Optical communication driven robot base station recharging method and system

Abstract

The invention discloses an optical communication driven robot base station recharging method and system, particularly relates to the technical field of optical communication, and is used for solving the problem of unreliable visible light communication signal identification caused by automatic exposure adjustment and environment interference in a complex light environment in the prior art. An image sequence containing visible light communication signals and an automatic exposure parameter sequence are collected, a brightness value sequence of a candidate light source area is established, modulation signals are demodulated, a first candidate set is obtained through screening on the basis of inter-symbol interference characteristics, and a second candidate set is obtained through analyzing motion trail smoothness characteristics. And finally, determining a real base station communication light source by evaluating the inverse correlation between the modulation signal and the automatic exposure parameter sequence, and calculating the course adjustment amount of the robot according to the position change to realize accurate recharging.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of optical communication technology, more particularly, the present application relates to a robot base station recharging method and system driven by optical communication. BACKGROUND

[0002] Visible light communication technology can be applied in mobile robot autonomous recharging system due to its potential in positioning and communication. The base station transmits modulated visible light signal as a guide beacon, and the robot captures and analyzes the signal using the optical sensor carried on the robot to realize autonomous approach and docking to the base station. The prior art solution usually relies on light source region recognition and light intensity signal demodulation of the image data collected by the sensor to obtain the guide information. However, the actual operating environment of the robot is complex, the environmental lighting conditions are uncertain, and the image sensor of the robot itself is automatically adjusted to adapt to the overall picture. These factors together constitute a challenge to the stability of the visible light communication link.

[0003] However, the prior art has defects in actual application. The non-uniformity and dynamic change of the environmental background light interact with the automatic adjustment mechanism of the robot image sensor to ensure regular imaging, which seriously interferes with the stable recognition of the communication light source and the accurate extraction of the signal characteristics, resulting in a decrease in the decoding reliability of the robot to the guide signal, and further affecting the success rate and robustness of the entire recharging process. The coupling interference of the complex light environment and the imaging system itself on the visible light communication signal cannot be overcome. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the present application provides a robot base station recharging method and system driven by optical communication to solve the problems raised in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions: The robot base station recharging method driven by optical communication comprises: S1, collecting an image sequence containing a visible light communication signal transmitted by the base station through the optical sensor of the robot, and recording an automatic exposure parameter sequence; S2, preprocessing the image sequence, identifying a candidate communication light source region with brightness higher than a dynamic threshold, and establishing a brightness value sequence of the candidate communication light source region; S3, demodulating the modulated signal from the brightness value sequence, obtaining an interference regularity index by analyzing the symbol interference characteristics of the modulated signal, and screening a first candidate set with the interference regularity index greater than a first threshold; S4, analyzing the smoothness of the motion trajectory of the first candidate set, calculating a trajectory smoothness index by evaluating the continuity of the trajectory point distribution, and screening a second candidate set with the trajectory smoothness index greater than a second threshold; S5. For the second candidate set, analyze the inverse correlation between the modulation signal of the candidate communication light source region and the automatic exposure parameter sequence, and calculate the inverse correlation index. S6. Select the candidate communication light source region with the highest inverse correlation index from the second candidate set as the real base station communication light source. Calculate the robot's heading adjustment amount based on the position change of the real base station communication light source and control the robot to move to complete the recharging.

[0006] Furthermore, the robot's optical sensors acquire image sequences containing visible light communication signals emitted by the base station, and record automatic exposure parameter sequences, including: The optical sensor acquires image sequences containing visible light communication signals emitted by the base station and records automatic exposure parameter sequences; The automatic exposure parameter sequence is generated by the optical sensor during the acquisition of the image sequence and includes exposure time parameters and gain parameters.

[0007] Furthermore, the image sequence is preprocessed to identify candidate communication light source regions with brightness exceeding a dynamic threshold, and a brightness value sequence for these candidate communication light source regions is established, including: Image sequences are preprocessed to eliminate environmental noise interference; A dynamic threshold is calculated based on the preprocessed image sequence, and the dynamic threshold is adaptively determined according to the overall brightness characteristics of the image sequence. Identify candidate communication light source regions in an image sequence whose brightness exceeds a dynamic threshold; A brightness value sequence of candidate communication light source regions is established. The brightness value sequence is formed by extracting the brightness values ​​of the candidate communication light source regions in the image sequence and arranging them in chronological order.

[0008] Furthermore, the modulated signal is demodulated from the brightness value sequence, and the interference regularity index is obtained by analyzing the inter-symbol interference characteristics of the modulated signal. A first candidate set with interference regularity indices greater than a first threshold is then selected, including: Demodulate the modulated signal from the brightness value sequence; Analyze the inter-symbol interference characteristics of the modulated signal and calculate the interference regularity index by comparing the amplitude changes of adjacent symbols; Compare the interference regularity index with the first threshold; A first candidate set is selected based on the interference regularity index being greater than a first threshold. The first candidate set includes candidate communication light source regions whose interference regularity index is greater than the first threshold.

[0009] Furthermore, the interference regularity index obtained by analyzing the inter-symbol interference characteristics of the modulated signal includes: extracting the symbol amplitude sequence from the modulated signal, calculating the amplitude change value of adjacent symbols, and calculating the interference regularity index by analyzing the statistical distribution characteristics of the amplitude change value.

[0010] Furthermore, the smoothness characteristics of the motion trajectory are analyzed for the first candidate set. A trajectory smoothness index is calculated by evaluating the continuity of the trajectory point distribution. A second candidate set with a trajectory smoothness index greater than a second threshold is then selected, including: Motion trajectories are constructed based on the position coordinates of candidate communication light source regions in consecutive image frames from the first candidate set. Extract the movement vectors between adjacent trajectory points in the motion trajectory and analyze the consistency of the changes in the direction of the movement vectors; The trajectory smoothness index is calculated by evaluating the stability of changes in the direction of the moving vector. The trajectory smoothness index is compared with a second threshold; candidate communication light source regions with trajectory smoothness indices greater than the second threshold are selected to form a second candidate set.

[0011] Furthermore, for the second candidate set, the inverse correlation between the modulation signal of the candidate communication light source region and the automatic exposure parameter sequence is analyzed, and the inverse correlation index is calculated, including: Extract the modulation signal of the candidate communication light source region from the second candidate set; Obtain the automatic exposure parameter sequence; Analyze the correlation between the changing trends of the modulation signal and the automatic exposure parameter sequence; The inverse correlation index is calculated by evaluating the degree of inverse relationship between the modulation signal and the changing trends of the automatic exposure parameter sequence; The modulation signal comes from the demodulation result of the brightness value sequence of the candidate communication light source region, and the automatic exposure parameter sequence comes from the exposure parameter record when the optical sensor acquires the image sequence.

[0012] Furthermore, the inverse correlation index is calculated by evaluating the degree of opposition between the changing trends of the modulation signal and the automatic exposure parameter sequence. This includes: extracting time series data of the modulation signal and the automatic exposure parameter sequence, analyzing the opposition between their changing trends, and obtaining the inverse correlation index by statistically analyzing the frequency of opposite changes.

[0013] Furthermore, the candidate communication light source region with the highest inverse correlation index is selected from the second candidate set as the real base station communication light source. The robot's heading adjustment is calculated based on the positional change of the real base station communication light source, and the robot is controlled to move and complete the recharging process, including: Compare the inverse correlation indices of each candidate communication light source region in the second candidate set; The candidate communication light source region with the highest inverse correlation index was selected as the actual base station communication light source. Extract the position coordinates of the real base station communication light source in consecutive image frames; The robot's heading adjustment is calculated based on the trend of position coordinate changes in consecutive image frames; Control commands are generated based on the robot's heading adjustment and sent to the robot's motion control system to complete the recharge operation.

[0014] On the other hand, the present invention provides an optical communication-driven robot base station recharging system, comprising: The sequence acquisition module is used to acquire image sequences containing visible light communication signals emitted by the base station through the robot's optical sensors, and to record automatic exposure parameter sequences; The sequence establishment module is used to preprocess the image sequence, identify candidate communication light source regions with brightness higher than the dynamic threshold, and establish a brightness value sequence of the candidate communication light source regions. The first candidate module is used to demodulate the modulated signal from the brightness value sequence, obtain the interference regularity index by analyzing the inter-symbol interference characteristics of the modulated signal, and screen the first candidate set whose interference regularity index is greater than the first threshold. The second candidate module is used to analyze the smoothness characteristics of the motion trajectory of the first candidate set, calculate the trajectory smoothness index by evaluating the continuity of the trajectory point distribution, and screen the second candidate set whose trajectory smoothness index is greater than the second threshold. The index calculation module is used to analyze the inverse correlation between the modulation signal of the candidate communication light source region and the automatic exposure parameter sequence in the second candidate set, and to calculate the inverse correlation index. The recharge control module is used to select the candidate communication light source region with the highest inverse correlation index from the second candidate set as the real base station communication light source, calculate the robot's heading adjustment amount based on the position change of the real base station communication light source, and control the robot to move to complete the recharge.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. By constructing a multi-level screening mechanism, accurate identification and analysis of visible light communication signals from base stations were achieved in complex lighting environments. First, the signal regularity was assessed by analyzing the inter-symbol interference characteristics of the modulation signal, effectively distinguishing between regular interference caused by automatic exposure adjustment and random environmental noise, significantly improving the reliability of signal feature extraction. Second, by evaluating the smoothness characteristics of the candidate light source's motion trajectory, false signals caused by sudden changes in ambient light or brief obstructions were eliminated, ensuring the continuity of the tracking target. By analyzing the inverse correlation between the modulation signal and the automatic exposure parameter sequence, coupling interference caused by automatic sensor adjustment could be accurately identified, thus achieving the identification of real base station signals at the signal level.

[0016] 2. By co-analyzing visible light communication signals and sensor parameters, stable recharge guidance in dynamic environments was achieved. Utilizing the inherent correlation between the automatic exposure parameter sequence and the modulation signal, an anti-interference signal analysis mechanism was constructed, enabling the robot to maintain accurate tracking of base station signals under strong ambient light changes. Through feature complementarity in the multi-level screening process, the system can not only effectively suppress environmental interference but also adapt to the dynamic adjustment characteristics of the sensor itself. Thus, while ensuring conventional imaging quality, the stability of the communication link is maintained, significantly improving the robot's recharge success rate and system robustness in real complex environments. Attached Figure Description

[0017] Figure 1 This is a flowchart of the optical communication-driven robot base station recharging method of the present invention; Figure 2 This is a schematic diagram of the structure of the optical communication-driven robot base station recharging system of the present invention. Detailed Implementation

[0018] 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.

[0019] Example 1: Figure 1 The present invention provides a method for recharging a robot base station driven by optical communication, comprising: S1. Acquire image sequences containing visible light communication signals emitted by the base station using the robot's optical sensors, and record the automatic exposure parameter sequence; S2. Preprocess the image sequence to identify candidate communication light source regions with brightness higher than the dynamic threshold, and establish a brightness value sequence of the candidate communication light source regions. S3. Demodulate the modulation signal from the brightness value sequence, obtain the interference regularity index by analyzing the inter-symbol interference characteristics of the modulation signal, and screen the first candidate set whose interference regularity index is greater than the first threshold. S4. Analyze the smoothness characteristics of the motion trajectory of the first candidate set, calculate the trajectory smoothness index by evaluating the continuity of the trajectory point distribution, and screen the second candidate set whose trajectory smoothness index is greater than the second threshold. S5. For the second candidate set, analyze the inverse correlation between the modulation signal of the candidate communication light source region and the automatic exposure parameter sequence, and calculate the inverse correlation index. S6. Select the candidate communication light source region with the highest inverse correlation index from the second candidate set as the real base station communication light source. Calculate the robot's heading adjustment amount based on the position change of the real base station communication light source and control the robot to move to complete the recharging.

[0020] S1. Acquire image sequences containing visible light communication signals emitted by the base station using the robot's optical sensors, and record the automatic exposure parameter sequence. Specifically, this is implemented as follows: An optical sensor acquires an image sequence containing visible light communication signals emitted by a base station. The optical sensor uses a complementary metal-oxide-semiconductor (CMOS) image sensor or a charge-coupled device (CCD) image sensor mounted on the robot. Images are acquired in consecutive frames at a frame rate of 30 frames per second. Each frame in the image sequence contains a modulated light signal emitted by the base station via visible light communication. The modulated light signal uses on / off keyed modulation. The base station's light source changes brightness at a preset frequency, for example, 1000 Hz. During acquisition, the optical sensor ensures its field of view is aligned with the base station to capture the complete modulated light signal. The image sequence is stored in RGB or grayscale format with a resolution of 1920 x 1080 pixels. During acquisition, the optical sensor automatically adjusts its position according to ambient lighting conditions. Imaging parameters are adjusted to ensure image quality and avoid overexposure or underexposure. The image sequence is stored in the robot's memory for subsequent processing steps. The acquisition of the optical sensor is triggered by the robot's control unit, and the acquisition duration is set according to the needs of the robot's recharging task, for example, the acquisition duration is 10 seconds. Each frame of the image sequence is captured by the lens and filter assembly of the optical sensor to reduce ambient stray light interference and ensure the clarity of visible light communication signals. During the acquisition process, the optical sensor may be affected by changes in ambient light or motion blur, but through an automatic adjustment mechanism, the stability of the image sequence can be maintained. For example, in a strong light environment, the optical sensor reduces overexposure by lowering the exposure time parameter, and in a low light environment, it enhances signal detection by increasing the gain parameter.

[0021] The automatic exposure parameter sequence is generated by the optical sensor during the image acquisition process. This sequence includes an exposure time parameter and a gain parameter. The exposure time parameter represents the exposure duration for each frame of the image from the optical sensor, measured in milliseconds (ms). For example, the exposure time parameter ranges from 1 ms to 100 ms. The gain parameter represents the signal amplification factor of the optical sensor, measured in decibels (dB) or ISO values. For example, the gain parameter ranges from ISO 100 to ISO 100. The generation of the automatic exposure parameter sequence is achieved through the internal automatic exposure control circuit of the optical sensor. The automatic exposure control circuit dynamically adjusts the exposure time parameter and gain parameter according to the overall brightness average value of the image sequence to maintain the image brightness within the target range. For example, the target brightness is set to the median gray value of the image, 128. The automatic exposure parameter sequence corresponds one-to-one with each frame of the image sequence and is recorded in the robot's storage unit. The specific values ​​of the exposure time parameter and gain parameter are obtained by reading the register of the optical sensor or the application programming interface. The automatic exposure parameter sequence is used for subsequent analysis steps and is used synchronously with the image sequence. The generation process of the automatic exposure parameter sequence is fully automated and requires no manual intervention. It is controlled by the robot's software driver, which calls the application programming interface of the optical sensor to read and store parameters in real time.

[0022] During the generation of the automatic exposure parameter sequence, the automatic exposure control circuit first calculates the average brightness of each frame. The average brightness is obtained by summing the grayscale values ​​of the image pixels and dividing by the total number of pixels. For example, for an image with a resolution of 1920 x 1080 pixels, the total number of pixels is 2,073,600. The calculation of the average brightness is completed in the image processing unit of the optical sensor. The automatic exposure control circuit compares the average brightness with a preset threshold. The preset threshold is set based on historical data or experimental calibration. For example, by testing under multiple lighting conditions, using a standard light source to simulate different conditions, measuring the average brightness of the image, and statistically analyzing the optimal range, the preset threshold is set to a grayscale value of 100. If the average brightness is lower than the preset threshold, the automatic exposure control circuit increases the exposure time parameter or gain parameter, for example, by increasing the exposure time parameter by 5 milliseconds and increasing the gain parameter by ISO 100. If the average brightness is higher than the preset threshold, the automatic exposure control circuit decreases the exposure time parameter or gain parameter, for example, by decreasing the exposure time parameter by 5 milliseconds and decreasing the gain parameter by ISO 100. The adjustment step size is set according to the performance of the optical sensor to ensure a smooth transition and avoid parameter abrupt changes.

[0023] The recording of the automatic exposure parameter sequence is accomplished by the robot's data acquisition module. This module records the exposure time and gain parameters at the same frame rate as the image sequence, forming time-series data. For example, each frame corresponds to a data point containing both the exposure time and gain parameter values. The automatic exposure parameter sequence is stored in array form, with each element containing two fields representing the exposure time and gain parameters respectively. The array index is aligned with the frame index of the image sequence to ensure data consistency. The automatic exposure parameter sequence is stored in binary or text file format for easy subsequent reading and processing. During acquisition, if the ambient light changes drastically, the automatic exposure control circuit will accelerate the adjustment frequency, for example, adjusting every frame, to quickly adapt to the conditions. The generation of the automatic exposure parameter sequence also considers the physical limitations of the optical sensor; for example, the maximum exposure time parameter is 100 milliseconds, the minimum exposure time parameter is 1 millisecond, the maximum gain parameter is ISO 6400, and the minimum gain parameter is ISO 100. These limits are set through hardware firmware to prevent parameters from going out of bounds. The parameter adjustment is based on real-time feedback. For example, the automatic exposure control circuit continuously monitors the variance of the brightness distribution. If the variance exceeds the variance threshold, for example, the variance threshold is set to 500, it indicates that the illumination is uneven. In this case, the gain parameter is adjusted first to reduce motion blur. The variance threshold is obtained by analyzing the brightness distribution variance of historical image data. For example, the variance value is recorded in multiple recharge tasks, and a fixed threshold is set to improve robustness.

[0024] When optical sensors acquire image sequences, the adjustment of the automatic exposure parameter sequence also involves noise suppression and stability assurance. For example, in low-light environments, increasing the gain parameter can introduce noise. The automatic exposure control circuit smooths parameter changes through a filtering algorithm, which uses a moving average method, such as using the average brightness of the most recent 5 frames as the adjustment basis to ensure parameter stability. Synchronization between the automatic exposure parameter sequence and the image sequence is achieved through timestamps generated by the robot's system clock with millisecond-level accuracy, ensuring data alignment in subsequent inverse correlation analysis. In robot recharging scenarios, the generation of the automatic exposure parameter sequence is designed to adapt to complex lighting environments, such as changes in indoor lighting or... Under outdoor sunlight interference, the reliability of visible light communication signals in the image sequence is maintained by dynamic adjustment. The specific values ​​of the automatic exposure parameter sequence are used for inverse correlation analysis in subsequent steps. For example, in S5, the changing trends of the modulation signal and the automatic exposure parameter sequence are compared. The automatic exposure parameter sequence is obtained by using standard image processing libraries, such as using OpenCV library functions to read sensor parameters. The setting of the brightness average preset threshold and variance threshold also takes into account the robot's motion state. For example, when moving quickly, the blurring is reduced by increasing the adjustment frequency. The threshold setting is based on historical operation data. For example, the lighting conditions are recorded in multiple recharging tasks, and the threshold is optimized to improve robustness.

[0025] S2. Preprocess the image sequence to identify candidate communication light source regions with brightness exceeding the dynamic threshold, and establish a brightness value sequence for the candidate communication light source regions. Specifically, this is implemented as follows: The image sequence is preprocessed to eliminate environmental noise interference. Gaussian filtering is used, where a 3×3 pixel kernel slides across each frame of the image sequence with a standard deviation of 1.5 to smooth the image and reduce random noise. Median filtering, with a 5×5 pixel window, is also included to remove salt-and-pepper noise. The preprocessing is performed in the robot's image processing unit. The input is the image sequence acquired in step S1, and the output is the preprocessed image sequence. The preprocessed image sequence preserves visible light communication signals while minimizing environmental interference. For example, in indoor environments, preprocessing effectively suppresses noise caused by light flicker, and in outdoor environments, it reduces the impact of sunlight reflection. The preprocessing is implemented by calling the Gaussian filtering and median filtering functions in the OpenCV library. Preprocessing parameters are set according to the image sequence's resolution and noise level. For example, for a 1920×1080 pixel image, the kernel size and standard deviation are determined experimentally to balance denoising and signal preservation.

[0026] A dynamic threshold is calculated based on the preprocessed image sequence. The dynamic threshold is adaptively determined according to the overall brightness characteristics of the image sequence. These overall brightness characteristics include the average brightness and brightness variance of each frame. The average brightness is obtained by dividing the sum of the grayscale values ​​of all pixels in each frame by the total number of pixels. For example, for a 1920×1080 pixel image, the total number of pixels is 2,073,600. The brightness variance is obtained by dividing the sum of the squares of the differences between the grayscale values ​​of each pixel and the average brightness by the total number of pixels. The dynamic threshold is calculated using a proportional adjustment method based on the average brightness; for example, the dynamic threshold is set to the average brightness multiplied by a certain ratio. The coefficient is added with an offset. The scaling factor is adaptively adjusted based on the range of brightness changes in the image sequence. For example, the scaling factor ranges from 1.1 to 1.5, and the offset ranges from 10 to 50. The specific values ​​of the scaling factor and offset are set according to the brightness distribution statistics of historical image data. For example, brightness characteristics are recorded in multiple recharge tasks, and parameters are optimized to improve threshold adaptability. The dynamic threshold is calculated independently on each frame of the image, but the temporal consistency of the sequence is considered. For example, a sliding window method is used with a window size of 5 frames. The average brightness of the frames within the window is calculated as the basis for adjustment to ensure that the dynamic threshold changes smoothly and avoids abrupt changes.

[0027] The system identifies candidate communication light source regions in an image sequence whose brightness exceeds a dynamic threshold. This identification process is achieved through binarization, which compares the pixel grayscale value of each frame with the dynamic threshold. If the pixel grayscale value is higher than the dynamic threshold, it is set to white; otherwise, it is set to black, forming a binary image. Candidate communication light source regions are extracted from this binary image using connected component analysis. Connected component analysis scans the binary image, marking adjacent white pixel regions, with each region considered a candidate communication light source region. Filtering of candidate communication light source regions is based on region area and shape features. For example, the region area must be greater than a minimum area threshold (set to 50 pixels) to avoid noise points. Region shape is evaluated using aspect ratio and roundness; for example, the aspect ratio must be between 0.5 and 2.0, and the roundness must be greater than 0.7 to ensure the region closely approximates the circular characteristics of a light source. The identification process is executed in real-time within the robot's processing unit. The input is the pre-processed image sequence and the dynamic threshold; the output is a list of candidate communication light source regions. Each region contains location coordinates and region attributes, such as the region's centroid coordinates and bounding box.

[0028] A brightness value sequence for candidate communication light source regions is established. This sequence is formed by extracting the brightness values ​​of these regions from the image sequence and arranging them chronologically. For each candidate region, the average brightness value is calculated by summing the grayscale values ​​of all pixels in the region and dividing by the number of pixels in the region. The brightness value sequence is arranged according to the time frame order of the image sequence. For example, for each candidate region, the average brightness value is recorded sequentially from the first frame to the last frame, forming a time-series data. The brightness value sequence is stored using an array or list structure, with one sequence corresponding to each candidate region. The sequence length is consistent with the number of frames in the image sequence. The brightness value sequence is used for demodulation of the modulation signal in subsequent steps, such as demodulating the modulation signal from the brightness value sequence in S3. The establishment process also includes region tracking to ensure the consistency of the same light source across different frames. For example, a position-based tracking algorithm is used, comparing the centroid distance between regions in adjacent frames. If the distance is less than a tracking threshold, the region is considered the same light source. The generation of the brightness value sequence is completed by the robot's data management module, ensuring data integrity and timing accuracy.

[0029] S3. Demodulate the modulated signal from the brightness value sequence, obtain the interference regularity index by analyzing the inter-symbol interference characteristics of the modulated signal, and screen the first candidate set whose interference regularity index is greater than the first threshold. The specific implementation is as follows: The modulated signal is demodulated from the luminance value sequence. The demodulation process employs an incoherent detection method, which detects envelope changes in the luminance value sequence. The luminance value sequence originates from the luminance value sequence of the candidate communication light source region established in step S2. Modulation demodulation includes symbol synchronization and symbol decision. Symbol synchronization determines the symbol boundary by identifying the rising and falling edges of the luminance value sequence. Symbol decision determines the symbol value by comparing the average luminance value within the symbol period with a decision threshold. For example, in binary on / off keyed modulation, if the average luminance value within the symbol period is higher than the decision threshold, it is judged as symbol 1; otherwise, it is judged as symbol 0. The decision threshold is set based on the overall amplitude range of the luminance value sequence. For example, the decision threshold is set to 70% of the maximum value of the luminance value sequence. The demodulation process is completed in the robot's signal processing unit, and the output is a symbol sequence of the modulated signal. Each symbol corresponds to an amplitude value, which is taken from the average value of the luminance values ​​within the symbol period. The demodulation process also includes filtering, such as using a low-pass filter to smooth the luminance value sequence. The cutoff frequency is set according to the modulation rate. For example, for a 1000 Hz modulated signal, the cutoff frequency is set to 1500 Hz to ensure demodulation accuracy. The symbol period is determined by analyzing the autocorrelation characteristics of the luminance value sequence. For example, the autocorrelation function of the sequence is calculated, and the interval corresponding to the periodic peak is found as the symbol period.

[0030] This paper analyzes the inter-symbol interference characteristics of modulated signals. Interference regularity indices are calculated by comparing the amplitude changes of adjacent symbols. A symbol amplitude sequence is extracted from the modulated signal, containing the amplitude value of each demodulated symbol. The amplitude change value between adjacent symbols is calculated by subtracting the absolute value of the amplitude difference between them. For example, for the i-th symbol and the (i+1)-th symbol in the amplitude sequence, the amplitude change value is the absolute value of the (i+1)-th symbol's amplitude value minus the amplitude value of the i-th symbol, where i is the symbol number. The interference regularity index is calculated by analyzing the statistical distribution characteristics of the amplitude change values. These statistical distribution characteristics include… The interference regularity index is calculated by dividing the standard deviation of the amplitude variation values ​​by the average value. This ratio is called the coefficient of variation and is used to quantify the relative fluctuation of amplitude variation. For example, the smaller the coefficient of variation, the more regular the interference between symbols. The calculation process is executed in the robot's processing unit. The input is the symbol amplitude sequence of the modulated signal, and the output is the interference regularity index. The value range of the interference regularity index is from 0 to positive infinity. In practical applications, it is usually less than 10. The calculation also considers the influence of the symbol sequence length. For example, for sequences with a length of less than 10 symbols, interpolation methods are used to supplement them to ensure statistical reliability.

[0031] The interference regularity index is compared with a first threshold, which is set through statistical analysis based on historical experimental data. Specifically, multiple sets of test signals with known characteristics are collected in a laboratory environment, including real base station signals and interference signals. The interference regularity index of each set of signals is calculated and a distribution map is plotted. The classification boundary is determined by analyzing the distribution characteristics. For example, the interference regularity index of real base station signals is mainly concentrated between 0.3 and 0.6, while the index of interference signals is mostly distributed above 0.7. Therefore, the first threshold is set to 0.5 to achieve the best classification effect. The setting of the first threshold also needs to consider the influence of the actual application environment. For example, the above experimental process is repeated under different lighting conditions to adjust the first threshold to maintain the screening accuracy. The first threshold is adjusted to 0.6 in strong light environment and to 0.4 in weak light environment. The comparison process is completed by the robot's comparison unit. If the interference regularity index is less than the first threshold, the modulation signal of the candidate communication light source area is considered to have regular interference characteristics, which are consistent with the characteristics of real base station communication light sources.

[0032] A first candidate set is selected based on the interference regularity index exceeding a first threshold. This first candidate set contains candidate communication light source regions whose interference regularity index exceeds the first threshold. The selection process involves iterating through the interference regularity indexes of all candidate communication light source regions and selecting regions with index values ​​greater than the first threshold. The generation of the first candidate set is completed in the robot's data management module, and the output is a list of candidate regions. Each region contains location information and a corresponding interference regularity index. The first candidate set is used for motion trajectory analysis in subsequent steps. For example, in S4, the smoothness characteristics of the motion trajectory are analyzed using the first candidate set. The selection process also includes deduplication. For example, if multiple candidate regions overlap, the region with the largest interference regularity index is selected to ensure the uniqueness of the first candidate set. The selection results are stored in the robot's storage unit for easy retrieval and processing later. A minimum interference regularity index requirement is also set during the selection process. For example, only candidate regions with interference regularity indices between 0.1 and 0.8 are retained to exclude the influence of outliers.

[0033] S4. Analyze the smoothness characteristics of the motion trajectory of the first candidate set, calculate the trajectory smoothness index by evaluating the continuity of the trajectory point distribution, and screen the second candidate set whose trajectory smoothness index is greater than the second threshold. The specific implementation is as follows: Motion trajectories are constructed based on the position coordinates of candidate communication light source regions in consecutive image frames within the first candidate set. The position coordinates are derived from the centroid coordinates of the candidate communication light source regions in the image sequence. The motion trajectory is formed by connecting the position coordinates of the same candidate communication light source region in different image frames in chronological order. Each motion trajectory contains a series of trajectory points, and the number of trajectory points corresponds to the number of frames in the image sequence. For example, for an image sequence lasting 10 seconds, calculated at a frame rate of 30 frames per second, the motion trajectory contains 300 trajectory points. A tracking algorithm is used to ensure trajectory continuity when constructing the motion trajectory. The tracking algorithm is implemented by comparing the position distance of candidate communication light source regions in adjacent frames. If the position distance is less than the tracking threshold, it is considered as the same light source. The tracking threshold is set according to the image resolution and shooting distance. For example, for a 1920×1080 pixel image, the tracking threshold is set to 30 pixels. The coordinate system of the motion trajectory takes the upper left corner of the image as the origin, the horizontal direction to the right as the positive x-axis, and the vertical direction downward as the positive y-axis. All coordinate values ​​are in pixels.

[0034] The movement vectors between adjacent trajectory points are extracted. The movement vectors are obtained by calculating the coordinate difference between adjacent trajectory points. For example, for trajectory points P1 and P2, the movement vector is the x-coordinate of P2 minus the x-coordinate of P1 as the horizontal component, and the y-coordinate of P2 minus the y-coordinate of P1 as the vertical component. The direction of the movement vector is obtained by calculating the angle between the vector and the positive x-axis, with the angle ranging from 0 to 360 degrees. The magnitude of the movement vector is obtained by calculating the square root of the sum of the squares of the horizontal and vertical components. The consistency of the direction change of the movement vector is analyzed. The degree of direction change is evaluated by calculating the absolute value of the difference in the direction angle between adjacent movement vectors. For example, for three consecutive trajectory points P1, P2, and P3, the direction angle θ1 of the movement vector from P1 to P2 is calculated first, and then the direction angle θ2 of the movement vector from P2 to P3 is calculated. The difference in direction angle is the absolute value of θ2 minus θ1, with the difference ranging from 0 to 180 degrees.

[0035] The trajectory smoothness index is calculated by evaluating the stability of the change in the direction of the moving vector. The trajectory smoothness index is obtained by statistically analyzing the standard deviation of the angular differences between all adjacent moving vectors. The smaller the standard deviation, the more stable the change in the direction of the moving vector. The calculation of the trajectory smoothness index also considers the influence of the magnitude of the moving vector. For example, the angular differences are weighted and averaged, with the weight being the reciprocal of the magnitude of the corresponding moving vector, to ensure that short-distance movement has a small impact on the smoothness assessment. The value range of the trajectory smoothness index is from 0 to positive infinity, and in practical applications it is usually less than 50 degrees. The sliding window method is also used in the calculation process to improve stability. For example, a window size of 5 moving vectors is used to calculate the moving average of the angular differences within the window, and then the standard deviation of the moving average is calculated as the trajectory smoothness index.

[0036] The trajectory smoothness index is compared with a second threshold, which is set through experimental analysis based on historical motion trajectory data. Specifically, multiple sets of motion trajectories with known characteristics are collected in a laboratory environment, including real base station light source trajectories and interference light source trajectories. The trajectory smoothness index of each set of trajectories is calculated and the distribution pattern is analyzed. For example, statistical analysis shows that the trajectory smoothness index of real base station light source trajectories is mainly concentrated between 5 and 15 degrees, while the trajectory smoothness index of interference light source trajectories is mostly distributed between 20 and 40 degrees. Therefore, the second threshold is set to 18 degrees to achieve the best classification effect. The setting of the second threshold also needs to consider the influence of the robot's motion state. For example, when the robot moves quickly, the second threshold is appropriately increased to 22 degrees, and when the robot moves slowly, the second threshold is decreased to 15 degrees. The comparison process is completed by the robot's comparison unit. If the trajectory smoothness index is less than the second threshold, the motion trajectory of the candidate communication light source area is considered to have smooth characteristics.

[0037] A second candidate set is formed by screening candidate communication light source regions whose trajectory smoothness index is greater than a second threshold. The screening process involves traversing the trajectory smoothness index of all candidate communication light source regions in the first candidate set and selecting regions whose index value is greater than the second threshold. The generation of the second candidate set is completed in the robot's data management module, and the output is a list of candidate regions. Each region contains location information, interference regularity index, and trajectory smoothness index. The second candidate set is used for inverse correlation analysis in subsequent steps. For example, in S5, the inverse correlation between the modulation signal of the candidate communication light source region and the automatic exposure parameter sequence is analyzed. The screening process also includes trajectory length verification. For example, only candidate regions with a number of trajectory points greater than the minimum number of trajectory points are retained. The minimum number of trajectory points is set to 20 points to ensure the reliability of motion trajectory analysis. The trajectory integrity requirement is also considered during the screening process. For example, the trajectory point missing rate is required to be less than 10% to exclude incomplete trajectories caused by occlusion or other reasons.

[0038] S5. For the second candidate set, analyze the inverse correlation between the modulation signal of the candidate communication light source region and the automatic exposure parameter sequence, and calculate the inverse correlation index. The specific implementation is as follows: The modulation signal of the candidate communication light source region in the second candidate set is extracted. The modulation signal comes from the demodulation result of the brightness value sequence of the candidate communication light source region. The brightness value sequence is obtained by the brightness value sequence of the candidate communication light source region established in step S2. The demodulation process adopts an incoherent detection method. The modulation signal is extracted by detecting the envelope change of the brightness value sequence. For example, for binary on / off keying modulation, the part of the brightness value sequence above the decision threshold is judged as symbol 1, and the part below the decision threshold is judged as symbol 0. The modulation signal is represented as the amplitude value sequence corresponding to the symbol sequence. The amplitude value is taken from the average value of the brightness value in each symbol period. The sampling frequency of the modulation signal is consistent with the frame rate of the image sequence. For example, for an image sequence of 30 frames per second, the sampling frequency of the modulation signal is 30 Hz.

[0039] The automatic exposure parameter sequence is obtained from the exposure parameter records of the optical sensor when acquiring the image sequence. The automatic exposure parameter sequence includes exposure time parameters and gain parameters. The exposure time parameter represents the exposure duration of each frame of the image from the optical sensor, in milliseconds. The gain parameter represents the signal amplification factor of the optical sensor, in ISO values. The automatic exposure parameter sequence corresponds one-to-one with each frame of the image sequence. The parameter values ​​are obtained by reading the registers of the optical sensor or the application programming interface. The sampling frequency of the automatic exposure parameter sequence is the same as the frame rate of the image sequence. For example, for an image sequence of 30 frames per second, the automatic exposure parameter sequence contains 30 sets of exposure time parameters and gain parameters per second.

[0040] The correlation between the changing trends of the modulation signal and the automatic exposure parameter sequence is analyzed. The changing trend is characterized by calculating the first-order difference between the two sequences. The first-order difference represents the numerical change between adjacent sampling points. For the modulation signal, the first-order difference is obtained by subtracting the amplitude of the previous sampling point from the amplitude of a certain sampling point. For the automatic exposure parameter sequence, the first-order difference of the exposure time parameter and the first-order difference of the gain parameter are calculated respectively. The correlation between the changing trends is determined by comparing the signs of the first-order differences of the two sequences. When the first-order difference of the modulation signal and the first-order difference of the automatic exposure parameter sequence have opposite signs, the changing trends are considered to be opposite.

[0041] The inverse correlation index is calculated by evaluating the degree of inverse relationship between the changing trends of the modulation signal and the automatic exposure parameter sequence. Time series data of the modulation signal and the automatic exposure parameter sequence are extracted, and the inverse correlation index is obtained by statistically analyzing the frequency of opposite changes. The frequency of opposite changes is defined as the proportion of sampling points with opposite trends in the total number of sampling points. For example, for a sequence of length N, the number M of sampling points with opposite first-order difference signs is counted, and the inverse correlation index is calculated as M divided by N. The inverse correlation index ranges from 0 to 1; a larger value indicates a stronger inverse correlation between the two sequences. A sliding window method is also used in the calculation to improve stability. For example, using a window size of 10 sampling points, the inverse correlation index within the window is calculated, and the average of the inverse correlation indices of all windows is taken as the final result. The calculation of the inverse correlation index also considers parameter weights. For example, the inverse correlation index is calculated separately for the exposure time parameter and the gain parameter, and then the weights are summed according to the importance of the parameters. The weight of the exposure time parameter is set to 0.6, and the weight of the gain parameter is set to 0.4. The weight allocation is determined experimentally based on the degree of influence of the parameters on the image brightness. The specific weight setting method is to test the inverse correlation index discrimination under different weight combinations in a laboratory environment and select the weight combination that makes the real base station and the interference source most obvious.

[0042] S6. Select the candidate communication light source region with the highest inverse correlation index from the second candidate set as the real base station communication light source. Calculate the robot's heading adjustment amount based on the positional change of the real base station communication light source, and control the robot to move and complete the recharging process. Specifically, the implementation is as follows: The inverse correlation index of each candidate communication light source region in the second candidate set is compared. The comparison process involves traversing the inverse correlation index values ​​of all candidate communication light source regions in the second candidate set. The inverse correlation index values ​​are derived from the inverse correlation index calculated in step S5. The comparison method employs a sorting algorithm, such as using a quicksort algorithm to arrange the candidate communication light source regions in descending order of their inverse correlation index values. During the sorting process, the inverse correlation index values ​​are compared in floating-point form with precision retained to two decimal places. The comparison process is completed in the robot's processing unit. The input is a list of candidate regions in the second candidate set, with each region containing an inverse correlation index value. The output is a sorted list of candidate regions. The time complexity of the sorting algorithm is O(n log n), where n is the number of candidate regions in the second candidate set. For example, when the second candidate set contains 10 candidate regions, the sorting process can be completed in milliseconds, ensuring real-time performance.

[0043] The candidate communication light source region with the highest inverse correlation index is selected as the real base station communication light source. The selection process involves choosing the candidate communication light source region with the largest inverse correlation index value from the sorted candidate region list. The criterion for the largest inverse correlation index value is the first element in the sorted list. If multiple candidate regions have the same highest inverse correlation index value, the candidate region with the highest trajectory smoothness index is selected. The trajectory smoothness index is derived from the trajectory smoothness index calculated in step S4. The determination of the real base station communication light source also considers position stability. For example, it is required that the variance of the position coordinates of the candidate region in consecutive image frames is less than the position variance threshold, which is set to 100 pixels squared. The selection process is executed in the robot's decision unit, and the output is the region identifier and attribute information of the real base station communication light source, including position coordinates, inverse correlation index value, and trajectory smoothness index.

[0044] The position coordinates of the real base station communication light source in consecutive image frames are extracted. The position coordinates are derived from the centroid coordinates of the real base station communication light source in the image sequence. The centroid coordinates are obtained by calculating the average value of the pixel coordinates of the candidate communication light source region. For example, for each image frame, the centroid coordinates of the real base station communication light source are obtained by summing the x-coordinates of all pixels in the region and dividing by the number of pixels, and the y-coordinates are obtained by summing the y-coordinates and dividing by the number of pixels. The extraction range of position coordinates covers the entire image sequence. The extraction process is completed by accessing the image sequence data stored in the robot's memory. The position coordinates are stored in two-dimensional coordinate form, with each coordinate point containing x-coordinate and y-coordinate values ​​in pixels. The extraction frequency of position coordinates is consistent with the frame rate of the image sequence. For example, for an image sequence of 30 frames per second, 30 sets of position coordinates are extracted per second.

[0045] The robot's heading adjustment is calculated based on the trend of position coordinate changes in consecutive image frames. The trend is characterized by calculating the first-order difference of the position coordinate sequence. The first-order difference represents the change in position coordinates between adjacent image frames. For example, for the x-coordinate sequence, the first-order difference is obtained by subtracting the x-coordinate value of the previous frame from the x-coordinate value of a certain frame. The same applies to the y-coordinate sequence. The robot's heading adjustment is obtained by calculating the difference between the direction angle of the position coordinate change vector and the robot's current heading angle. The direction angle is obtained by calculating the arctangent of the position coordinate change vector. The unit is degrees. The robot's current heading angle comes from the robot's inertial measurement unit. The calculation of the heading adjustment also considers the relative distance between the robot and the real base station communication light source. For example, when the relative distance is less than a distance threshold, the heading adjustment is reduced. The distance threshold is set to 200 pixels. The value range of the heading adjustment is -180 degrees to +180 degrees. Negative values ​​indicate left turn adjustment, and positive values ​​indicate right turn adjustment.

[0046] Control commands are generated based on the robot's heading adjustment and sent to the robot's motion control system to complete the recharging operation. Control command generation involves converting the heading adjustment into wheel speed control signals. For example, for a differential drive robot, the left and right wheel speeds are calculated by adding the heading adjustment to the base speed and multiplying by a proportional coefficient. This proportional coefficient is determined by the robot's kinematic model; for example, it might be set to 0.1. The control commands are sent to the robot's motion control system as digital signals via a serial communication interface. The sending frequency matches the robot's control cycle; for example, if the control cycle is 100 milliseconds, 10 control commands are sent per second. The recharging operation involves the robot moving towards the direction of the actual base station's communication light source until the robot's charging port physically contacts the base station's charging contacts. During the movement, the robot continuously monitors changes in its position coordinates and dynamically adjusts the heading adjustment to ensure the smoothness and accuracy of the movement path. A precision alignment procedure is also initiated when approaching the base station, for example, using infrared sensors for assisted positioning to improve the docking success rate.

[0047] Example 2: Figure 2 A schematic diagram of the optical communication-driven robot base station recharging system of the present invention is provided. The optical communication-driven robot base station recharging system includes: The sequence acquisition module is used to acquire image sequences containing visible light communication signals emitted by the base station through the robot's optical sensors, and to record automatic exposure parameter sequences; The sequence establishment module is used to preprocess the image sequence, identify candidate communication light source regions with brightness higher than the dynamic threshold, and establish a brightness value sequence of the candidate communication light source regions. The first candidate module is used to demodulate the modulated signal from the brightness value sequence, obtain the interference regularity index by analyzing the inter-symbol interference characteristics of the modulated signal, and screen the first candidate set whose interference regularity index is greater than the first threshold. The second candidate module is used to analyze the smoothness characteristics of the motion trajectory of the first candidate set, calculate the trajectory smoothness index by evaluating the continuity of the trajectory point distribution, and screen the second candidate set whose trajectory smoothness index is greater than the second threshold. The index calculation module is used to analyze the inverse correlation between the modulation signal of the candidate communication light source region and the automatic exposure parameter sequence in the second candidate set, and to calculate the inverse correlation index. The recharge control module is used to select the candidate communication light source region with the highest inverse correlation index from the second candidate set as the real base station communication light source, calculate the robot's heading adjustment amount based on the position change of the real base station communication light source, and control the robot to move to complete the recharge.

[0048] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0049] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0050] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission; wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission includes infrared, microwave, etc. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center containing one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0051] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0052] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0053] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0054] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0055] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0056] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0057] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for recharging a robot base station driven by optical communication, characterized in that, include: S1. Acquire image sequences containing visible light communication signals emitted by the base station using the robot's optical sensors, and record the automatic exposure parameter sequence; S2. Preprocess the image sequence to identify candidate communication light source regions with brightness higher than the dynamic threshold, and establish a brightness value sequence of the candidate communication light source regions. S3. Demodulate the modulation signal from the brightness value sequence, obtain the interference regularity index by analyzing the inter-symbol interference characteristics of the modulation signal, and screen the first candidate set whose interference regularity index is greater than the first threshold. S4. Analyze the smoothness characteristics of the motion trajectory of the first candidate set, calculate the trajectory smoothness index by evaluating the continuity of the trajectory point distribution, and screen the second candidate set whose trajectory smoothness index is greater than the second threshold. S5. For the second candidate set, analyze the inverse correlation between the modulation signal of the candidate communication light source region and the automatic exposure parameter sequence, and calculate the inverse correlation index. S6. Select the candidate communication light source region with the highest inverse correlation index from the second candidate set as the real base station communication light source. Calculate the robot's heading adjustment amount based on the position change of the real base station communication light source and control the robot to move to complete the recharging.

2. The method for recharging a robot base station driven by optical communication according to claim 1, characterized in that, The robot's optical sensors acquire image sequences containing visible light communication signals emitted by the base station and record automatic exposure parameter sequences, including: The optical sensor acquires image sequences containing visible light communication signals emitted by the base station and records automatic exposure parameter sequences; The automatic exposure parameter sequence is generated by the optical sensor during the acquisition of the image sequence and includes exposure time parameters and gain parameters.

3. The method for recharging a robot base station driven by optical communication according to claim 1, characterized in that, Image sequence preprocessing is performed to identify candidate communication light source regions with brightness exceeding a dynamic threshold, and a brightness value sequence for these candidate communication light source regions is established, including: Image sequences are preprocessed to eliminate environmental noise interference; A dynamic threshold is calculated based on the preprocessed image sequence, and the dynamic threshold is adaptively determined according to the overall brightness characteristics of the image sequence. Identify candidate communication light source regions in an image sequence whose brightness exceeds a dynamic threshold; A brightness value sequence of candidate communication light source regions is established. The brightness value sequence is formed by extracting the brightness values ​​of the candidate communication light source regions in the image sequence and arranging them in chronological order.

4. The method for recharging a robot base station driven by optical communication according to claim 1, characterized in that, The modulated signal is demodulated from the brightness value sequence. Inter-symbol interference characteristics of the modulated signal are analyzed to obtain interference regularity indices. A first candidate set with interference regularity indices greater than a first threshold is selected, including: Demodulate the modulated signal from the brightness value sequence; Analyze the inter-symbol interference characteristics of the modulated signal and calculate the interference regularity index by comparing the amplitude changes of adjacent symbols; Compare the interference regularity index with the first threshold; A first candidate set is selected based on the interference regularity index being greater than a first threshold. The first candidate set includes candidate communication light source regions whose interference regularity index is greater than the first threshold.

5. The method for recharging a robot base station driven by optical communication according to claim 4, characterized in that, The interference regularity index obtained by analyzing the inter-symbol interference characteristics of the modulated signal includes: extracting the symbol amplitude sequence from the modulated signal, calculating the amplitude change value of adjacent symbols, and calculating the interference regularity index by analyzing the statistical distribution characteristics of the amplitude change value.

6. The method for recharging a robot base station driven by optical communication according to claim 1, characterized in that, The smoothness characteristics of the motion trajectory are analyzed for the first candidate set. A trajectory smoothness index is calculated by evaluating the continuity of the trajectory point distribution. A second candidate set with a trajectory smoothness index greater than a second threshold is then selected, including: Motion trajectories are constructed based on the position coordinates of candidate communication light source regions in consecutive image frames from the first candidate set. Extract the movement vectors between adjacent trajectory points in the motion trajectory and analyze the consistency of the changes in the direction of the movement vectors; The trajectory smoothness index is calculated by evaluating the stability of changes in the direction of the moving vector. The trajectory smoothness index is compared with a second threshold; candidate communication light source regions with trajectory smoothness indices greater than the second threshold are selected to form a second candidate set.

7. The method for recharging a robot base station driven by optical communication according to claim 1, characterized in that, For the second candidate set, the inverse correlation between the modulation signal of the candidate communication light source region and the automatic exposure parameter sequence is analyzed, and the inverse correlation index is calculated, including: Extract the modulation signal of the candidate communication light source region from the second candidate set; Obtain the automatic exposure parameter sequence; Analyze the correlation between the changing trends of the modulation signal and the automatic exposure parameter sequence; The inverse correlation index is calculated by evaluating the degree of inverse relationship between the modulation signal and the changing trends of the automatic exposure parameter sequence; The modulation signal comes from the demodulation result of the brightness value sequence of the candidate communication light source region, and the automatic exposure parameter sequence comes from the exposure parameter record when the optical sensor acquires the image sequence.

8. The method for recharging a robot base station driven by optical communication according to claim 7, characterized in that, The inverse correlation index is calculated by evaluating the degree of opposition between the changing trends of the modulation signal and the automatic exposure parameter sequence. This includes: extracting time series data of the modulation signal and the automatic exposure parameter sequence, analyzing the opposition between their changing trends, and obtaining the inverse correlation index by statistically analyzing the frequency of opposite changes.

9. The method for recharging a robot base station driven by optical communication according to claim 1, characterized in that, The candidate communication light source region with the highest inverse correlation index is selected from the second candidate set as the real base station communication light source. The robot's heading adjustment is calculated based on the positional change of the real base station communication light source, and the robot is controlled to move and complete the recharging process, including: Compare the inverse correlation indices of each candidate communication light source region in the second candidate set; The candidate communication light source region with the highest inverse correlation index was selected as the actual base station communication light source. Extract the position coordinates of the real base station communication light source in consecutive image frames; The robot's heading adjustment is calculated based on the trend of position coordinate changes in consecutive image frames; Control commands are generated based on the robot's heading adjustment and sent to the robot's motion control system to complete the recharge operation.

10. A robot base station recharging system driven by optical communication, used to implement the robot base station recharging method driven by optical communication as described in any one of claims 1-9, characterized in that, include: The sequence acquisition module is used to acquire image sequences containing visible light communication signals emitted by the base station through the robot's optical sensors, and to record automatic exposure parameter sequences; The sequence establishment module is used to preprocess the image sequence, identify candidate communication light source regions with brightness higher than the dynamic threshold, and establish a brightness value sequence of the candidate communication light source regions. The first candidate module is used to demodulate the modulated signal from the brightness value sequence, obtain the interference regularity index by analyzing the inter-symbol interference characteristics of the modulated signal, and screen the first candidate set whose interference regularity index is greater than the first threshold. The second candidate module is used to analyze the smoothness characteristics of the motion trajectory of the first candidate set, calculate the trajectory smoothness index by evaluating the continuity of the trajectory point distribution, and screen the second candidate set whose trajectory smoothness index is greater than the second threshold. The index calculation module is used to analyze the inverse correlation between the modulation signal of the candidate communication light source region and the automatic exposure parameter sequence in the second candidate set, and to calculate the inverse correlation index. The recharge control module is used to select the candidate communication light source region with the highest inverse correlation index from the second candidate set as the real base station communication light source, calculate the robot's heading adjustment amount based on the position change of the real base station communication light source, and control the robot to move to complete the recharge.

Images