An intelligent robot for power marketing based on large model technology

Abstract

A kind of power marketing intelligent robot based on large model technology, the robot is equipped with control system, control system includes image acquisition module and central control analysis module, image acquisition module utilizes camera real-time capture user facial expression and is transmitted to central control analysis module, central control analysis module calculates the head inclination of user by large model algorithm, judges the height deviation of screen and line of sight;Respectively using expression recognition algorithm detects user squint, frown action;Distance adjusting module controls mobile mechanism to adjust the distance between screen and user, height adjusting module drives lifting device to calibrate line of sight angle;When squint is detected, automatically optimize screen brightness, contrast and font size;Recognize frown, extend current page residence time, and superimposed expansion explanation information.The present application can accurately understand user intention, real-time adjust interactive rhythm, significantly improve the use comfort of different user groups, realize intelligent, adaptive power marketing service experience.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, specifically to an intelligent robot for electricity marketing based on large model technology. Background Technology

[0002] Electricity marketing is the process by which power companies, in a changing market environment, aim to meet people's electricity consumption needs by providing electricity products and related services through a series of market-related business activities. In the field of electricity marketing, intelligent telemarketing robots can be used for customer consultation, electricity product promotion, and fault handling, helping companies achieve more efficient, accurate, and personalized sales and driving innovation in the electricity marketing industry. Essentially, an intelligent telemarketing robot is a software program based on artificial intelligence technology, designed to simulate human language and behavior, engage in intelligent dialogue with customers, and achieve business goals such as sales, promotion, and customer service.

[0003] For example, patent document CN117557222A discloses an intelligent bill collection robot system based on electricity marketing. This invention, through the user classification module and IVR voice platform in the intelligent bill collection robot system, can realize batch intelligent call reminders and intelligent call collection for users' electricity bill payments. It can also intelligently notify users of electricity payment policies via telephone calls, which greatly reduces the labor costs required by power companies in the electricity bill collection process. While improving the efficiency and success rate of electricity bill collection, it also makes the collection records traceable, further improving work efficiency and user satisfaction with the power company's services. In addition, through the voice interaction module in the IVR voice platform, it can also realize intelligent interaction with users, allowing users to understand and process electricity bill-related data through more channels and more conveniently, enhancing the user experience.

[0004] With the advancement of technology, the technology of electricity marketing robots has become increasingly mature. In addition to virtual robots used online, as mentioned above, more and more physical electricity marketing robots are being deployed in the offline stores of power companies. Compared with traditional human service, telemarketing robots can provide services 24 hours a day without rest, greatly increasing the coverage and frequency of sales activities. Moreover, telemarketing robots can strictly follow preset scripts and logic in their conversations, ensuring consistency and professionalism in every communication, and avoiding the inconsistent communication results that may be caused by factors such as emotions and experience in human service.

[0005] However, in practical use, electricity marketing is an activity that needs to be carried out continuously for a period of time, and the user's emotions (comfort level) directly affect their patience and thus the conversion rate. Therefore, providing a high level of comfort is crucial for electricity marketing robots. However, most intelligent robots currently on the market can only make some basic adaptive adjustments, such as adjusting their height according to the user's height, which is clearly insufficient for electricity marketing. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent robot for electricity marketing based on large-scale modeling technology, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an intelligent robot for electricity marketing based on large-scale modeling technology, comprising:

[0008] The control system includes:

[0009] The image acquisition module is configured to capture the user's facial expression images in real time via a camera and transmit them to the central control and analysis module.

[0010] The central control analysis module includes:

[0011] The first analysis unit is used to analyze the head tilt angle of the user when viewing the screen based on a large model algorithm, and to determine the height deviation between the robot and the user's line of sight.

[0012] The second analysis unit is used to detect whether the user has a squinting expression using an expression recognition algorithm;

[0013] The third analysis unit is used to detect whether the user has a frowning expression using an expression recognition algorithm;

[0014] The distance adjustment module is connected to the first analysis unit and controls the moving mechanism to perform forward or backward or left or right displacement according to the height deviation, so that the robot screen maintains a preset distance range from the user.

[0015] The height adjustment module drives the lifting mechanism to adjust the vertical height of the robot screen based on the height deviation output by the first analysis unit, so that the robot screen maintains a suitable angle with the user's line of sight.

[0016] The screen adjustment module dynamically adjusts the screen's brightness, contrast, and font size in response to the squinting detection signal from the second analysis unit.

[0017] The response adjustment module, after receiving the frown detection signal from the third analysis unit, triggers the delay pause control of the current page on the screen and generates extended explanation information to be overlaid and displayed on the paused page.

[0018] Furthermore, the control system also includes:

[0019] Voice prompt module: used to remind the user to relax and look at the screen when the user interacts with the robot; the voice prompt module uses text-to-speech technology to convert preset text information into voice output;

[0020] Infrared ranging module: used to measure the distance between the user and the robot; the infrared ranging module emits infrared light through an infrared sensor and receives the reflected light, and calculates the distance based on the time difference or phase difference;

[0021] Distance judgment module: Based on the data provided by the infrared ranging module, it determines whether the distance between the user and the robot is within a suitable range; the distance judgment module adjusts the robot's movement parameters by comparing the measured distance with a preset threshold using a PID control algorithm.

[0022] Furthermore, the first analysis unit analyzes the user's head tilt angle when viewing the screen based on a large model algorithm to determine the height deviation between the robot and the user's line of sight, including:

[0023] Feature point detection: Using Haar features + Adaboost classifier, key feature points of the user's head are detected from the user's facial image. The feature points include facial features such as the corners of the eyes, the corners of the mouth, and the tip of the nose.

[0024] Facial contour fitting: After detecting feature points, these points are used to fit the contour of the user's face. The fitting methods include ellipse fitting and polygon fitting.

[0025] Head pose estimation: After obtaining the facial contour or feature points, the head pose is estimated using geometric algorithms or machine learning algorithms, that is, the tilt angle of the head is calculated. The geometric algorithm is based on the relative positional relationship of the facial feature points and uses geometric principles to calculate the rotation angle of the head, or uses the fitting results of the facial contour to estimate the tilt direction of the head.

[0026] Tilt angle judgment and adjustment: The calculated tilt angle is compared with the preset suitable tilt angle range. If the tilt angle is not within the suitable range, the height adjustment module is triggered to adjust the height of the robot accordingly.

[0027] Furthermore, the machine learning algorithm involves training a regression model, using detected feature points or facial contours as input and head tilt angle as output. During the testing phase, newly captured facial images are input into the trained model to predict the head tilt angle.

[0028] Furthermore, the second analysis unit detects whether the user has a squinting expression using an expression recognition algorithm, including:

[0029] Feature extraction: Using image processing technology, key facial features of the user are extracted from the preprocessed image. These features include the shape, size, and position of the eyes, as well as the skin texture around the eyes.

[0030] Squinting detection: Based on the extracted feature information, machine learning or deep learning algorithms are used to analyze the eye region, specifically including the following algorithms:

[0031] 1) Geometric feature-based method: By calculating geometric features such as the width and height of the eye area and the degree of eye opening and closing, it can be determined whether the user is squinting;

[0032] 2) Template matching method: A template image of squinting state is predefined, and the extracted eye area is matched with the template. The degree of matching is used to determine whether the user is squinting.

[0033] 3) Deep learning-based method: The deep learning model is trained on a large number of squinting and non-squinting images, so that the model can automatically learn the feature representation of squinting and achieve accurate recognition of squinting state in practical applications.

[0034] Judgment and Feedback: Based on the squinting detection results, if it is determined that the user is squinting or has other facial expressions indicating that they cannot see the screen clearly, the corresponding feedback mechanism is triggered.

[0035] Furthermore, the feedback mechanism specifically works as follows: if the user is squinting, the distance adjustment module is triggered a second time to allow the robot to move autonomously. During the movement, the robot detects changes in the user's facial expressions in real time until the robot moves to the closest distance within the safe observation range. If the problem is still not resolved, the screen adjustment module is triggered to resolve the problem by adjusting the screen brightness, contrast, font size, etc.

[0036] Furthermore, the third analysis unit detects whether the user has a frowning expression using an expression recognition algorithm, including:

[0037] Feature extraction: Using image processing technology, key facial features are extracted from the preprocessed image, especially the feature changes in the eyebrow, eye, and mouth areas, including the curvature of the eyebrows, the opening and closing of the eyes, and the shape of the mouth.

[0038] Facial Expression Classification and Recognition: Based on the extracted feature information, machine learning or deep learning algorithms are used to classify and recognize facial expressions. Specifically, the algorithm is trained and learned from a large number of labeled facial expression images, and can automatically learn the feature representations of different expressions, and achieve accurate recognition of specific expressions in practical applications.

[0039] Frowning detection and judgment: Based on facial expression classification and recognition, special attention is paid to frowning expressions. According to the algorithm output, if it is determined that the user has a frowning expression, the corresponding feedback mechanism is triggered.

[0040] Furthermore, the robot includes a body and an electrically controlled slide rail. A head is fixedly connected to the top of the body, and the head includes a display screen, a camera, a speaker, and a microphone. The camera and speaker are located on the front of the display screen, and the microphone is fixedly connected to the top of the display screen. Legs are embedded in the bottom of the body, and the legs include lifting cylinders and electrically controlled rollers. The bottom end of the lifting cylinder forms an embedded structure with the bottom of the body, and the electrically controlled rollers are fixedly connected to the bottom end of the lifting cylinder. An operating compartment is opened on the front of the body, and an operating keyboard is installed inside the operating compartment. A disinfection component is installed on the inner top wall of the operating compartment. The electrically controlled slide rail is fixedly installed on the front of the body, and a transparent cover plate is mounted on the top of the inner drive of the electrically controlled slide rail. The top of the electrically controlled slide rail is flush with the bottom of the operating compartment, and the electrically controlled slide rail and the transparent cover plate form a sliding structure. Hands are fixedly connected to the outer walls of the left and right sides of the body.

[0041] Furthermore, the disinfection assembly includes a sprayer, a diverter tube, atomizing nozzles, a mounting bracket, and an alcohol storage tank. The sprayer is positioned directly above the operation keyboard, with a diverter tube at the top and atomizing nozzles evenly spaced at the bottom of the diverter tube. The mounting bracket is welded and fixed to the inner wall of the operation chamber, and an alcohol storage tank is engaged between the mounting bracket and the inner wall of the operation chamber. The alcohol storage tank is connected to the sprayer via a flexible hose.

[0042] Furthermore, the hand assembly includes an air supply pump, an air inlet pipe, an air extraction pump, and an air extraction pipe. The air supply pump is fixedly installed on the left outer wall of the machine body, and the air outlet of the air supply pump is fixedly connected to the air inlet pipe. One end of the air inlet pipe passes through the left outer wall of the machine body and extends to the operating compartment. The air extraction pump is fixedly installed on the right outer wall of the machine body, and the air inlet of the air extraction pump is fixedly connected to the air extraction pipe. One end of the air extraction pipe passes through the right outer wall of the machine body and extends to the operating compartment.

[0043] This invention provides an intelligent robot for electricity marketing based on large-scale modeling technology, which has the following beneficial effects:

[0044] 1. This invention not only incorporates an infrared ranging adjustment function, enabling the robot to autonomously adjust its distance to ensure the user's viewing experience is within a suitable range, thus preventing the user from being too close or too far from the robot and affecting their screen viewing experience, but also features a central control analysis module. Through a first analysis unit, a second analysis unit, and a third analysis unit, the module intelligently analyzes the user's head tilt angle and facial expressions while viewing the screen, triggering corresponding feedback mechanisms. This allows the robot to more flexibly adapt to the needs and preferences of different users, providing a more comfortable interactive experience. In particular, through large-scale modeling technology and machine learning algorithms, the robot can more accurately understand the user's intentions and needs, providing a more intelligent response rhythm, thereby improving customer satisfaction and enhancing the brand image and market competitiveness of power companies.

[0045] 2. This invention features a transparent cover to shield the operating chamber when not in use, reducing the risk of dust accumulation and damage to the keyboard. Combined with a disinfection component, the keyboard can be disinfected after each use, significantly reducing the risk of disease transmission and enhancing hygiene and safety. The simultaneous operation of the air supply and extraction pumps ensures air exchange within the operating chamber, removing moisture and the pungent odor of alcohol caused by evaporation. This further prevents moisture from affecting the keyboard's sensitivity and alcohol odor from impacting the user's senses. Attached Figure Description

[0046] Figure 1 This is a logic block diagram of the control system of an intelligent power marketing robot based on large model technology according to the present invention.

[0047] Figure 2 This is a schematic diagram of the first analysis unit of an intelligent power marketing robot based on large model technology according to the present invention;

[0048] Figure 3 This is a schematic diagram of the second analysis unit of an intelligent power marketing robot based on large model technology according to the present invention;

[0049] Figure 4 This is a flowchart illustrating the third analysis unit of an intelligent power marketing robot based on large-scale modeling technology according to the present invention.

[0050] Figure 5 This is a three-dimensional structural diagram of an intelligent power marketing robot based on large-scale model technology according to the present invention.

[0051] Figure 6 This is a schematic diagram of the disinfection component of an intelligent power marketing robot based on large model technology according to the present invention;

[0052] Figure 7 This invention relates to an intelligent robot for electricity marketing based on large-scale modeling technology. Figure 6 Enlarged schematic diagram of the structure at point A in the middle.

[0053] In the diagram: 1. Body; 2. Head; 201. Display screen; 202. Camera; 203. Speaker; 204. Microphone; 3. Legs; 301. Lifting cylinder; 302. Electrically controlled rollers; 4. Operating compartment; 5. Operating keyboard; 6. Disinfection components; 601. Sprayer; 602. Diverter pipe; 603. Atomizing nozzle; 604. Mounting bracket; 605. Alcohol storage tank; 7. Electrically controlled slide rail; 8. Transparent cover; 9. Hands; 901. Air pump; 902. Air inlet pipe; 903. Air extraction pump; 904. Air extraction pipe. Detailed Implementation

[0054] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0055] like Figures 1-4 As shown, an intelligent robot for electricity marketing based on large-scale modeling technology includes a control system mounted on the robot to control its activities. The control system includes:

[0056] Voice prompt module: Used to remind the user to relax and look at the screen via voice when interacting with the robot. In this embodiment, the voice prompt module uses text-to-speech (TTS) technology to convert preset text information into voice output. Specifically, it uses pre-recorded voice messages or text-to-speech technology to generate prompts, which are then played through speaker 203.

[0057] Infrared ranging module: Used to measure the distance between the user and the robot. In this embodiment, the infrared ranging module emits infrared light through an infrared sensor and receives the reflected light, calculating the distance based on the time difference or phase difference.

[0058] Distance Judgment Module: Based on data provided by the infrared ranging module, the distance judgment module determines whether the distance between the user and the robot is within a suitable range. In this embodiment, the distance judgment module adjusts the robot's movement parameters using a PID control algorithm by comparing the measured distance with a preset threshold. Specifically, the measured distance is compared with a preset suitable distance range; if it exceeds the range, the corresponding adjustment module is triggered.

[0059] Distance Adjustment Module: When the distance is not within the appropriate range, the robot can autonomously move to adjust the distance. The distance adjustment module controls the robot's movement mechanism (electrically controlled roller 302) and moves it forward, backward, left, and right according to the result of the distance judgment module to adjust the distance between the robot and the user.

[0060] Image acquisition module: Uses a camera to capture real-time facial expression images of the user;

[0061] The central control analysis module specifically includes

[0062] The first analysis unit analyzes the user's head tilt angle while viewing the screen using a large-scale model algorithm to determine whether the robot's height is appropriate. In this embodiment, the first analysis unit utilizes image processing technology to analyze the user's head tilt angle and compares it with a preset suitable angle. The specific processing flow is as follows:

[0063] Feature point detection: A Haar feature + Adaboost classifier approach (by constructing Haar features and training the Adaboost classifier to detect facial feature points) is used to detect key feature points of the user's head from the user's facial image. Feature points include easily identifiable and stable facial features such as the corners of the eyes, the corners of the mouth, and the tip of the nose. In this embodiment, the captured user's facial image is preprocessed, including denoising, grayscale conversion, and binarization, to facilitate subsequent image analysis.

[0064] Facial contour fitting: After detecting feature points, these points are used to fit the contour of the user's face. Fitting methods include ellipse fitting (assuming that the facial contour is approximately elliptical and fitting the ellipse parameters through methods such as least squares) and polygon fitting (connecting the detected feature points into polygons to approximate the facial contour).

[0065] Head pose estimation: After obtaining facial contours or feature points, geometric algorithms or machine learning algorithms are used to estimate the head pose, i.e., to calculate the head tilt angle. Geometric algorithms: Based on the relative positional relationships of facial feature points, such as the distance and angle from the corners of the eyes and mouth to the tip of the nose, geometric principles are used to calculate the head rotation angle. Alternatively, the fitting results of the facial contour, such as the major and minor axes of an ellipse, are used to estimate the head tilt direction. Machine learning algorithms: A regression model, such as Support Vector Regression (SVR) or Random Forest Regression, is trained. The detected feature points or facial contours are used as input, and the head tilt angle is used as the output. During the testing phase, newly captured facial images are input into the trained model to predict the head tilt angle.

[0066] Tilt angle judgment and adjustment: The calculated tilt angle is compared with the preset suitable tilt angle range. If the tilt angle is not within the suitable range, the height adjustment module is triggered to adjust the height of the robot accordingly.

[0067] Height adjustment module: Based on the judgment result of the first analysis unit, the robot's height is adjusted accordingly by controlling the robot's lifting mechanism (lifting cylinder 301) to maintain a suitable angle with the user's line of sight.

[0068] The second analysis unit analyzes the user's facial expressions using an expression recognition algorithm to determine if the user is squinting or has other facial expressions indicating difficulty seeing the screen. In this embodiment, the facial expression recognition and analysis process of the second analysis unit is as follows:

[0069] Feature extraction: Using image processing techniques, such as edge detection and contour extraction, key facial features of the user are extracted from the preprocessed image. These features include the shape, size, and position of the eyes, as well as the skin texture around the eyes.

[0070] Squinting detection: Based on the extracted feature information, machine learning or deep learning algorithms are used to analyze the eye region, specifically including the following algorithms:

[0071] 1) Geometric feature-based method: By calculating geometric features such as the width and height of the eye area and the degree of eye opening and closing, it can be determined whether the user is squinting;

[0072] 2) Template matching method: A template image of squinting state is predefined, and the extracted eye area is matched with the template. The degree of matching is used to determine whether the user is squinting.

[0073] 3) Deep learning-based methods: Deep learning models such as convolutional neural networks (CNN) are used to train a large number of squinting and non-squinting images, enabling the model to automatically learn the feature representation of squinting and achieve accurate recognition of squinting state in practical applications.

[0074] Judgment and Feedback: Based on the squinting detection results, if the system determines that the user is squinting or has difficulty seeing the screen, a corresponding feedback mechanism is triggered. Specifically, if the user is squinting (i.e., has an expression indicating difficulty seeing the screen), the distance adjustment module is triggered secondarily, allowing the robot to move autonomously. During this movement, the robot continuously monitors changes in the user's facial expressions until it reaches the closest safe observation range. If the problem persists, the screen adjustment module is triggered to resolve the issue by adjusting screen brightness, contrast, font size, etc.

[0075] Screen adjustment module: responsible for adjusting the brightness, contrast, and font size of the screen to solve the problem of users not being able to see the screen clearly and improve readability; in this embodiment, it is specifically adjusted by software that controls the screen display to adjust the brightness, contrast, and font size displayed on the screen.

[0076] The third analysis unit analyzes the user's facial expressions using an expression recognition algorithm to determine if the user is frowning or showing signs of confusion or distress. In this embodiment, the facial expression recognition and analysis process of the third analysis unit is as follows:

[0077] Feature extraction: Using image processing techniques, such as edge detection, contour extraction, and texture analysis, key facial features are extracted from the preprocessed image, especially the feature changes in areas such as eyebrows, eyes, and mouth, including the curvature of eyebrows, the opening and closing of eyes, and the shape of the mouth.

[0078] Facial Expression Classification and Recognition: Based on the extracted feature information, machine learning algorithms (such as Support Vector Machine (SVM), K-Nearest Neighbors (KNN)) or deep learning algorithms (such as Convolutional Neural Network (CNN), Recurrent Neural Network (RNN)) are used to classify and recognize facial expressions. Specifically, the algorithm is trained and learned from a large number of labeled facial expression images, and can automatically learn the feature representations of different expressions, and achieve accurate recognition of specific expressions in practical applications.

[0079] Frowning detection and judgment: Based on facial expression classification and recognition, special attention is paid to confused and embarrassed expressions such as frowning. According to the algorithm output, if it is determined that the user has a frowning expression, the corresponding feedback mechanism is triggered.

[0080] Response Adjustment Module: Based on the frown detection results, the module pauses the current page on the screen for a delay and provides a more patient and detailed explanation or suggestion to respond to the user's emotional state.

[0081] In summary, after the robot is activated, it first prompts the user via a voice prompt module to look at the screen in a relaxed and comfortable state, and then triggers the infrared ranging module to measure the distance. The distance judgment module then determines whether the distance between the user and the robot is appropriate based on the distance measurement result. If the distance is not within the appropriate range, the distance adjustment module is triggered, allowing the robot to autonomously move and adjust the distance to ensure the user is within a suitable viewing distance. After the distance is adjusted appropriately, the image acquisition module uses a camera to acquire real-time images of the user's facial expressions, and the central control analysis module analyzes the images. The first analysis unit uses an algorithm to analyze the user's head tilt angle while viewing the screen to determine whether the robot's height is appropriate, and then adjusts it accordingly through the height adjustment module. Meanwhile, the second analysis unit uses algorithms to determine if the user is squinting (i.e., unable to see the screen clearly). If so, it prioritizes triggering the distance adjustment module to allow the robot to move autonomously. During the movement, it monitors the user's facial expressions in real time until the robot moves to the closest distance within a safe observation range. If the problem is still not resolved, it triggers the screen adjustment module to adjust the screen's brightness, contrast, and font size to solve the problem. In addition, the third analysis unit uses algorithms to determine if the user is frowning (i.e., looking confused or distressed). If so, it triggers the response adjustment module to control the delay of the current screen page and generates extended explanatory information to be overlaid on the paused page, such as providing a more patient and detailed explanation or suggestion to respond to the user's emotional state.

[0082] like Figures 5-7 As shown, the robot includes a body 1 and an electrically controlled slide rail 7. A head 2 is fixedly connected to the top of the body 1, and the head 2 includes a display screen 201, a camera 202, a speaker 203, and a microphone 204. The camera 202 and the speaker 203 are located on the front of the display screen 201, and the microphone 204 is fixedly connected to the top of the display screen 201. Legs 3 are embedded in the bottom of the body 1, and the legs 3 include a lifting cylinder 301 and electrically controlled rollers 302. The bottom end of the lifting cylinder 301 is connected to the body 1. The bottom of the unit forms an embedded structure, and the bottom end of the lifting cylinder 301 is fixedly connected to an electrically controlled roller 302. An operating chamber 4 is opened on the front of the unit 1, and an operating keyboard 5 is installed inside the operating chamber 4. A disinfection component 6 is installed on the inner top wall of the operating chamber 4. The disinfection component 6 includes a sprayer 601, a diverter pipe 602, an atomizing nozzle 603, a mounting bracket 604, and an alcohol storage tank 605. The sprayer 601 is positioned directly above the operating keyboard 5, and a diverter pipe 602 is installed on the top of the sprayer 601. The bottom of the diverter pipe 602, etc. An atomizing nozzle 603 is provided. A mounting bracket 604 is welded and fixed to the inner wall of the operating chamber 4, and an alcohol storage tank 605 is engaged between the mounting bracket 604 and the inner wall of the operating chamber 4. The alcohol storage tank 605 is connected to the sprayer 601 via a hose. An electrically controlled slide rail 7 is fixedly installed on the front of the machine body 1, and a transparent cover plate 8 is mounted on the top of the inner drive of the electrically controlled slide rail 7. The top of the electrically controlled slide rail 7 is flush with the bottom of the operating chamber 4, and the electrically controlled slide rail 7 and the transparent cover plate 8 form a sliding structure. The left and right outer walls of the machine body 1 are fixedly connected to... Hand unit 9 includes an air pump 901, an air inlet pipe 902, an air extraction pump 903, and an air extraction pipe 904. The air pump 901 is fixedly installed on the left outer wall of the body 1, and the air outlet of the air pump 901 is fixedly connected to the air inlet pipe 902. One end of the air inlet pipe 902 passes through the left outer wall of the body 1 and extends to the operating chamber 4. The air extraction pump 903 is fixedly installed on the right outer wall of the body 1, and the air inlet of the air extraction pump 903 is fixedly connected to the air extraction pipe 904. One end of the air extraction pipe 904 passes through the right outer wall of the body 1 and extends to the operating chamber 4.

[0083] The specific operation is as follows: When the user wakes up the robot through the touch screen 201, the transparent cover 8 moves under the control of the electric sliding rail 7, exposing the operation chamber 4, so that the operator can touch the operation keyboard 5 to perform interactive operations. After the power marketing service is completed, the transparent cover 8 returns to its original position, causing the operation chamber 4 to close again. Then the disinfection component 6 is automatically activated, and the sprayer 601 sprays the alcohol in the alcohol storage tank 605 onto the operation keyboard 5 through the atomizing nozzle 603 to sterilize and disinfect it. After the alcohol evaporates, the air pump 901 and the air pump 903 operate simultaneously to allow fresh air to enter the operation chamber 4 and remove the moisture and irritating odor of alcohol caused by alcohol evaporation from the operation chamber 4.

[0084] The embodiments of the present invention are given for illustrative and descriptive purposes only, and are not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of the invention, and to enable those skilled in the art to understand the invention and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. A smart robot for electricity marketing based on large-scale modeling technology, characterized in that, include: The control system includes: The image acquisition module is configured to capture the user's facial expression images in real time via a camera and transmit them to the central control and analysis module. The central control analysis module includes: The first analysis unit is used to analyze the head tilt angle of the user when viewing the screen based on a large model algorithm, and to determine the height deviation between the robot and the user's line of sight. The second analysis unit is used to detect whether the user has a squinting expression using an expression recognition algorithm; The third analysis unit is used to detect whether the user has a frowning expression using an expression recognition algorithm; The distance adjustment module is connected to the first analysis unit and controls the moving mechanism to perform forward or backward or left or right displacement according to the height deviation, so that the robot screen maintains a preset distance range from the user. The height adjustment module drives the lifting mechanism to adjust the vertical height of the robot screen based on the height deviation output by the first analysis unit, so that the robot screen maintains a suitable angle with the user's line of sight. The screen adjustment module dynamically adjusts the screen's brightness, contrast, and font size in response to the squinting detection signal from the second analysis unit. The response adjustment module, after receiving the frown detection signal from the third analysis unit, triggers the delay pause control of the current page on the screen and generates extended explanation information to be overlaid and displayed on the paused page. The second analysis unit detects whether the user has a squinting expression using an expression recognition algorithm, including: Feature extraction: Using image processing technology, key facial features of the user are extracted from the preprocessed image. These features include the shape, size, and position of the eyes, as well as the skin texture around the eyes. Squinting detection: Based on the extracted feature information, machine learning or deep learning algorithms are used to analyze the eye region, specifically including the following algorithms: 1) Geometric feature-based method: By calculating the width, height, and eye opening / closing degree of the eye region, it can be determined whether the user is squinting; 2) Template matching method: A template image of squinting state is predefined, and the extracted eye area is matched with the template. The degree of matching is used to determine whether the user is squinting. 3) Deep learning-based method: The deep learning model is trained on a large number of squinting and non-squinting images, so that the model can automatically learn the feature representation of squinting and achieve accurate recognition of squinting state in practical applications. Judgment and Feedback: Based on the squinting detection results, if it is determined that the user is squinting and cannot see the screen clearly, the corresponding feedback mechanism is triggered; The third analysis unit detects whether the user has a frowning expression using an facial expression recognition algorithm, including: Feature extraction: Using image processing technology, key facial features are extracted from the preprocessed image, which are the feature changes in the eyebrow, eye, and mouth areas, including the curvature of the eyebrows, the opening and closing of the eyes, and the shape of the mouth. Facial Expression Classification and Recognition: Based on the extracted feature information, machine learning or deep learning algorithms are used to classify and recognize facial expressions. Specifically, the algorithm is trained and learned from a large number of labeled facial expression images, and can automatically learn the feature representations of different expressions, and achieve accurate recognition of specific expressions in practical applications. Frowning detection and judgment: Based on facial expression classification and recognition, special attention is paid to frowning expressions. If the algorithm outputs results in a judgment that the user is frowning, the corresponding feedback mechanism is triggered.

2. The intelligent robot for power marketing based on large-scale modeling technology according to claim 1, characterized in that, The control system further includes: Voice prompt module: used to remind the user to relax and look at the screen when the user interacts with the robot; the voice prompt module uses text-to-speech technology to convert preset text information into voice output; Infrared ranging module: used to measure the distance between the user and the robot; the infrared ranging module emits infrared light through an infrared sensor and receives the reflected light, and calculates the distance based on the time difference or phase difference; Distance judgment module: Based on the data provided by the infrared ranging module, it determines whether the distance between the user and the robot is within a suitable range; the distance judgment module adjusts the robot's movement parameters by comparing the measured distance with a preset threshold using a PID control algorithm.

3. The intelligent robot for power marketing based on large-scale modeling technology according to claim 1, characterized in that, The first analysis unit analyzes the user's head tilt angle when viewing the screen based on a large model algorithm, and determines the height deviation between the robot and the user's line of sight, including: Feature point detection: Using Haar features + Adaboost classifier, key feature points of the user's head are detected from the user's facial image. The feature points include facial features such as the corners of the eyes, the corners of the mouth, and the tip of the nose. Facial contour fitting: After detecting feature points, these points are used to fit the contour of the user's face. Fitting methods include ellipse fitting and polygon fitting. Head pose estimation: After obtaining the facial contour or feature points, the head pose is estimated using geometric algorithms or machine learning algorithms, that is, the tilt angle of the head is calculated. The geometric algorithm is based on the relative positional relationship of the facial feature points and uses geometric principles to calculate the rotation angle of the head, or uses the fitting results of the facial contour to estimate the tilt direction of the head. Tilt angle judgment and adjustment: The calculated tilt angle is compared with the preset suitable tilt angle range. If the tilt angle is not within the suitable range, the height adjustment module is triggered to adjust the height of the robot accordingly.

4. The intelligent robot for electricity marketing based on large-scale modeling technology according to claim 3, characterized in that, The machine learning algorithm trains a regression model by taking detected feature points or facial contours as input and head tilt angle as output. During the testing phase, newly captured facial images are input into the trained model to predict the head tilt angle.

5. The intelligent robot for electricity marketing based on large-scale modeling technology according to claim 1, characterized in that, The feedback mechanism is as follows: if the user squints, the distance adjustment module is triggered a second time to make the robot move towards the user autonomously. During the movement, the robot detects changes in the user's facial expression in real time until the robot moves to the closest distance within the safe observation range. If the problem is not resolved, the screen adjustment module is triggered to resolve the problem by adjusting the screen brightness, contrast, and font size.

6. The intelligent robot for electricity marketing based on large-scale modeling technology according to claim 1, characterized in that, The robot includes a body and an electrically controlled slide rail. A head is fixedly connected to the top of the body, and the head includes a display screen, a camera, a speaker, and a microphone. The camera and speaker are located on the front of the display screen, and the microphone is fixedly connected to the top of the display screen. Legs are embedded in the bottom of the body, and the legs include lifting cylinders and electrically controlled rollers. The bottom end of the lifting cylinder is embedded in the bottom of the body, and the electrically controlled rollers are fixedly connected to the bottom end of the lifting cylinder. An operating compartment is opened on the front of the body, and an operating keyboard is installed inside the operating compartment. A disinfection component is installed on the inner top wall of the operating compartment. The electrically controlled slide rail is fixedly installed on the front of the body, and a transparent cover is mounted on the top of the inner drive of the electrically controlled slide rail. The top of the electrically controlled slide rail is flush with the bottom of the operating compartment, and the electrically controlled slide rail and the transparent cover form a sliding structure. Hands are fixedly connected to the outer walls of the left and right sides of the body.

7. The intelligent robot for electricity marketing based on large-scale modeling technology according to claim 6, characterized in that, The disinfection assembly includes a sprayer, a diverter tube, atomizing nozzles, a mounting bracket, and an alcohol storage tank. The sprayer is positioned directly above the control keyboard, with a diverter tube at the top and atomizing nozzles evenly spaced at the bottom of the diverter tube. The mounting bracket is welded and fixed to the inner wall of the control chamber, and an alcohol storage tank is fitted between the mounting bracket and the inner wall of the control chamber. The alcohol storage tank is connected to the sprayer via a flexible hose.

8. The intelligent robot for electricity marketing based on large-scale modeling technology according to claim 7, characterized in that, The hand unit includes an air supply pump, an air inlet pipe, an air extraction pump, and an air extraction pipe. The air supply pump is fixedly installed on the left outer wall of the machine body, and the air outlet of the air supply pump is fixedly connected to the air inlet pipe. One end of the air inlet pipe passes through the left outer wall of the machine body and extends to the operating compartment. The air extraction pump is fixedly installed on the right outer wall of the machine body, and the air inlet of the air extraction pump is fixedly connected to the air extraction pipe. One end of the air extraction pipe passes through the right outer wall of the machine body and extends to the operating compartment.

Images