A virtual digital human interaction method and interactive display terminal

Abstract

This invention relates to the field of virtual digital human interaction technology, specifically disclosing a virtual digital human interaction method and an interactive display terminal. The method includes: S1, real-time acquisition of image data from the user reception area to determine if a user has an interaction need; S2, when a user's interaction need is detected, the virtual digital human interacts with the user, simultaneously acquiring environmental data; S3, analysis of the acquired image data and environmental data, and analysis of the possible cause of interaction obstruction when potential interaction obstruction is detected. If the possible cause of interaction obstruction is environmental noise, step S4 is executed; otherwise, a remote assistance command is generated. The interaction method proposed in this invention can reduce the probability of false wake-up of the virtual digital human in complex environments, is suitable for public areas with complex environments and high environmental noise, and effectively reduces the remote assistance trigger rate, saving human customer service resources.

Description

Technical Field

[0001] This invention relates to the field of virtual digital human interaction technology, specifically to a virtual digital human interaction method and an interactive display terminal. Background Technology

[0002] Virtual digital humans refer to digital virtual characters that exist in the non-physical world, created by comprehensively utilizing technologies such as computer graphics, graphics rendering, motion capture, deep learning, and speech synthesis. With the rapid development of artificial intelligence and virtual reality technologies, virtual digital humans are gradually being applied to scenarios such as customer service and tour guidance.

[0003] While existing virtual digital humans can meet customer service and navigation needs to some extent during user interaction, there are still some shortcomings. For example, traditional interaction methods rely on a single signal wake-up (such as voice wake-up or button trigger) to determine user needs, which can easily lead to misjudgment in some complex scenarios. Also, in noisy environments, if the user is someone who has difficulty interacting with text (such as elderly people with poor eyesight or illiterate people), the virtual digital human is easily affected by noise interference when interacting with the user through voice, which can hinder the interaction and affect the user experience and interaction efficiency. Summary of the Invention

[0004] The purpose of this invention is to provide an interactive method and interactive display terminal for virtual digital humans, solving the following technical problems:

[0005] The question of how to improve the interaction efficiency of virtual digital humans in complex environments.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A method for interacting with a virtual digital human, the method comprising:

[0008] S1. Real-time acquisition of image data from the user reception area to determine whether there is a user's interactive need;

[0009] S2. When a user's interaction needs are detected, the virtual digital human interacts with the user and collects environmental data simultaneously.

[0010] S3. Analyze the collected image data and environmental data. When a possible interaction obstruction is detected, analyze and determine the possible cause of the interaction obstruction. If the possible cause of the interaction obstruction is environmental noise, then execute step S4; otherwise, generate a remote assistance command.

[0011] S4. The virtual digital human moves to the appropriate area according to the dynamic path, interacts with the user again, and sends friendly prompts during the movement, while monitoring the user's location.

[0012] S5. If the user is still unable to complete the interactive operation, a remote assistance command will be generated.

[0013] Furthermore, the process of determining whether a user has an interaction need includes:

[0014] By analyzing the collected image data, the angle between the user's line of sight and the normal of the interactive interface within the reception area, as well as the straight-line distance between the user and the interactive terminal, can be obtained.

[0015] When the angle between the user's line of sight and the normal of the interactive interface is less than the preset angle and the duration is greater than or equal to the preset first time period, the user is judged to be staring at the interactive interface.

[0016] When the straight-line distance between the user and the interactive terminal is less than the preset distance and the duration is greater than or equal to the preset second time period, the user is judged to be in a state where interaction is convenient.

[0017] When a user is simultaneously in an interactive state and a staring interactive interface state, it is determined that the user has an interactive need.

[0018] Furthermore, in step S3, the process of analyzing the collected image data and environmental data includes:

[0019] Q=A*ω1+B*ω2+C*ω3 (1)

[0020]

[0021] The characteristic value Q of user interaction obstruction is obtained by analyzing and calculating using formulas (1)-(4);

[0022] Where A is the unrecognized voice coefficient of the virtual digital human, B is the negative feedback posture coefficient of the user, C is the user response latency rate, ω1, ω2, and ω3 are the weight coefficients corresponding to the factors that hinder the interaction, and S in S represents the number of voices received by the virtual digital human. out The number of successfully recognized voices for the virtual digital human, where N is the preset number of negative user feedback gestures, i∈[1,N], W i Let T be the negative feedback attitude evaluation value of the i-th user. s T is the time from when an interactive command is issued to the user's response for the virtual digital human. std This serves as the preset baseline control time;

[0023] The user interaction obstruction feature value Q is compared with the preset interaction obstruction threshold Q. risk Perform a comparison;

[0024] If Q≥Q risk This indicates that the user's interaction may be blocked;

[0025] Conversely, it indicates that the user is interacting normally with the virtual digital human.

[0026] Furthermore, the process of analyzing and determining the possible reasons for user interaction disruption includes:

[0027] The interference intensity of ambient noise is quantified by acoustic features. If the quantified interference intensity value is greater than the preset maximum interference intensity threshold, it is determined that the reason for the user's interaction may be blocked by ambient noise.

[0028] If the interference intensity value is greater than the preset minimum interference intensity threshold and less than or equal to the preset maximum interference intensity threshold, the voice signal quality will be further analyzed.

[0029] If the interference intensity value is less than or equal to the preset minimum interference intensity threshold, it is determined that the reason for the user's interaction being blocked is not environmental noise.

[0030] Furthermore, the process of obtaining the interference intensity value includes:

[0031]

[0032] The interference intensity value is obtained by analysis and calculation using formulas (5)-(6).

[0033] Where M is the number of environmental sound sampling points, j∈[1,M], D j E represents the instantaneous sampled decibel value of the j-th sampling point of the ambient sound. s E represents the average decibel value of ambient sound. std E is the preset control decibel value. b This is the baseline decibel value for ambient background noise.

[0034] Furthermore, the process of further analyzing the quality of the speech signal includes:

[0035]

[0036] The speech quality coefficient ρ is obtained by analysis and calculation using formula (7);

[0037] Among them, R v Human voice frequency band spectral entropy, R n For the non-human voice frequency band spectral entropy, G s For speech signal-to-noise ratio;

[0038] Compare the voice quality coefficient ρ with the preset judgment coefficient ρ1;

[0039] If ρ < ρ1, then the reason why the user's interaction may be blocked is environmental noise;

[0040] Conversely, if the user's interaction is blocked, it is determined that the reason is not environmental noise.

[0041] Further, in step S4, the process of monitoring the user's state position includes:

[0042]

[0043] Analyze and calculate the distance d between the virtual digital human and the user through formula (8) s ;

[0044] where (x1, y1) are the coordinates of the virtual digital human and (x2, y2) are the user's coordinates;

[0045] When the distance d between the virtual digital human and the user is monitored s ≥ d1, the virtual digital human stops moving and continuously monitors the user's position;

[0046] Conversely, the virtual digital human continues to move until it moves to an appropriate area;

[0047] When the virtual digital human stops moving, within the third preset time period, if the distance d between the virtual digital human and the user at any time point s < d2, the virtual digital human continues to move according to the dynamic path;

[0048] Conversely, the virtual digital human returns to the initial parking position according to the dynamic path.

[0049] Further, the process of generating the remote assistance instruction includes priority evaluation:

[0050]

[0051] Analyze and calculate the priority evaluation coefficient F through formula (9);

[0052] where P is the number of times of interaction blockage, P std is the preset number of interaction comparison times, k ∈ [1, P], Q k is the kth interaction blockage eigenvalue, and L is the user type adjustment coefficient;

[0053] In the process of generating the remote assistance instruction, generate the remote assistance instruction corresponding to the priority according to the evaluation coefficient of the priority.

[0054] An interactive display terminal, comprising:

[0055] A mobile base equipped with universal wheels and universal drive wheels;

[0056] A sensor group including a microphone, a camera, and a lidar. The microphone is used to collect the sound information around the virtual digital human, the camera is used to collect the image information of the reception area, and the lidar is used to collect the obstacle information around the virtual digital human;

[0057] A processing control unit is used to execute an interaction method for a virtual digital human.

[0058] The display interface is used to show the appearance and movements of the virtual digital human and supports user screen touch interaction.

[0059] The beneficial effects of this invention are:

[0060] (1) The interactive method proposed in this invention can reduce the probability of false wake-up of virtual digital humans in complex places. It is suitable for complex public areas with high environmental noise, such as shopping malls, stations, and hospitals. Normally, the device can be set up in public areas to make it easy for users to find the area to use. When the environmental noise is too high and affects the interaction, it can automatically lead the user to the appropriate area for interaction. It also effectively reduces the remote assistance trigger rate, saves human customer service resources, and supports 24-hour uninterrupted service, improving service quality and efficiency.

[0061] (2) This invention accurately identifies interaction needs and dynamically optimizes strategies through multimodal perception and real-time intelligent analysis, significantly improving the efficiency and reliability of human-computer interaction, and is suitable for intelligent terminal applications in complex scenarios.

[0062] (3) This invention calculates the spatial distance between the virtual digital human and the user in real time, which has the effect of dynamically adjusting the movement strategy, ensuring that the user is always within the optimal interaction range and improving the interaction experience.

[0063] (4) This invention integrates historical interaction data, obstruction features and user type features through a priority evaluation formula, which has the effect of intelligently identifying high-priority needs, triggering remote assistance instructions first, and improving service response efficiency. Attached Figure Description

[0064] The invention will now be further described with reference to the accompanying drawings.

[0065] Figure 1 This is a flowchart of the steps of an interactive method for a virtual digital human proposed in this invention;

[0066] Figure 2 This is a schematic block diagram of an interactive display terminal proposed in this invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] Please see Figure 1 As shown, in one embodiment, an interaction method for a virtual digital human is provided, the method comprising:

[0069] S1. Combining camera image analysis, microphone array sound source localization, pressure sensor and other devices, real-time image data of the user reception area is collected. The user reception area is a designated area near a pre-set virtual digital human. Through machine learning model analysis of user dwell time, gaze direction, gestures and other multi-dimensional data, it is determined whether the user has interactive needs.

[0070] S2. When a user's interaction needs are detected, the virtual digital human interacts with the user and collects environmental data simultaneously, including but not limited to environmental noise, surrounding obstacles, and light intensity.

[0071] S3. Analyze the collected image data and environmental data. Through multi-dimensional data analysis, determine whether there is any interaction obstruction during the interaction between the virtual digital human and the user. If possible interaction obstruction is detected, analyze and determine the possible cause of the interaction obstruction. If the possible cause of the interaction obstruction is environmental noise, then execute step S4. Otherwise, generate a remote assistance command and then conduct a video call with customer service.

[0072] S4. The virtual digital human constructs an environmental map using LiDAR, uses intelligent algorithms to plan and generate dynamic paths, moves to an appropriate area according to the dynamic path, and interacts with the user again. During the movement, friendly prompts are issued simultaneously to prompt the user to follow the virtual digital human and to monitor the user's location.

[0073] S5. If the user is still unable to complete the interactive operation, a remote assistance instruction will be generated, and then a video call will be made with customer service.

[0074] Through the above technical solution, this embodiment provides an interaction method for virtual digital humans. The method involves real-time image acquisition and analysis of the user reception area to determine if a user has an interaction need. When an interaction need is detected, the virtual digital human interacts with the user, simultaneously collecting environmental data. If the interaction process is suspected of being obstructed, the method analyzes the possible causes. If the obstruction is due to environmental noise, the virtual digital human guides the user to a suitable area via a dynamic path, enabling interaction. If the obstruction is not due to environmental noise or if interaction cannot be completed even after moving to a suitable area, a remote assistance command is generated, triggering a video call with online customer service to provide service. This interaction method reduces the probability of false wake-up of the virtual digital human in complex environments. It is suitable for complex and noisy public areas such as shopping malls, stations, and hospitals. The device is typically placed in public areas for easy user discovery. When excessive environmental noise affects interaction, it automatically guides the user to a suitable area for interaction, effectively reducing the remote assistance trigger rate, saving human customer service resources, and supporting 24 / 7 uninterrupted service, thus improving service quality and efficiency.

[0075] In one embodiment, the process of determining whether a user has an interaction need includes:

[0076] By analyzing the collected image data, and taking the normal direction of the interactive interface as the reference axis (such as the direction perpendicular to the screen), the angle between the line connecting the user's pupils and the normal of the interactive interface, as well as the straight-line distance between the user and the interactive terminal, are calculated using a convolutional neural network.

[0077] When the angle between the user's line of sight and the normal of the interactive interface is less than the preset angle (usually preset to 30 degrees) and the duration is greater than or equal to the preset first time period (usually preset to 5 seconds), the user is judged to be staring at the interactive interface.

[0078] When the straight-line distance between the user and the interactive terminal is less than the preset distance (usually preset to 1.5 meters) and the duration is greater than or equal to the preset second time period (usually preset to 3 seconds), the user is judged to be in a state where interaction is convenient.

[0079] When a user is simultaneously in an interactive state and a staring interactive interface state, it is determined that the user has an interactive need.

[0080] Step S3, the process of analyzing the collected image data and environmental data, includes:

[0081] Q=A*ω1+B*ω2+C*ω3 (1)

[0082]

[0083] The characteristic value Q of user interaction obstruction is obtained by analyzing and calculating using formulas (1)-(4);

[0084] Where A is the unrecognized voice coefficient of the virtual digital human, B is the negative feedback posture coefficient of the user, C is the user response latency rate, ω1, ω2, and ω3 are the weight coefficients corresponding to the factors that hinder the interaction, which can be set according to experimental data, and S in S represents the total amount of speech received by the virtual digital human. out The number of successfully recognized voices for the virtual digital human, where N is the preset number of negative user feedback gestures, obtained empirically, i∈[1,N], W i The negative feedback posture evaluation value for the i-th user can be obtained through empirical presets or by training a convolutional neural network. For example, the negative posture evaluation value for frowning is preset to 0.8, the negative posture evaluation value for shaking one's head is preset to 0.9, and the negative posture evaluation value for listening is preset to 1. s T is the time from when an interactive command is issued to the user's response for the virtual digital human. std The preset baseline control time can be obtained based on experience.

[0085] The user interaction obstruction feature value Q is compared with the preset interaction obstruction threshold Q. risk The preset interaction obstruction threshold Q is compared. risk It can be obtained by setting based on experimental data;

[0086] If Q≥Q risk This indicates that the user's interaction may be blocked;

[0087] Conversely, it indicates that the user is interacting normally with the virtual digital human.

[0088] The process of analyzing and determining the possible reasons for user interaction obstacles includes:

[0089] The interference intensity of ambient noise is quantified by acoustic features. If the quantified interference intensity value is greater than the preset maximum interference intensity threshold, it is determined that the reason for the user's interaction may be blocked by ambient noise.

[0090] If the interference intensity value is greater than the preset minimum interference intensity threshold and less than or equal to the preset maximum interference intensity threshold, the voice signal quality will be further analyzed.

[0091] If the interference intensity value is less than or equal to the preset minimum interference intensity threshold, it is determined that the reason for the user's interaction being blocked is not environmental noise.

[0092] The process of obtaining the interference intensity value includes:

[0093]

[0094] The interference intensity value is obtained by analysis and calculation using formulas (5)-(6).

[0095] Where M is the number of environmental sound sampling points, which is obtained based on empirical presets, j∈[1,M], D j E is the instantaneous decibel value of the j-th sampling point of the ambient sound, obtained through a microphone. s E represents the average decibel value of ambient sound. std E is a preset control decibel value, obtained based on experience. b The baseline decibel value for ambient noise can be obtained through long-term monitoring and statistics when there are no users.

[0096] The process of further analyzing the quality of the speech signal includes:

[0097]

[0098] The speech quality coefficient ρ is obtained by analysis and calculation using formula (7);

[0099] Among them, R v The spectral entropy of the human voice band is a quantitative indicator of the spectral complexity of the dominant human voice frequency band (typically 300–3400 Hz) in a speech signal. It is obtained by calculating the distribution entropy value of the spectral energy within the dominant human voice frequency band. R n G is the spectral entropy of the non-human voice frequency band, a quantitative indicator of the spectral complexity outside the human voice frequency band (such as low-frequency noise <300Hz or high-frequency noise >3400Hz). It is obtained by calculating the distribution entropy value of the spectral energy outside the human voice frequency band. s The speech signal-to-noise ratio (SNR) is obtained by calculating the ratio of the speech signal power to the background noise power.

[0100] Compare the voice quality coefficient ρ with the preset judgment coefficient ρ1;

[0101] If ρ < ρ1, then the reason why the user's interaction may be blocked is environmental noise;

[0102] Conversely, if the user's interaction is blocked, it is determined that the reason is not environmental noise.

[0103] Through the above technical solution, this embodiment provides a method for determining whether there is an interaction obstruction. The method accurately identifies interaction needs and dynamically optimizes strategies through multimodal perception and real-time intelligent analysis, significantly improving the efficiency and reliability of human-computer interaction, and is suitable for intelligent terminal applications in complex scenarios.

[0104] In one embodiment, step S4, the process of monitoring the user's location status, includes:

[0105]

[0106] The distance d between the virtual digital human and the user is obtained through analytical calculation using formula (8). s ;

[0107] Among them, (x1, y1) are the coordinates of the virtual digital human, obtained through calculating the environmental map by lidar, and (x2, y2) are the coordinates of the user, obtained through calculating the environmental map by lidar;

[0108] When the distance d between the virtual digital human and the user is monitored s ≥ d1, the virtual digital human stops moving and continuously monitors the user's position. The d1 is the first distance threshold, which is the trigger distance for whether the virtual digital human needs to stop moving and is preset according to experience;

[0109] On the contrary, the virtual digital human continues to move until it moves to an appropriate area;

[0110] When the virtual digital human stops moving, within the third preset time period (generally preset to 10 seconds), if at any time point the distance d between the virtual digital human and the user s < d2, the virtual digital human continues to move according to the dynamic path. The d2 is the second distance threshold, which is the sensitive distance for judging whether the virtual digital human needs to move again after stopping;

[0111] On the contrary, the virtual digital human returns to the initial parking position according to the dynamic path.

[0112] The process of generating the remote assistance instruction includes priority evaluation:

[0113]

[0114] The priority evaluation coefficient F is obtained through analytical calculation using formula (9);

[0115] Among them, P is the number of times of interaction blockage, obtained by statistical analysis of historical data. It should be noted that the statistically obtained data is the number of times of interaction blockage of the current user, P std is the preset interaction comparison number, preset according to experience, k ∈ [1, P], Q k is the kth interaction blockage eigenvalue, and L is the user type adjustment coefficient, which can be obtained by identifying the collected image information. For example, the adjustment coefficient corresponding to the elderly is 1, and the adjustment coefficient corresponding to the young is 0.2;

[0116] During the process of generating the remote assistance instruction, according to the evaluation coefficient of the priority, a remote assistance instruction corresponding to the priority is generated.

[0117] Through the above technical solution, this embodiment provides a virtual digital human interaction method with dynamic distance monitoring and priority assessment. The method calculates the spatial distance between the virtual digital human and the user in real time, which has the effect of dynamically adjusting the movement strategy, ensuring that the user is always within the optimal interaction range and improving the interaction experience. By integrating historical interaction data, obstruction characteristics and user type characteristics through the priority assessment formula, it has the effect of intelligently identifying high-priority needs, triggering remote assistance commands first, and improving service response efficiency.

[0118] Please see Figure 2 As shown, in one embodiment, an interactive display terminal is provided, comprising:

[0119] Mobile base, equipped with omnidirectional wheels and omnidirectional drive wheels;

[0120] The sensor group includes a microphone, a camera, and a lidar. The microphone is used to collect sound information around the virtual digital human, the camera is used to collect image information of the reception area, and the lidar is used to collect obstacle information around the virtual digital human.

[0121] A processing control unit is used to execute an interaction method for a virtual digital human.

[0122] The display interface is used to show the appearance and movements of the virtual digital human and supports user screen touch interaction.

[0123] The interactive terminal integrates multimodal interaction technology (voice / vision / touch), an autonomous navigation system, and virtual digital human intelligent algorithms to achieve accurate perception and natural interaction in dynamic scenarios, improve service response efficiency, reduce labor costs, and has significant effects in complex environments.

[0124] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. An interaction method for a virtual digital human, characterized in that, The method includes: S1. Real-time acquisition of image data from the user reception area to determine whether there is a user's interactive need; S2. When a user's interaction needs are detected, the virtual digital human interacts with the user and collects environmental data simultaneously. S3. Analyze the collected image data and environmental data. When a possible interaction obstruction is detected, analyze and determine the possible cause of the interaction obstruction. If the possible cause of the interaction obstruction is environmental noise, then execute step S4; otherwise, generate a remote assistance command. S4. The virtual digital human moves to the appropriate area according to the dynamic path, interacts with the user again, and sends friendly prompts during the movement, while monitoring the user's location. S5. If the user is still unable to complete the interactive operation, a remote assistance command will be generated. The process of analyzing and determining the possible reasons for user interaction obstacles includes: The interference intensity of ambient noise is quantified by acoustic features. If the quantified interference intensity value is greater than the preset maximum interference intensity threshold, it is determined that the reason for the user's interaction may be blocked by ambient noise. If the interference intensity value is greater than the preset minimum interference intensity threshold and less than or equal to the preset maximum interference intensity threshold, the voice signal quality will be further analyzed. If the interference intensity value is less than or equal to the preset minimum interference intensity threshold, it is determined that the reason for the user's interaction being blocked is not environmental noise.

2. The interaction method for a virtual digital human according to claim 1, characterized in that, The process of determining whether a user has an interaction need includes: By analyzing the collected image data, the angle between the user's line of sight and the normal of the interactive interface within the reception area, as well as the straight-line distance between the user and the interactive terminal, can be obtained. When the angle between the user's line of sight and the normal of the interactive interface is less than the preset angle and the duration is greater than or equal to the preset first time period, the user is judged to be staring at the interactive interface. When the straight-line distance between the user and the interactive terminal is less than the preset distance and the duration is greater than or equal to the preset second time period, the user is judged to be in a state where interaction is convenient. When a user is simultaneously in an interactive state and a staring interactive interface state, it is determined that the user has an interactive need.

3. The interaction method for a virtual digital human according to claim 2, characterized in that, Step S3, the process of analyzing the collected image data and environmental data, includes: ； The characteristic values ​​of user interaction obstruction are obtained by analyzing and calculating using formulas (1)-(4). ; in, A represents the unrecognized voice coefficient of the virtual digital human, B represents the negative feedback posture coefficient of the user, and C represents the user response latency rate. , , To determine the weight coefficients corresponding to the factors hindering interaction, The amount of speech received by the virtual digital human. The number of voices successfully recognized for the virtual digital human. The preset number of negative user feedback gestures. , Let i be the negative feedback attitude evaluation value of the i-th user. The time from issuing an interactive command to the user's response for the virtual digital human. This serves as the preset baseline control time; User interaction obstruction characteristics Interaction blocked threshold Perform a comparison; like This indicates that the user's interaction may be blocked; Conversely, it indicates that the user is interacting normally with the virtual digital human.

4. The interaction method for a virtual digital human according to claim 3, characterized in that, The process of obtaining the interference intensity value includes: ； The interference intensity value is obtained by analysis and calculation using formulas (5)-(6). ; Where M is the number of environmental sound sampling points, , Let J be the instantaneous decibel value of the j-th sampling point of the ambient sound. The average decibel value of ambient sound. To preset the control decibel value, This is the baseline decibel value for ambient background noise.

5. The interaction method for a virtual digital human according to claim 4, characterized in that, The process of further analyzing the quality of the speech signal includes: ； The speech quality coefficient is obtained by analysis and calculation using formula (7). ; in, The spectral entropy of the human voice frequency band, For non-human voice frequency band spectral entropy, For speech signal-to-noise ratio; Voice quality coefficient With preset judgment coefficient Perform a comparison; like If so, the reason why the user's interaction may be blocked is determined to be environmental noise; Conversely, if the user's interaction is blocked, it is determined that the reason is not environmental noise.

6. The interaction method for a virtual digital human according to claim 5, characterized in that, Step S4, the process of monitoring the user's location status includes: ； The distance between the virtual digital person and the user is obtained by analyzing and calculating using formula (8). ; in, For virtual digital human coordinates, User coordinates; When the distance between the virtual digital human and the user is detected At that time, the virtual digital human stops moving and continues to monitor the user's location; Conversely, the virtual digital human continues to move until it reaches the appropriate area; When the virtual digital human stops moving, within the third preset time period, if the distance between the virtual digital human and the user at any point in time... Then the virtual digital human continues to move according to the dynamic path; Conversely, the virtual digital human returns to its initial parking position based on a dynamic path.

7. The interaction method for a virtual digital human according to claim 6, characterized in that, The remote assistance command generation process includes priority evaluation: ； The priority evaluation coefficient is obtained by analyzing and calculating using formula (9). ; in, The number of times the interaction was blocked. To preset the number of interactive comparisons, , The eigenvalue of the k-th interaction is blocked. Adjust the coefficients for user types; During the remote assistance instruction generation process, remote assistance instructions of corresponding priorities are generated based on the priority evaluation coefficient.

8. An interactive display terminal, characterized in that, include: Mobile base, equipped with omnidirectional wheels and omnidirectional drive wheels; The sensor group includes a microphone, a camera, and a lidar. The microphone is used to collect sound information around the virtual digital human, the camera is used to collect image information of the reception area, and the lidar is used to collect obstacle information around the virtual digital human. A processing control unit is configured to execute an interaction method for a virtual digital human according to any one of claims 1-7; The display interface is used to show the appearance and movements of the virtual digital human and supports user screen touch interaction.

Images