodiments of the present invention achieve the following beneficial effects:

[0184] Key action screening ensures the value of key action changes, allowing personnel to obtain information closely related to work decisions through the key actions of the simulated subject; through time alignment, it ensures that the action changes and position changes of different simulated subjects can be synchronized on the same timeline, thereby improving the authenticity and consistency of the simulation process; by optimizing the perspective set composed of the selected immersive perspectives, it can provide a diversified and efficient perspective experience while ensuring the least perspective selection and the smoothest switching, thereby enhancing the simulated personnel's overall understanding of the work situation and reducing visual fatigue and information overload during the simulation process.

[0185] Constraint B1 ensures that different immersive perspectives do not overlap in time, thereby avoiding information duplication and visual redundancy, allowing personnel to experience complete and diverse work situations at different time points; constraint B2 ensures that the immersive perspective at each time point belongs to the least category of simulation subjects, which not only reduces the frequency of perspective switching, but also improves the fluency and immersion of the simulation; the combination of constraints B1 and B2 allows the use of optimized perspectives to focus on different immersive perspectives to lead personnel to view the work simulation sandbox, which can provide rich information feedback and effectively guide personnel to deeply analyze and understand the complex situations in the work simulation sandbox, thereby providing effective support for the decision-making of corresponding work plans in the future.

[0186] In one embodiment, the step S55 of generating an immersion trigger constraint for each immersion perspective includes:

[0187] Constraint C1: When a person adjusts his / her viewing angle of the work simulation sandbox and the adjustment time is earlier than the change time associated with the immersive perspective, the time difference does not exceed the time difference threshold;

[0188] and / or,

[0189] Constraint C2: The visual content overlap between the viewing perspective adjusted by the operator and the immersive perspective exceeds the overlap threshold.

[0190] In constraint C1, when a person views the work simulation sandbox, a viewing perspective is generated; the adjustment time refers to the time when the person adjusts the viewing perspective of the work simulation sandbox; the time difference threshold can be 20 seconds; when the time difference between the whole time and the change time associated with the immersive perspective does not exceed the time difference threshold, it means that the person adjusts the viewing perspective shortly before the triggerable time (associated change time) of the immersive perspective, and may want to view the work simulation sandbox more deeply, and the immersive perspective can be triggered;

[0191] In constraint C2, the overlap threshold may be 65%; when the visual content overlap between the viewing perspective adjusted by the person and the immersive perspective exceeds the overlap threshold, it indicates that the person has expressed his / her intention of immersive viewing through the perspective adjustment, and the immersive perspective may be triggered.

[0192] The embodiments of the present invention achieve the following beneficial effects:

[0193] By constraining the coordination of C1 and C2, it is possible to accurately determine the timing of leading personnel to view the work simulation sandbox through an immersive perspective, which improves the user experience and is more intelligent.

[0194] In one embodiment, the generating of the second auxiliary content based on the AI ​​assistant according to the work assistance requirement includes:

[0195] Perform vector transformation on work assistance requirements to obtain the requirement content vector;

[0196] Calculate the similarity between the demand content vector and the document vector in the vector database;

[0197] Generate second auxiliary content based on the document content corresponding to the document vector with the greatest similarity;

[0198] The steps for obtaining the document vector in the vector database are as follows:

[0199] Receive multi-format documents uploaded by personnel; the formats of multi-format documents include at least: TXT, DOCX, PDF;

[0200] Perform vector conversion on the document content of multi-format documents to obtain document vectors;

[0201] The step of converting the document content of the multi-format document into a vector to obtain a document vector includes:

[0202] The document content of multi-format documents is separated into text and converted into document vectors using an embedding model.

[0203] The embedding model at least includes: bge-large-zh-v1.5 Chinese embedding model.

[0204] The vector database uses the PGVector vector library for vector storage.

[0205] After the second auxiliary content is generated based on the document content corresponding to the document vector with the greatest similarity, the generated second auxiliary content is optimized through the chatGLM API.

[0206] When the document content of the multi-format document is converted into a vector, key information in the document content is automatically identified and extracted according to the element direction specified by the user.

[0207] The disadvantages of the prior art to be solved by the embodiments of the present invention are as follows:

[0208] High reliance on manual operations: In the work process, from data collection, input to archiving, a large amount of work relies on manual operations, which is not only inefficient but also prone to errors due to human factors.

[0209] Low information processing efficiency: When handling cases, staff need to read and analyze a large number of documents and manually extract key information. This process is time-consuming and affects the speed of case handling.

[0210] Limited accuracy of information extraction: When manually analyzing documents, it is easy to miss important information or extract inaccurate information due to subjective judgment.

[0211] Difficulty in associating information across documents: When handling complex business, information associations between multiple documents are involved. It is difficult to manually sort out these information associations, which may lead to insufficient mining of case clues.

[0212] The embodiment of the present invention is to significantly improve the efficiency and quality of work with the help of artificial intelligence technology. This goal is mainly achieved through the following advanced functions:

[0213] Through the automated document processing process, the present invention can quickly and accurately extract key information, discover potential correlations and patterns, and help staff explore clues in depth;

[0214] The use of cutting-edge natural language processing technology helps staff quickly grasp the content of documents and effectively absorb relevant knowledge, thereby improving work efficiency;

[0215] Relying on advanced information integration technology, the present invention can identify and integrate related documents, sort out the key elements in the documents, and construct a complex relationship map between the elements; through precise algorithms and pattern recognition, the tool can reduce human errors, ensure the accuracy of the extracted information, and improve the reliability of case information.

[0216] The ultimate goal of these functions is to use artificial intelligence technology to reduce the work pressure of staff, comprehensively improve execution efficiency, and promote technological innovation to meet the challenges of the new era and provide solid technological support for the stability of social order and public safety.

[0217] The implementation aspects of the embodiments of the present invention are as follows:

[0218] 1. Document parsing

[0219] Document upload and text parsing

[0220] Operation: Upload documents in TXT, DOCX, or PDF (PDF only supports text format). The size of a single file should not exceed 10M. A maximum of 10 files can be uploaded.

[0221] Purpose: To build a vector library to facilitate parsing of corresponding files.

[0222] Technology: Read the document content, perform text segmentation, and then use the same embedded model to vectorize the text. (Text segmentation is a tool that breaks down continuous text data into smaller, more meaningful parts. In the field of natural language processing (NLP), text segmentation is an important task that can help us better understand the structure and content of the text, and then conduct further text analysis and processing. The same embedded model refers to the bge-large-zh-v1.5 model)

[0223] Purpose: Convert documents of various formats into vector form to facilitate subsequent data processing and retrieval.

[0224] Data Storage:

[0225] Operation: Save the vectorized text data in the PGVector vector library.

[0226] Purpose: To ensure the uniformity and searchability of knowledge base data.

[0227] 2. User Questions and Answers

[0228] User input handling:

[0229] Action: Vectorize the user's input question.

[0230] Purpose: Convert user questions into vector form for data matching.

[0231] Determine the matching scope:

[0232] Operation: The user selects the files that he wants to parse.

[0233] Purpose: To match the file name selected by the user with the documents in the vector library and determine the parsing scope.

[0234] Vector library matching:

[0235] Action: Perform similarity matching between the user question and the content of the files in the selected vector library.

[0236] Logic: If the corresponding content is matched, the corresponding content and the user question are combined to generate a prompt, and the chatGLM API is called to return the answer based on the matched vector library content.

[0237] Purpose: To obtain accurate answers related to the parsed document.

[0238] Answer optimization and presentation:

[0239] This technical solution applies AI and NLP technologies to build an efficient information retrieval system that can answer user questions based on specified documents.

[0240] 3. Factor Extraction

[0241] Operation: The user specifies the direction of the elements to be extracted or allows the AI ​​to extract the elements in the document autonomously.

[0242] Logic: Build a prompt based on the document content selected by the user, and call the chatGLM API to extract the corresponding elements in the document.

[0243] Purpose: To help users quickly organize document information and obtain key information points.

[0244] The embodiments of the present invention achieve the following beneficial effects:

[0245] 1. Innovation in unified parsing and vectorization of multi-format documents

[0246] This technical solution innovatively realizes the unified parsing and vectorization processing of documents in various formats such as TXT, DOCX, PDF, etc. By using the bge-large-zh-v1.5 model, the document content in different formats is converted into vector form for easy storage and retrieval.

[0247] 2. Intelligent question answering based on vector library

[0248] The present invention allows users to upload documents and build a vector library, and then achieves similarity matching with the document content in the vector library by vectorizing the questions input by the users, so as to query the accurate answers in the selected documents.

[0249] 3. Flexible element extraction function

[0250] The present invention allows the user to specify the direction of extracting elements, and utilizes chatGLM to quickly organize document information and extract key information points.

[0251] 4. Efficiency of data storage and retrieval

[0252] The technical solution uses the PGVector vector library to store vectorized text data, ensuring the uniformity and efficient retrieval of knowledge base data. This data storage method not only optimizes the storage structure, but also improves the speed of data retrieval.

[0253] The embodiment of the present invention provides a personnel work assistance system based on an AI assistant, such as Figure 2 As shown, including:

[0254] The first auxiliary module 1 is used to assist personnel with their work based on an AI assistant when they are working; wherein the AI ​​assistant is an artificial intelligence model trained using a large amount of work experience related to the work profile of the personnel;

[0255] The first auxiliary module provides work assistance to personnel based on an AI assistant, including:

[0256] Obtain multimodal work records of personnel in real time;

[0257] Based on AI assistant, the first auxiliary content is generated according to multimodal work records;

[0258] Pushing first auxiliary content to personnel;

[0259] and / or,

[0260] Receives personnel input on job assistance needs;

[0261] Based on AI assistant, generate second auxiliary content according to work assistance needs;

[0262] Push secondary auxiliary content to the person.

[0263] The personnel work assistance system based on AI assistant also includes:

[0264] The second auxiliary module is used to:

[0265] When personnel use the work simulation sandbox, whether to enter the standard assistance timing is determined based on the real-time simulated content of the work simulation performed by the personnel in the work simulation sandbox;

[0266] When entering, the auxiliary basis is extracted from the real-time simulated content based on the auxiliary basis extraction template of the standard auxiliary timing;

[0267] Based on the AI ​​assistant, determine the auxiliary simulation content according to the auxiliary basis;

[0268] Generate interactive control timelines for auxiliary simulation content;

[0269] Based on the auxiliary simulation content, the corresponding work simulation is carried out in the work simulation sandbox, and during the simulation process, the corresponding interactive control is carried out with the personnel based on the interactive control timeline.

[0270] The second auxiliary module generates an interactive control timeline of auxiliary simulation content, including:

[0271] Get the simulation progress timeline of auxiliary simulation content;

[0272] Based on the time interval division constraint, a plurality of first target time intervals are divided on the simulation progress time axis;

[0273] Traverse each first target time interval in sequence;

[0274] Each time the traversal is completed, multiple immersive perspectives are determined based on the action changes and position changes of multiple simulated subjects within the first target time interval traversed on the simulation progress timeline;

[0275] Generate immersion trigger constraints for each immersion perspective;

[0276] Determine a second target time interval corresponding to the traversed first target time interval from the blank time axis;

[0277] Setting each immersive perspective and its respective immersive trigger constraint within a second target time interval;

[0278] After traversing each first target time interval, a blank timeline in which all immersive perspectives and respective immersive trigger constraints are fully set is used as an interactive control timeline.

[0279] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A personnel work assistance method based on AI assistant, characterized in that: include: When people are working, AI assistants are used to assist them. AI assistants are artificial intelligence models trained using a large amount of work experience related to people's work profiles. The AI ​​assistant-based work assistance for personnel includes: Real-time acquisition of multi-modal work records of personnel; Based on AI assistant, the first auxiliary content is generated according to multimodal work records; Pushing first auxiliary content to personnel; and / or, Receives personnel input on job assistance needs; Based on AI assistant, generate second auxiliary content according to work assistance needs; Push secondary auxiliary content to the person.

2. The personnel work assistance method based on AI assistant as claimed in claim 1, characterized in that: The generating of the second auxiliary content based on the AI ​​assistant according to the work auxiliary demand includes: Perform vector transformation on work assistance requirements to obtain the requirement content vector; Calculate the similarity between the demand content vector and the document vector in the vector database; Generate second auxiliary content based on the document content corresponding to the document vector with the greatest similarity; The steps for obtaining the document vector in the vector database are as follows: Receive multi-format documents uploaded by personnel; the formats of multi-format documents include at least: TXT, DOCX, PDF; Perform vector conversion on the document content of multi-format documents to obtain document vectors; The step of converting the document content of the multi-format document into a vector to obtain a document vector includes: The document content of multi-format documents is separated into text and converted into document vectors using an embedding model.

3. The personnel work assistance method based on AI assistant as claimed in claim 2, characterized in that: The embedding model at least includes: bge-large-zh-v1.5 Chinese embedding model.

4. The personnel work assistance method based on AI assistant as claimed in claim 2, characterized in that: The vector database uses the PGVector vector library for vector storage.

5. The personnel work assistance method based on AI assistant as claimed in claim 2, characterized in that: After the second auxiliary content is generated based on the document content corresponding to the document vector with the greatest similarity, the generated second auxiliary content is optimized through the chatGLM API.

6. The personnel work assistance method based on AI assistant as claimed in claim 2, characterized in that: When the document content of the multi-format document is converted into a vector, key information in the document content is automatically identified and extracted according to the element direction specified by the user.

7. The personnel work assistance method based on AI assistant as claimed in claim 1, characterized in that: Also includes: When personnel use the work simulation sandbox, whether to enter the standard assistance timing is determined based on the real-time simulated content of the work simulation performed by the personnel in the work simulation sandbox; When entering, the auxiliary basis is extracted from the real-time simulated content based on the auxiliary basis extraction template of the standard auxiliary timing; Based on the AI ​​assistant, determine the auxiliary simulation content according to the auxiliary basis; Generate interactive control timelines for auxiliary simulation content; Based on the auxiliary simulation content, the corresponding work simulation is carried out in the work simulation sandbox, and during the simulation process, the corresponding interactive control is carried out with the personnel based on the interactive control timeline.

8. The personnel work assistance method based on AI assistant as claimed in claim 7, characterized in that: The interactive control timeline for generating auxiliary simulation content includes: Get the simulation progress timeline of auxiliary simulation content; Based on the time interval division constraint, a plurality of first target time intervals are divided on the simulation progress time axis; Traverse each first target time interval in sequence; Each time the traversal is completed, multiple immersive perspectives are determined based on the action changes and position changes of multiple simulated subjects within the first target time interval traversed on the simulation progress timeline; Generate immersion trigger constraints for each immersion perspective; Determine a second target time interval corresponding to the traversed first target time interval from the blank time axis; Setting each immersive perspective and its respective immersive trigger constraint within a second target time interval; After traversing each first target time interval, a blank timeline in which all immersive perspectives and respective immersive trigger constraints are fully set is used as an interactive control timeline.

9. The personnel work assistance method based on AI assistant as claimed in claim 8, characterized in that: The time interval division constraints include: Different first target time intervals each completely and independently include continuous simulation time periods of at least N simulation situations; wherein N is a positive integer greater than or equal to 2; as well as, There is at least one attribute association relationship between any two simulation situations in consecutive simulation time periods included in the same first target time interval.

10. A personnel work assistance system based on AI assistant, characterized in that: include: The first auxiliary module is used to assist personnel with their work based on an AI assistant when they are working; wherein the AI ​​assistant is an artificial intelligence model trained using a large amount of work experience related to the work profile of the personnel; The first auxiliary module provides work assistance to personnel based on an AI assistant, including: Real-time acquisition of multi-modal work records of personnel; Based on AI assistant, the first auxiliary content is generated according to multimodal work records; Pushing first auxiliary content to personnel; and / or, Receives personnel input on job assistance needs; Based on AI assistant, generate second auxiliary content according to work assistance needs; Push secondary auxiliary content to the person.

Images