Desktop lighting robot control system

Abstract

The invention relates to a desktop lighting robot control system. The desktop lighting robot control system comprises a camera, a voice module, an information processing module, a posture sensing module, a motion control module and a spectrum detection and light source control module, a user image is collected through a camera, a voice module captures a user voice signal, an information processing module resolves a user face position and a gesture control instruction, and a posture sensing module obtains user head posture information and visual point distance information; the motion control module generates a control instruction according to the information and drives the robot to adjust the position and the light source; the spectrum detection and light source control module utilizes a deep learning technology to predict and restore environment spectrum information and intelligently adjust a light source so as to adapt to different scene requirements. According to the invention, through attitude calculation and spectrum prediction technologies, intelligent following and dimming are realized, and a comfortable and stable illumination environment is provided.

Description

Technical Field

[0001] The invention relates to the field of robots, and in particular to a desktop lighting robot control system. Background Art

[0002] In recent years, deep learning technology has made significant progress in the field of target detection. It mainly imitates the learning mechanism of the human brain to autonomously learn features from massive data and use them to detect targets. The R-CNN algorithm is the first algorithm among many deep learning algorithms to be applied to the field of target detection. It generates candidate regions through methods such as selective search, and uses convolutional neural networks and support vector machines for feature extraction and classification. Fast R-CNN, as an optimized version of R-CNN, avoids repeated feature calculations. It extracts image feature maps at one time and extracts features for candidate regions, while improving speed and performance. Faster R-CNN further introduces a region proposal network, which predicts the target location and bounding box by sliding a window on the feature map. The generated candidate regions are then used for classification and regression of Fast R-CNN. Thanks to this efficient candidate region generation capability and accurate target recognition capability, Faster R-CNN can divide target detection into two subtasks and achieve end-to-end training and reasoning. Therefore, it has become a benchmark in this field and has been widely used in many fields such as security monitoring and autonomous driving.

[0003] FOC technology, or Field-Oriented Control, is an advanced motor control technology. It uses a variable frequency drive (VFD) to accurately control the direction and size of the magnetic field of a three-phase AC motor to achieve efficient and accurate torque and speed control. FOC technology converts the three-phase motor model into a two-phase model through Clarke and Park transformations, and then controls the d-axis and q-axis currents to orient the motor magnetic field. This technology makes the motor torque smooth, low noise, high efficiency, and has a high-speed dynamic response. FOC technology is widely used in industrial drives, household appliances, electric vehicles, wind power generation and other fields, improving production efficiency and equipment performance. It is one of the important technologies in the field of modern motor control.

[0004] Attitude solution technology is a technology that obtains the attitude information of an object through sensors and processes and calculates it through algorithms. It is widely used in robots, drones, virtual reality and other fields. Attitude solution technology includes two parts: attitude acquisition and attitude processing. Attitude acquisition obtains the attitude information of an object through sensors, cameras and other devices, such as accelerometers, gyroscopes, etc. Attitude processing processes and calculates the acquired attitude information through algorithms to determine the position, direction and attitude of the object. This technology is widely used in various fields. In the field of robotics, it can help robots perform autonomous navigation and task execution. In order to improve accuracy, researchers have proposed a variety of algorithms and models, such as Kalman filtering and particle filtering, to deal with sensor noise and uncertainty. The continuous improvement of these algorithms and models has made attitude solution technology play an increasingly important role in various fields.

[0005] Digital spectral sensors are commonly used to detect the principle of spectral information, which is mainly based on the photoelectric effect and photoelectric detection technology. It integrates multiple photodiodes, each of which is sensitive to light of different wavelengths, forming multiple channels. When light is irradiated onto the sensor, light of different wavelengths is detected by the photodiodes of each channel, generating currents proportional to the light intensity. These currents are converted into digital signals, i.e. spectral information, by the built-in analog-to-digital converter (ADC). By reading these digital signals, the intensity of light of different wavelengths can be obtained, thereby obtaining the spectral information of the object. Digital spectral sensors have the characteristics of high sensitivity, high resolution and high stability, and are widely used in color detection, material analysis, environmental monitoring and other fields. By analyzing the spectral information, the color temperature, irradiance and other characteristics of the environmental spectrum can be understood, providing important information for scientific research, industry, medical and other fields.

[0006] Traditional lighting methods often only focus on the brightness and color of light, while ignoring the impact of light on human health. Modern lighting systems pay more attention to the quality, distribution and changes of light, and provide people with a comfortable and harmless lighting environment through scientific design and reasonable layout. This lighting method can not only meet people's visual needs, but also adjust people's biological clocks, improve emotional states, and improve work efficiency. In various places such as homes, schools, and offices, suitable light sources can play an important role. By choosing suitable lamps, adjusting light brightness and color temperature, etc., a healthy and comfortable lighting environment can be created, which is essential for protecting our physical and mental health. Summary of the invention

[0007] In order to solve the above problems, the present invention proposes a desktop lighting robot control system, which realizes intelligent following and dimming through posture solution and spectrum prediction technology, provides a stable and comfortable lighting environment, and is suitable for office and education scenarios.

[0008] The specific plan is as follows:

[0009] A desktop lighting robot control system, comprising: a camera, a voice module, an information processing module, a posture sensing module, a motion control module, several motors, a spectrum sensor and a spectrum detection and light source control module;

[0010] The camera is used to collect user images;

[0011] The voice module is used to capture the user's voice signal;

[0012] The information processing module is used to receive a user image and a user voice signal, calculate the user's facial position and gesture control instructions based on the user image, and calculate the user's voice based on the user voice signal;

[0013] The posture sensing module is used to collect posture signals and distance signals, and calculate the user's head posture information and visual point distance information based on the posture signals and distance signals;

[0014] The motion control module is used to receive and analyze the user's facial position, gesture control instructions, user voice, user head posture information and visual point distance information, and generate motor control instructions and spectrum detection and light source control module control instructions;

[0015] The plurality of motors are used to receive and analyze motor control instructions to drive the desktop lighting robot;

[0016] The spectral sensor is used to collect spectral information;

[0017] The spectrum detection and light source control module is used to receive and parse the spectrum detection and light source control module control instructions, predict and restore the spectrum information through deep learning technology, obtain ambient light information, and control the light source of the desktop lighting robot based on the ambient light information.

[0018] Furthermore, the desktop lighting robot control system also includes: a lateral motion track, a power module, a lateral motion motor, a camera gimbal, a light source gimbal, a lateral motor synchronous wheel, a synchronous belt idler wheel and an idler wheel fixing plate, a main body pulley and pulley, a sensor mounting bracket, a mounting fixing buckle, a limit switch and a cooling fan;

[0019] The power module includes voltage boost, voltage reduction, power detection and heat dissipation control parts;

[0020] The lateral motion motor includes a stepper motor, a synchronous wheel and a stepper motor drive module with a closed-loop control function;

[0021] The camera gimbal and the light source gimbal have the same structure, including a yaw axis brushless motor, a pitch axis brushless motor and a brushless motor FOC driver board;

[0022] The power module and the lateral motion motor are respectively fixed at two ends of the lateral motion track, and the lateral motion motor pulls the robot body to move through a synchronous belt;

[0023] The lateral motor synchronous wheel is fixed to the output shaft of the lateral motion motor;

[0024] The synchronous belt idler is fixed to the idler fixing plate by bolts;

[0025] The idler wheel fixing plate is fixed to one end of the transverse motion track close to the power module by bolts;

[0026] The robot body is connected to the track through the main pulley and the pulley, the information processing module is located above the pulley plate, and the motion control module is fixed to the front side of the pulley plate through the tenons and screws;

[0027] The cooling fan is mounted on the idler wheel fixing plate by screws and blows toward the information processing module;

[0028] The mounting and fixing buckle is fixed on the transverse motion track by bolts;

[0029] The limit switch is installed on the fixing buckles on both sides;

[0030] The camera gimbal and the light source gimbal are connected to the mounting bracket of the motion control module through the same bracket.

[0031] Furthermore, the generating of the motor control instruction specifically includes:

[0032] The motion control module calculates the motion angles of the two axes corresponding to the camera gimbal and the rotation angle of the lateral stepper motor according to the user's head posture information and the coordinate information of the face in the user's image, and generates a first control instruction; at the same time, it integrates the visual point distance information, calculates the motion angles of the light source gimbal's pitch axis brushless motor and the yaw axis brushless motor, and generates a second control instruction; the first control instruction and the second control instruction are transmitted to each motor through the control bus.

[0033] Furthermore, the prediction and restoration of spectral information by deep learning technology specifically includes:

[0034] S1, mixing the spectral power distribution data of red light, green light, blue light and white light, and using the mixed sample data as a training data set;

[0035] S2, sampling the training data set, obtaining the spectral power distribution data of the digital spectrum sensor on the specified channel as the input of the prediction restoration model, and obtaining the spectral power distribution data of the specified multiple of the specified number of channels of the digital spectrum sensor at equal intervals as the output of the prediction restoration model;

[0036] S3, normalizing the spectral power distribution data used for predicting the input and output of the restoration model to obtain spectral power distribution data scaled to the interval of 0 to 1;

[0037] S4, inputting the spectral power distribution data scaled to the interval of 0 to 1 into the prediction restoration model for training, and obtaining a trained prediction restoration model;

[0038] S5, inputting the test data into the trained prediction restoration model to obtain prediction data, comparing the prediction data with the real data to obtain the final calibration coefficient, and obtaining the final prediction restoration model;

[0039] S6, deploy the final prediction and restoration model in the spectrum detection and light source control module.

[0040] Furthermore, the normalization process is specifically formulated as follows:

[0041]

[0042] Where x represents the raw data obtained by the digital spectral sensor, x max and x min Represent the maximum and minimum values ​​in the predicted data respectively.

[0043] Furthermore, the prediction restoration model is trained, including:

[0044] Use MLP, RBF, LSTM or 1D-CNN algorithms to train the prediction restoration model and reconstruct the spectral power distribution data.

[0045] The present invention adopts the above technical solution and has the following beneficial effects:

[0046] (1) The present invention uses a wireless attitude sensing module and an information processing module to accurately obtain the user's location information, realize intelligent following, and ensure that the lighting is always focused on the user's location;

[0047] (2) The present invention uses spectral data prediction and restoration technology to accurately predict and restore ambient light information, ensure the stability and comfort of the lighting environment, reduce light pollution, and protect the user's eyesight;

[0048] (3) The present invention automatically adjusts the brightness of the light source according to the ambient light information, provides suitable lighting conditions, and improves the user experience, especially in office and education scenarios, ensuring that users work and study under the best light. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 Schematic diagram of a posture sensing module according to an embodiment of the present invention;

[0050] Figure 2This is a schematic diagram of the structure of a desktop lighting robot according to an embodiment of the present invention;

[0051] Figure 3 It is a framework diagram of the interaction between modules in an embodiment of the present invention;

[0052] Figure 4 A schematic diagram of a mixture of 60% white light SPD and 40% red light SPD according to an embodiment of the present invention;

[0053] Figure 5 This is an example diagram of a neural network model according to an embodiment of the present invention;

[0054] Figure 6 This is a diagram showing the prediction effects of different neural network models according to an embodiment of the present invention;

[0055] Figure 7 This is a comparison chart of real data, predicted data and calibrated data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The present invention is further described in detail below in conjunction with embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0057] like Figure 1 , Figure 2 and Figure 3 As shown, the present invention provides a desktop lighting robot control system, comprising:

[0058] Including: camera, voice module, information processing module, posture sensing module, motion control module, several motors, spectrum sensor and spectrum detection and light source control module;

[0059] The camera is used to collect user images;

[0060] The voice module is used to capture user voice signals;

[0061] The information processing module is used to receive a user image and a user voice signal, calculate the user's facial position and gesture control instructions based on the user image, and calculate the user's voice based on the user voice signal;

[0062] The posture sensing module is used to collect posture signals and distance signals, and calculate the user's head posture information and visual point distance information based on the posture signals and distance signals;

[0063] The motion control module is used to receive and analyze the user's face position, gesture control instructions, user voice, user head posture information and visual point distance information, and generate motor control instructions and spectrum detection and light source control module control instructions; each motor receives and analyzes the motor control instructions to drive the desktop lighting robot;

[0064] The plurality of motors are used to receive and analyze motor control instructions to drive the desktop lighting robot;

[0065] The spectral sensor is used to collect spectral information;

[0066] The spectrum detection and light source control module is used to receive and parse the spectrum detection and light source control module control instructions, predict and restore the spectrum information through deep learning technology, obtain ambient light information, and control the light source of the desktop lighting robot based on the ambient light information.

[0067] Specifically, the desktop lighting robot further includes:

[0068] A lateral motion track (18), a power module (15), a lateral motion motor (17), a camera gimbal (6), (7), a light source gimbal (8), (9), an image and posture acquisition module (10), an audio acquisition module (2), a motion control module (3), a lateral motor synchronous wheel (16), a synchronous belt idler wheel (19) and an idler wheel fixing plate (20), a main body pulley (4) and a pulley (5), a sensor mounting bracket (22), a mounting fixing buckle (13), (14), a limit switch (11), (12) and a cooling fan (21);

[0069] The power module includes voltage boost, voltage reduction, power detection and heat dissipation control parts;

[0070] The lateral motion motor includes a stepper motor, a synchronous wheel and a stepper motor drive module with a closed-loop control function;

[0071] The camera gimbal and the light source gimbal have the same structure, including a yaw axis brushless motor, a pitch axis brushless motor and a brushless motor FOC driver board;

[0072] The power module and the lateral motion motor are respectively fixed at two ends of the lateral motion track, and the lateral motion motor pulls the robot body to move through a synchronous belt;

[0073] The lateral motor synchronous wheel is fixed to the output shaft of the lateral motion motor;

[0074] The synchronous belt idler is fixed to the idler fixing plate by bolts;

[0075] The idler wheel fixing plate is fixed to one end of the transverse motion track close to the power module by bolts;

[0076] The robot body is connected to the track through the main pulley and the pulley, the information processing module is located above the pulley plate, and the motion control module is fixed to the front side of the pulley plate through the tenons and screws;

[0077] The cooling fan is mounted on the idler wheel fixing plate by screws and blows toward the information processing module;

[0078] The mounting and fixing buckle is fixed on the transverse motion track by bolts;

[0079] The limit switch is installed on the fixing buckles on both sides;

[0080] The camera gimbal and the light source gimbal are connected to the mounting bracket of the motion control module through the same bracket;

[0081] The motion control module includes a main controller and a wireless transmission module. The motion control module parses the position information and control instructions obtained by the posture solving technology, generates each motor control instruction, and transmits it to each motor through the main controller and the wireless transmission module.

[0082] Specifically, the generating of the motor control instruction includes:

[0083] The motion control module calculates the motion angles of the two axes corresponding to the camera gimbal and the rotation angle of the lateral stepping motor according to the user's head posture information and the coordinate information of the face in the user's image, and generates a first control instruction; at the same time, the visual point distance information is integrated to calculate the motion angles of the light source gimbal's pitch axis brushless motor and the yaw axis brushless motor, and generate a second control instruction; the first control instruction and the second control instruction are transmitted to the gimbal motor 1, the gimbal motor 2, the gimbal motor 3, the gimbal motor 4, the lateral motor and the light source control module (spectral detection and light source control module) through the control bus.

[0084] Specifically, the method of realizing intelligent following and intelligent dimming of the desktop lighting robot is as follows:

[0085] First, the user wears the wearable posture sensing module on the head; then, the camera collects images, the voice module captures the user's voice signal, and transmits the image signal and voice signal to the information processing module to solve the user's facial position information and the user's voice and gesture control instructions; next, the posture sensing module collects posture signals and distance signals, and solves the user's head posture information and visual point distance information; then, the information processing module transmits the user's facial position information and the user's voice and gesture control instructions to the motion control module through the communication bus, and the posture sensing module transmits the user's head posture information and visual point distance information to the motion control module through the wireless communication module; on this basis, the motion control module parses the above information and instructions, generates each motor control instruction, and transmits it to each motor for execution through the control bus; at the same time, the spectral detection and light source control module collects spectral signals through a digital spectral sensor, predicts and restores the spectral information through deep learning technology, accurately calculates the ambient light information, and controls the light source to ensure that the fill light process is gentle and imperceptible, and the lighting environment is stable.

[0086] Specifically, people will move during the use of the desktop lighting robot. In order to realize the tracking function and intelligent fill light function, the tracking function of the camera pan and light source pan is realized through the motion control of the motor. The motion control module calculates the motion angles of the two axes corresponding to the camera pan and the rotation angle of the lateral stepper motor according to the user's head posture information (i.e., the angle information of the yaw and pitch axes) and the coordinate information of the face in the collected image, and generates a control instruction. At the same time, the aforementioned visual point distance information is integrated to calculate the motion angles of the pitch motor and the yaw motor of the light source pan, and also generate a control instruction. After the instruction is generated, it is transmitted to each motor for execution through the control bus. The motion control module calculates the motion angles of the two axes corresponding to the camera pan and the rotation angle of the lateral stepper motor according to the user's head posture information (i.e., the angle information of the yaw and pitch axes) and the coordinate information of the face in the collected image, and generates a control instruction. At the same time, the aforementioned visual point distance information is integrated to calculate the motion angles of the pitch motor and the yaw motor of the light source pan, and also generate a control instruction. After the instruction is generated, it is transmitted to each motor for execution through the control bus.

[0087] Specifically, the method of predicting and restoring the collected spectral signal by using the spectral data prediction and restoration technology specifically includes:

[0088] S1, mixing the spectral power distribution data of red light, green light, blue light and white light, and using the mixed sample data as a training data set;

[0089] S2, sampling the training data set, obtaining the spectral power distribution data of the digital spectrum sensor on the specified channel as the input of the prediction restoration model, and obtaining the spectral power distribution data of the specified multiple of the specified number of channels of the digital spectrum sensor at equal intervals as the output of the prediction restoration model;

[0090] S3, normalizing the spectral power distribution data used for predicting the input and output of the restoration model to obtain spectral power distribution data scaled to the interval of 0 to 1;

[0091] S4, inputting the spectral power distribution data scaled to the interval of 0 to 1 into the prediction restoration model for training, and obtaining a trained prediction restoration model;

[0092] S5, inputting the test data into the trained prediction restoration model to obtain prediction data, comparing the prediction data with the real data to obtain the final calibration coefficient, and obtaining the final prediction restoration model;

[0093] S6, deploy the final prediction and restoration model in the spectrum detection and light source control module.

[0094] Specifically, the normalization process is performed as follows:

[0095]

[0096] Where x represents the raw data obtained by the digital spectral sensor, x max and x min Represent the maximum and minimum values ​​in the predicted data respectively.

[0097] Specifically, the prediction restoration model is trained, including:

[0098] Use MLP, RBF, LSTM or 1D-CNN algorithms to train the prediction restoration model and reconstruct the spectral power distribution data.

[0099] Specifically, the implementation method of running the final prediction and restoration model on the spectrum detection and light source control module or other embedded devices is as follows:

[0100] First, a spectrometer was used to measure the SPD of red light, green light, blue light, and white light at room temperature as initial data.

[0101] Next, any two or three different colors of light are combined in 2% increments, such as 2% white light combined with 98% red light, 4% white light combined with 96% red light, 2% white light combined with 2% red light combined with 96% green light, 4% white light combined with 2% red light combined with 94% green light, and so on. Figure 4 This is a diagram of a combination of 60% white light and 40% red light; finally, a training data set of 5006 samples was obtained in total.

[0102] Next, each sample is sampled to obtain the SPD at 8 wavelengths corresponding to the response of the spectral sensor as the input of Jupiter training data; then, 81 points of SPD data are obtained at equal intervals of 5nm as the output of model training; on this basis, the obtained SPD is normalized. The present invention uses linear normalization, which can scale the data to between 0 and 1.

[0103] Afterwards, MLP, RBF, LSTM, 1D-CNN and other algorithms were used for model training and SPD reconstruction. Examples of neural network models are as follows: Figure 5 As shown, according to Figure 6 The reconstruction effects of each model shown are shown by selecting the most suitable neural network; the input of the neural network model is the value of the channel wavelength of the visible light range of the corresponding spectral sensor extracted from the actual measured SPD, Figure 6 (a) represents SPD reconstruction using MLP; (b) represents SPD reconstruction using RBF; (c) represents SPD reconstruction using LSTM; (d) represents SPD reconstruction using 1D-CNN.

[0104] Finally, the original data obtained by the spectral sensor is used for prediction. The predicted 81 data are compared with the real data, and the ratio is calculated as the calibration coefficient. The average of the calibration coefficients calculated from 50 sets of data measured in different environments is taken as the final calibration coefficient. The comparison of the real data, predicted data and calibrated predicted data is shown in Figure 2. Figure 7 shown.

[0105] Although the present invention has been specifically shown and described in conjunction with the preferred embodiments, it should be understood by those skilled in the art that various changes may be made to the present invention in form and details without departing from the spirit and scope of the present invention as defined by the appended claims, all of which are within the scope of protection of the present invention.

Claims

1. A desktop lighting robot control system, characterized in that: include: Camera, voice module, information processing module, posture sensing module, motion control module, several motors, spectrum sensor and spectrum detection and light source control module; The camera is used to collect user images; The voice module is used to capture the user's voice signal; The information processing module is used to receive a user image and a user voice signal, calculate the user's facial position and gesture control instructions based on the user image, and calculate the user's voice based on the user voice signal; The posture sensing module is used to collect posture signals and distance signals, and calculate the user's head posture information and visual point distance information based on the posture signals and distance signals; The motion control module is used to receive and analyze the user's facial position, gesture control instructions, user voice, user head posture information and visual point distance information, and generate motor control instructions and spectrum detection and light source control module control instructions; The plurality of motors are used to receive and analyze motor control instructions to drive the desktop lighting robot; The spectral sensor is used to collect spectral information; The spectrum detection and light source control module is used to receive and parse the spectrum detection and light source control module control instructions, predict and restore the spectrum information through deep learning technology, obtain ambient light information, and control the light source of the desktop lighting robot based on the ambient light information.

2. The desktop lighting robot control system according to claim 1, characterized in that: Also includes: Transverse motion track, power module, transverse motion motor, camera gimbal, light source gimbal, transverse motor synchronous wheel, synchronous belt idler wheel and idler wheel fixing plate, main body pulley and pulley, sensor mounting bracket, mounting fixing buckle, limit switch and cooling fan; The power module includes boost, buck, power detection and heat dissipation control parts; The lateral motion motor includes a stepper motor, a synchronous wheel and a stepper motor drive module with a closed-loop control function; The camera gimbal and the light source gimbal have the same structure, including a yaw axis brushless motor, a pitch axis brushless motor and a brushless motor FOC driver board; The power module and the lateral motion motor are respectively fixed at two ends of the lateral motion track, and the lateral motion motor pulls the robot body to move through a synchronous belt; The lateral motor synchronous wheel is fixed to the output shaft of the lateral motion motor; The synchronous belt idler is fixed to the idler fixing plate by bolts; The idler wheel fixing plate is fixed to one end of the transverse motion track close to the power module by bolts; The robot body is connected to the track through the main pulley and the pulley, the information processing module is located above the pulley plate, and the motion control module is fixed to the front side of the pulley plate through the tenons and screws; The cooling fan is mounted on the idler wheel fixing plate by screws and blows toward the information processing module; The mounting and fixing buckle is fixed on the transverse motion track by bolts; The limit switch is installed on the fixing buckles on both sides; The camera gimbal and the light source gimbal are connected to the mounting bracket of the motion control module through the same bracket.

3. The desktop lighting robot control system according to claim 2, characterized in that: The generating of the motor control instruction specifically includes: The motion control module calculates the motion angles of the two axes corresponding to the camera gimbal and the rotation angle of the lateral stepper motor according to the user's head posture information and the coordinate information of the face in the user's image, and generates a first control instruction; at the same time, it integrates the visual point distance information, calculates the motion angles of the light source gimbal's pitch axis brushless motor and the yaw axis brushless motor, and generates a second control instruction; the first control instruction and the second control instruction are transmitted to each motor through the control bus.

4. The desktop lighting robot control system according to claim 1, characterized in that: The prediction and restoration of spectral information by deep learning technology specifically includes: S1, mixing the spectral power distribution data of red light, green light, blue light and white light, and using the mixed sample data as a training data set; S2, sampling the training data set, obtaining the spectral power distribution data of the digital spectrum sensor on the specified channel as the input of the prediction restoration model, and obtaining the spectral power distribution data of the specified multiple of the specified number of channels of the digital spectrum sensor at equal intervals as the output of the prediction restoration model; S3, normalizing the spectral power distribution data used for predicting the input and output of the restoration model to obtain spectral power distribution data scaled to the interval of 0 to 1; S4, inputting the spectral power distribution data scaled to the interval of 0 to 1 into the prediction restoration model for training, and obtaining a trained prediction restoration model; S5, inputting the test data into the trained prediction restoration model to obtain prediction data, comparing the prediction data with the real data to obtain the final calibration coefficient, and obtaining the final prediction restoration model; S6, deploy the final prediction and restoration model in the spectrum detection and light source control module.

5. The desktop lighting robot control system according to claim 4, characterized in that: The specific formula for the normalization process is as follows: Where x represents the raw data obtained by the digital spectral sensor, x max and x min Represent the maximum and minimum values ​​in the predicted data respectively.

6. The desktop lighting robot control system according to claim 4, characterized in that: Training the prediction and restoration model includes: Use MLP, RBF, LSTM or 1D-CNN algorithms to train the prediction restoration model and reconstruct the spectral power distribution data.

Images