Scatter DTOF chip error compensation method

Abstract

The invention relates to the technical field of DTOF chip error compensation. The scattered point DTOF chip error compensation method comprises the steps that N first depth values and N first histograms of each scattered point are obtained, the first rising edge width of the highest peak is calculated, a first preset depth value is obtained according to a first preset distance, and therefore a first error value between the first depth value and the first preset depth value is calculated, n first error values and N first rising edge widths of each scatter point are obtained, the N first error values and the N first rising edge widths of each scatter point are input into a mathematical model, a fitting formula is generated, and an error compensation coefficient of each scatter point is obtained through calculation; and inputting the error compensation coefficient of each scatter point and the width of the second rising edge into a fitting formula to obtain a second error value of each scatter point under the current distance, and adding the second error value to the second depth value to obtain a final measurement depth. The method has the effect of improving the overall ranging performance.

Description

Technical Field

[0001] The present invention relates to the technical field of DTOF chip error compensation, and in particular to a scattered point DTOF chip error compensation method. Background Art

[0002] With the rapid development of laser ranging technology, direct time of flight (DTOF) technology has been widely used in many applications due to its high precision and fast response capabilities, such as robot navigation, autonomous driving, 3D reconstruction and industrial measurement. DTOF technology calculates the distance of an object by emitting light pulses and measuring the time difference between emission and reception. This method not only improves the accuracy and speed of ranging, but also expands its application scenarios. However, traditional methods are difficult to adapt to multiple measurement environments at the same time, resulting in insufficient ranging accuracy. Specifically, changes in signal strength can cause deviations in measurement time, especially when measuring at close range. This deviation is particularly obvious. Therefore, when dealing with variable measurement conditions, existing technologies often require manual adjustments or the use of complex calibration processes, which not only increases the complexity and implementation cost of the system, but also limits its reliability and flexibility in practical applications. Summary of the invention

[0003] In order to improve the overall ranging performance, the present application provides a scattered point DTOF chip error compensation method.

[0004] The above-mentioned invention objective of the present application is achieved through the following technical solutions: A scattered point DTOF chip error compensation method, comprising: Under N different preset distances, obtain N first depth values ​​and N first histograms of each scattered point; For each of the scattered points, according to one of the first histograms, a first rising edge width of the highest peak is calculated, and a first preset depth value is obtained according to a first preset distance, thereby calculating a first error value between the first depth value and the first preset depth value, thereby obtaining the N first error values ​​and the N first rising edge widths of each of the scattered points; Inputting the N first error values ​​and the N first rising edge widths of each scattered point into a mathematical model to generate a fitting formula, calculating the error compensation coefficient of each scattered point by the fitting formula, and storing the error compensation coefficient in the chip; In the real-time distance measurement process, a second histogram of each scattered point is obtained, and a second rising edge width and a second depth value of the highest peak are determined according to the second histogram; Inputting the error compensation coefficient of each scattered point and the width of the second rising edge into the fitting formula to obtain a second error value of each scattered point at the current distance; The second error value is added to the second depth value to obtain a final measured depth.

[0005] By adopting the above technical solution, the error compensation coefficient of each scattered point can be obtained by fitting the depth error value of each scattered point and the rising edge width of the highest peak of the histogram, thereby effectively correcting the errors caused by signal waveform distortion or environmental changes in the ranging process. By combining the error compensation coefficient with real-time measurement data, the system can accurately correct the depth value of each scattered point and achieve the final high-precision measurement, significantly improving the ranging accuracy and system stability, especially when facing different ranging distances, environmental changes and other factors, it can still maintain excellent performance and ensure the accuracy of the measurement results. In addition, the error compensation coefficient storage method in the solution enables the system to have better adaptability, and can automatically adjust to different ranging conditions, improve the robustness and adaptability of the system, and reduce the need for manual adjustment.

[0006] In a preferred example of the present application, obtaining N first depth values ​​and N first histograms of each scattered point at N different preset distances includes: At a preset distance, obtaining a first histogram of each scatter point; Based on the first histogram, a highest peak is obtained, and according to the highest peak, a first depth value is obtained; After obtaining the one first depth value, the preset distance is changed to obtain the N first depth values ​​and the N first histograms of each scattered point.

[0007] By adopting the above technical solution, the depth information of each scattered point can be effectively obtained at multiple preset distances, and the depth value of each scattered point can be obtained by histogram analysis. First, a first histogram of each scattered point is obtained at a preset distance, and the corresponding first depth value is calculated according to the highest peak of the first histogram, thereby realizing preliminary depth measurement. As the preset distance changes, N different depth values ​​and corresponding N first histograms are obtained.

[0008] In a preferred example of the present application: the first rising edge width and the second rising edge width include: The first rising edge width is obtained by calculating the time bin difference between the highest point of the highest peak in the first histogram and the starting point of the highest peak; The second rising edge width is obtained by calculating the time bin difference between the highest point of the highest peak in the second histogram and the starting point of the highest peak.

[0009] By adopting the above technical solution, the rising edge width of each scattered point in the first histogram and the second histogram can be accurately calculated. During the measurement process, the system first analyzes the first histogram to find the highest peak, and determines the first rising edge width by calculating the time bin difference between the highest point and the starting point of the peak. Subsequently, a similar calculation is performed based on the second histogram to obtain the second rising edge width.

[0010] In a preferred example of the present application: the first depth value and the second depth value include: The first depth value is a first time bin corresponding to the highest point of the highest peak in the first histogram; The second depth value is a second time bin corresponding to the highest point of the highest peak in the second histogram.

[0011] By adopting the above technical solution, the depth value of each scattered point can be accurately calculated based on the highest point of the highest peak in the first histogram and the second histogram. When processing the first histogram, the highest peak of the signal is first located, and the time bin position corresponding to its highest point is found. This time bin is the first depth value, which represents the depth information of the scattered point at a specific distance. Subsequently, for the second histogram, the system also locates the highest point of the highest peak and determines the corresponding time bin. This time bin is the second depth value, which represents the actual depth during the measurement process.

[0012] In a preferred example of the present application: the mathematical model includes: the mathematical model is a linear mathematical model.

[0013] By adopting the above technical solution, the linear mathematical model can be used to simplify the complex data fitting process, making the calculation of the error compensation coefficient more efficient and accurate. In this solution, the linear model effectively expresses the error characteristics of the scattered depth value by determining the linear relationship between the error and the rising edge width.

[0014] In a preferred example of the present application, the fitting formula includes: The fitting formula is y=a·x+b, wherein y is a characterizing error value, x is a characterizing rising edge width of the highest peak, and a and b are the error compensation coefficients.

[0015] By adopting the above technical solution, the relationship between the error value y in the fitting formula and the peak rising edge width x is determined by coefficients a and b. These coefficients are obtained by fitting historical data, ensuring the accuracy and consistency of the error compensation coefficient.

[0016] In a preferred example of the present application: the N different preset distances also include: N-1 different preset distances under the assumed distance are short distances; and 1 preset distance outside the assumed distance is a long distance.

[0017] By adopting the above technical solution, the N-1 preset distances within the assumed distance are regarded as close distances. These distances are closer to the target object and can usually provide more accurate measurement results, while the long distance is outside the assumed distance and is usually affected by greater environmental factors, so special processing methods are required. By distinguishing between close distances and long distances, different compensation strategies and error models can be selected according to different distance ranges, thereby improving the system's adaptability to different ranging scenarios.

[0018] In a preferred example of the present application: the long distance includes: at the long distance, selecting the scattered points in the scatter point array that meet the preset scatter point conditions as the central scatter points, and the remaining scatter points are edge scatter points.

[0019] By adopting the above technical solution, at long distances, the edge scattered points on the scattered point array will not receive the echo signal of the reflector due to the long distance, so the scattered points with higher signal quality and measurement accuracy must be screened out through the preset scattered point conditions. The error compensation coefficient methods for the center scattered points and the edge scattered points are slightly different and need to be distinguished.

[0020] In a preferred example of the present application: the edge scatter points include: for the edge scatter points, first removing the first error value and the first rising edge width at the long distance, and then inputting N-1 first error values ​​and N-1 first rising edge widths of the edge scatter points at the close distance into the mathematical model to generate a fitting formula, and calculating the error compensation coefficient of the edge scatter points through the fitting formula.

[0021] By adopting the above technical solution, for edge scatter points, the first error value and the first rising edge width at a long distance are first removed to avoid the influence of errors caused by noise or interference in the long-distance measurement data on the compensation process. Then, multiple error values ​​and rising edge width data of edge scatter points at a close distance are input into the mathematical model to re-estimate its error compensation coefficient. Through this process, the mathematical model can be fitted according to the effective data at a close distance to generate a new fitting formula, which can adjust the error compensation coefficient according to the actual situation, thereby providing a more accurate correction value for the edge scatter points during the compensation process.

[0022] In a preferred example of the present application: the method for selecting the central scatter point includes: taking the current scatter point as the center, selecting a 6×6 area as the scatter point area to be measured; obtaining the third depth values ​​of m scatter points in the scatter point area to be measured; calculating m differences between the third depth values ​​of the m scatter points and the depth values ​​corresponding to the true distance; counting the number of scatter points whose m differences are less than 40 mm, and if the number of scatter points is greater than 6, the current scatter point is considered to be the central scatter point.

[0023] By adopting the above technical solution, the central scattered points can be effectively screened out during the measurement process, further improving the measurement accuracy. First, with the current scattered point as the center, a 6×6 area is selected as the scattered point area to be measured, so that more data can be collected in a smaller range, so as to more accurately evaluate the effectiveness of the current scattered point. Then, by obtaining the third depth value of the m scattered points in the area, the specific measurement data of the scattered points in the area can be obtained. Subsequently, the difference between the third depth value of these scattered points and the depth value corresponding to the true distance is calculated to obtain m differences, which are used to evaluate the measurement error of each scattered point. Finally, by counting the number of scattered points whose difference is less than 40mm, if the number of central scattered points that meet the conditions exceeds 6, the current scattered point is considered to be the central scattered point, thereby ensuring that the screened scattered points can meet the requirements of high-precision measurement.

[0024] The present application also provides a scattered-point DTOF chip, comprising a transmitter, a scattered-point array, a histogram module, and a controller, wherein the controller executes any of the methods described above.

[0025] The present application also provides an electronic device, including the scattered-point DTOF chip as described above.

[0026] In summary, the present application includes at least one of the following beneficial technical effects: 1. By adopting the above technical solution, the error compensation coefficient of each scattered point can be obtained by fitting the depth error value of each scattered point and the rising edge width of the highest peak of the histogram, thereby effectively correcting the errors caused by signal waveform distortion or environmental changes in the ranging process. By combining the error compensation coefficient with real-time measurement data, the system can accurately correct the depth value of each scattered point and achieve the final high-precision measurement, significantly improving the ranging accuracy and system stability, especially when facing different ranging distances, environmental changes and other factors, it can still maintain excellent performance and ensure the accuracy of the measurement results. In addition, the error compensation coefficient storage method in the solution enables the system to have better adaptability, and can automatically adjust to different ranging conditions, improve the robustness and adaptability of the system, and reduce the need for manual adjustment.

[0027] 2. At long distances, due to the long distance, environmental factors and noise may have a greater impact, and some scattered points may not receive the signal of the reflector, so it is necessary to distinguish between central scattered points and edge scattered points. For edge scattered points, the linear fitting formula is slightly different. The central scattered points that meet the conditions will be used for subsequent error compensation calculations, while the scattered points that do not meet the conditions are judged as edge scattered points. The calculation method of the error compensation coefficient of edge scattered points is slightly different from that of the central scattered points, but in the ranging process, the edge scattered points also need to output depth values, and depth error compensation is required.

[0028] 3. The central scattered points can be effectively screened out during the measurement process to further improve the measurement accuracy. First, with the current scattered point as the center, a 6×6 area is selected as the scattered point area to be measured, so that more data can be collected in a smaller range to more accurately evaluate the effectiveness of the current scattered point. Then, by obtaining the third depth value of the m scattered points in the area, the specific measurement data of the scattered points in the area can be obtained. Subsequently, the difference between the third depth value of these scattered points and the depth value corresponding to the true distance is calculated to obtain m differences, which are used to evaluate the measurement error of each scattered point. Finally, by counting the number of scattered points with a difference less than 40 mm, if the number of central scattered points that meet the conditions exceeds 6, the current scattered point is considered to be the central scattered point, thereby ensuring that the screened scattered points can meet the requirements of high-precision measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a flow chart of a scattered point DTOF chip error compensation method in one embodiment of the present application; Figure 2 is a graph showing the relationship between the rising edge width and the error value of scattered points at different distances in one embodiment of the present application; Figure 3 is a histogram of any scattered points in an embodiment of the present application; Figure 4 is a scatter point array including central scatter points and edge scatter points in one embodiment of the present application; Figure 5 is a graph showing the relationship between the rising edge width and the error value at different distances of edge scattered points in one embodiment of the present application; Figure 6 This is a block diagram of a scattered-point DTOF chip in one embodiment of the present application. DETAILED DESCRIPTION

[0030] The present application is further described in detail below in conjunction with the accompanying drawings.

[0031] In one embodiment, if Figure 1 As shown, the present application discloses a scattered point DTOF chip error compensation method, which specifically includes the following steps: S10: Under N different preset distances, obtain N first depth values ​​and N first histograms of each scattered point.

[0032] Specifically, at different preset distances, first set multiple preset distances (such as 300mm, 500mm, 800mm, etc.), each preset distance corresponds to a fixed measurement range. Then, based on the laser ranging principle, the laser sensor emits a pulse signal and receives the signal reflected from the surface of the object. For each scattered point, the relationship between the intensity of the received signal and time will be converted into a histogram, and the reflected signal intensity at different time points will be recorded. The depth of each scattered point is obtained through the histogram, see Figure 3 .

[0033] S20: For each scatter point, according to a first histogram, the first rising edge width of the highest peak is calculated, and the first preset depth value is obtained according to the first preset distance, thereby calculating the first error value between the first depth value and the first preset depth value, thereby obtaining N first error values ​​and N first rising edge widths for each scatter point.

[0034] Specifically, when analyzing the first histogram, first identify the peak in the histogram and locate the starting point and maximum point of the signal peak. Then, calculate the rising edge width of the peak, that is, the time difference between the time point when the signal starts to rise and the time point when it reaches the maximum value, see Figure 3 . This width can be obtained by analyzing the sampling points on the time axis to determine the speed of the crest rise and the shape of the crest. Next, based on the preset distance, using the relationship of distance = speed of light * time, calculate the first preset depth value at the distance, and compare it with the actually measured first depth value to calculate the difference between the two, which is the first error value. This error value reflects the depth measurement error caused by equipment deviation, environmental influence, etc. Through multiple calculations at different preset distances, N first error values ​​and N first rising edge widths of the scattered point at different preset distances are finally obtained.

[0035] S30: inputting the N first error values ​​and the N first rising edge widths of each scattered point into a mathematical model, generating a fitting formula, calculating the error compensation coefficient of each scattered point through the fitting formula, and storing the error compensation coefficient in the chip.

[0036] Specifically, the error value and rising edge width of each scattered point are used as input data and input into a preset mathematical model. This model can be a linear regression model or a polynomial fitting model. The most appropriate model is selected according to the distribution and relationship of the data. Figure 2The purpose of the fitting model is to derive a functional relationship through the known error value and the rising edge width, which represents the dependence between the error and the rising edge width. Through the fitting process, a fitting formula is generated to calculate the error compensation coefficient. Finally, the obtained error compensation coefficient is saved in the memory unit of the chip to ensure that the error compensation value of each scattered point can be quickly called in the subsequent measurement process to ensure real-time measurement accuracy.

[0037] S40: In the real-time distance measurement process, a second histogram of each scattered point is obtained, and a second rising edge width and a second depth value of the highest peak are determined according to the second histogram.

[0038] Specifically, during real-time measurement, each time a laser pulse is emitted and reflected back, a new histogram is generated, called the second histogram. This histogram records the relationship between the intensity of the new reflected signal and time. By analyzing the histogram, the peaks therein are identified, and the starting point and maximum point of the peaks are located. Next, the second rising edge width of the peak is calculated, that is, the time difference from the starting point of the peak to the maximum point. This width is calculated by detecting the time point from when the signal starts to rise to when it reaches the maximum value, and the corresponding time difference is calculated to finally determine the second rising edge width. At the same time, based on the time information, the relationship between the speed of light and the propagation time is used to calculate the second depth value, which refers to the time depth value from the emission of the laser signal to the arrival of the target object and then reflected back from the target object.

[0039] S50: Inputting the error compensation coefficient of each scattered point and the width of the second rising edge into a fitting formula to obtain a second error value of each scattered point at the current distance.

[0040] Specifically, first, read the error compensation coefficient stored for each scattered point from the chip memory. The compensation coefficient is calculated in the previous fitting process and usually exists in the form of a constant or polynomial coefficient. Secondly, obtain the peak data in the second histogram in the real-time ranging process. The time bin corresponding to the highest point of the peak is the second depth value, and calculate the rising edge width of the second peak, that is, the time difference between the signal from the initial rising point to the peak point. Then, input the error compensation coefficient and the second rising edge width into the established fitting formula together, and calculate the second error value of each scattered point at the current distance according to the formula. This calculation process relies on the previously defined fitting relationship, in which the fitting coefficient is optimized based on the test data of multiple scattered points. Therefore, the input error compensation coefficient and the second rising edge width will be calculated by the formula to finally obtain the error value of each scattered point at the current distance, see Figure 2 .

[0041] S60: Add the second error value to the second depth value to obtain a final measured depth.

[0042] Specifically, the second error value measured in real time is added to the second depth value to obtain the final measured depth. The error value here is calculated by a fitting formula, which indicates that the error caused by device deviation or environmental change needs to be added to the actual measured depth value for correction. Through addition, the error is compensated to the actual depth value to obtain a more accurate measurement result. This final measured depth represents the actual distance between the object and the ranging device after error compensation.

[0043] In one embodiment, obtaining N first depth values ​​and N first histograms of each scattered point at N different preset distances includes: S101: Under a preset distance, obtain a first histogram of each scattered point.

[0044] Specifically, first, determine the preset distance, for example, set a distance point (such as 300mm), and then use the sensor to measure the scattered points within the distance range. The measurement data corresponding to each scattered point will generate a histogram, which represents the distribution of the number of photons reflected back by the laser signal at different depths (times). By sampling the intensity of the reflected signal, with the depth as the horizontal axis and the intensity of the reflected signal as the vertical axis, the histogram corresponding to the scattered point is drawn, see Figure 3 It can be seen that each scatter point will generate a corresponding histogram at a preset distance.

[0045] S102: obtaining a highest peak based on a first histogram, and obtaining a first depth value according to the highest peak.

[0046] Specifically, the waveform data of the signal is extracted from the first histogram, and the highest point of the signal intensity in the histogram, i.e., the "peak", is found by analyzing the signal strength values ​​of each time bin in the histogram. Then, the depth position of the highest point is determined, which usually corresponds to the strongest point of the reflected signal. According to the depth value of the horizontal coordinate of the histogram, the depth value of the highest peak is mapped to the distance calculation formula to calculate the depth value corresponding to the current scatter point. For example, through the relationship between the speed of light and time ( Where c is the speed of light), the distance at this moment is calculated according to the depth value, that is, the first depth value of the scattered point is obtained. This depth value represents the measured depth of the scattered point at the preset distance.

[0047] S103: After obtaining one first depth value, changing the preset distance, thereby obtaining N first depth values ​​and N first histograms for each scattered point.

[0048] Specifically, after obtaining the first depth value, the preset distance of the measurement system is adjusted according to the set measurement strategy. For example, if the initial preset is 300mm, it is adjusted to different measurement distances such as 500mm, 800mm, and 1000mm. Each time a preset distance is adjusted, the sensor will re-measure and generate new scattered point data to obtain a new histogram. For each new distance, a new histogram will be re-acquired and the corresponding depth value will be calculated. Each acquired histogram and calculated depth value will be saved in the chip storage unit based on the corresponding distance, and finally the depth values ​​and histograms of multiple scattered points corresponding to multiple preset distances will be obtained.

[0049] In one embodiment, in step S20, the first rising edge width and the second rising edge width include: The first rising edge width is obtained by calculating the time bin difference between the highest point of the highest peak in the first histogram and the starting point of the highest peak.

[0050] Specifically, first, extract the data of the first histogram, which has been obtained by sampling the time distribution of the reflected signal, where the horizontal axis is the time bin (usually the time bin, indicating the measured time period), and the vertical axis is the signal strength. From the histogram, find the highest peak, that is, the point with the largest signal strength. Then, determine the position of the highest point of the peak, which corresponds to a specific time point in the time bin. Next, determine the starting point of the peak, which refers to the earliest moment when the peak starts to rise from zero, that is, the position where the signal strength begins to increase and reaches a significant increase. Calculate the time difference from the starting point to the highest point, that is, the time bin difference, which represents the time it takes for the peak to reach the maximum value from the beginning. This difference is the first rising edge width, which reflects the rising time of the signal peak. The first rising edge width is obtained through the time bin difference.

[0051] The second rising edge width is obtained by calculating the time bin difference between the highest point of the highest peak in the second histogram and the starting point of the highest peak.

[0052] Specifically, first, extract the data of the second histogram, which is obtained by sampling the reflected signal of the second measurement, with the horizontal axis being the time bin and the vertical axis being the signal intensity. Find the peak with the highest signal intensity from the second histogram, that is, the maximum value point of the signal. Then, determine the highest point of the peak, that is, the moment when the signal intensity is maximum in the second histogram. Next, find the starting point of the peak, which is the moment when the signal intensity begins to rise significantly, that is, the position where the peak signal gradually increases from low intensity. Calculate the time difference from the starting point to the highest point of the peak, which is the second rising edge width, indicating the time taken for the signal to rise from the beginning to reach the maximum intensity. The time bin difference is the second rising edge width.

[0053] In one embodiment, the first depth value and the second depth value include: the first depth value is a first time bin corresponding to a highest point of a highest peak in a first histogram.

[0054] Specifically, first, obtain the data of the first histogram, which is obtained by sampling the reflected signal, with the horizontal axis being the time bin and the vertical axis being the signal intensity. In the histogram, determine the highest point of the signal intensity, that is, find the position of the peak. The time bin corresponding to the highest point is the first depth value, which indicates the time scale when the reflected signal reaches the maximum intensity. For example, if the maximum signal intensity in the first histogram appears at a time bin of 500, then the depth value corresponding to this time bin is the first depth value. This depth value indicates the time when the laser pulse is reflected to the target during measurement, and it can reflect the distance between the target and the sensor. The corresponding target depth value can be inferred from the time bin position. The second depth value is the second time bin corresponding to the highest point of the highest peak in the second histogram.

[0055] Specifically, by acquiring the data of the second histogram, the histogram is obtained by sampling the reflected signal of the second measurement, the horizontal axis is the time bin, and the vertical axis is the signal intensity. In the second histogram, find the highest point of the signal intensity, that is, the highest peak position. The time bin corresponding to the highest point is the second depth value, which indicates the time scale when the maximum intensity of the reflected signal occurs. For example, if the maximum signal intensity in the second histogram appears at a time bin of 800, then the time value corresponding to this time bin is the second depth value. By calculating the relationship between the time bin position and the speed of light, the actual distance between the reflecting target and the measuring device can be obtained.

[0056] In one embodiment, in step S30, the mathematical model is a linear mathematical model.

[0057] Specifically, by defining the basic goal of error compensation, the error compensation coefficient is calculated by the known measurement error and some measurement parameters (such as the rising edge width). This goal can be achieved through a mathematical model, and a linear mathematical model is used here. The linear mathematical model assumes that the relationship between the error compensation coefficient and the input parameter is linear. It is assumed that the relationship between the error compensation coefficient (y) and the rising edge width (x) can be expressed by a simple linear equation, for example: y = a x + b, where y represents the error compensation coefficient, x represents the rising edge width, a and b are constant coefficients in the linear model, a represents the slope, which represents the strength of the relationship between the error and the rising edge width, and b represents the bias term, which represents the error value when the rising edge width is zero. In order to determine these coefficients (a and b), the scattered data measured at different preset distances during the calibration phase can be used for fitting. The most suitable coefficient value is calculated by the least squares method or other regression analysis methods to obtain the parameters of the linear model. The basic principle of the least squares method is to obtain the best fitting curve by minimizing the sum of squares of the prediction errors, that is, by the formula: Among them, y i is the known error value, x i is the corresponding rising edge width, N is the total number of sample points, and the optimal values ​​of a and b are solved by minimizing the sum of squared errors. After fitting, a linear model is obtained.

[0058] In one embodiment, in step S40, the fitting formula is y=a·x+b, wherein y represents the error value, x represents the rising edge width representing the highest peak, and a and b are error compensation coefficients.

[0059] Specifically, the error value (y) is first defined as the difference between the actual depth value obtained during the measurement process and the theoretical depth value, that is, the degree to which the measured value deviates from the expected result under a certain condition. By recording the relationship between these error values ​​and the corresponding rising edge width (x), the error compensation coefficient can be calculated for each scatter point. The rising edge width (x) usually refers to the time area covered by the rising segment of the highest peak in the histogram of the reflected signal, that is, the time difference from the beginning of the signal rising to the maximum value. By analyzing the relationship between different rising edge widths and error values, a suitable fitting curve can be found for each measurement scenario.

[0060] In one embodiment, the N different preset distances further include: N-1 different preset distances under the assumed distance are short distances, and 1 preset distance outside the assumed distance is a long distance.

[0061] Specifically, first set an assumed distance, which is used as a reference point for measurement. It can be a common measurement distance in a conventional use environment or a standard distance preset during design, such as 1000mm. Then, select multiple other distances close to the assumed distance as close distances. These distances are usually slightly shorter or longer than the assumed distances, and the range can be adjusted according to the specific application and design requirements of the measurement system. For example, assuming that the distance is 1000mm, you can select multiple distances such as 900mm, 950mm, 1050mm, and 100mm as close distances. These distances are usually adjusted by adjusting the working distance setting of the measuring device or gradually adjusted according to the measurement capabilities of the system. Ultimately, all selected close distances should cover the measurement area near the assumed distance. The way to obtain the long distance is usually to adjust the target detection range of the sensor at the hardware level, or to set the target area of ​​the measurement distance in the software.

[0062] In one embodiment, at a long distance, scatter points in the scatter point array that meet preset scatter point conditions are selected as central scatter points, and the remaining scatter points are edge scatter points.

[0063] Specifically, when measuring at a long distance, it is first necessary to extract all scattered point data from the scattered point array. These scattered point data are usually collected by laser ranging sensors or related measuring equipment. Next, these scattered points are screened according to the preset scattered point conditions. The preset scattered point conditions may include, for example, the signal strength, reflectivity, depth value and actual distance error range of the scattered points. For example, a condition is set that requires the signal strength of the scattered points to be greater than a certain threshold (such as 50%), or the error between the depth value of the scattered points and the theoretical depth value is less than a certain range (such as 0.1mm). These scattered points that meet the conditions are the central scattered points. For scattered points that do not meet these conditions, they are usually marked as edge scattered points due to weak signals, poor reflections or large errors. Edge scattered points may receive more environmental noise at a long distance. The screening method is usually implemented through algorithm automation, and by judging and comparing the data of each scattered point, it is finally determined which scattered points meet the conditions and are retained.

[0064] In one embodiment, for edge scatter, see Figure 5 First, remove the first error value and the first rising edge width at a long distance, and then input the N-1 first error values ​​and N-1 first rising edge widths of the edge scatter points at a close distance into the mathematical model to generate a fitting formula, and calculate the error compensation coefficient of the edge scatter points through the fitting formula.

[0065] Specifically, for each edge scatter point, its corresponding error value (first error value) and first rising edge width will be obtained in the long-distance measurement, and these values ​​are usually calculated by calculating the characteristic data such as the signal strength and time response curve of the scatter point. These error values ​​and rising edge widths of the edge scatter points at a long distance do not meet the preset conditions, so these data need to be removed at this stage, that is, they are not used in the subsequent fitting process. The removal method can be to exclude the data of these scatter points that do not meet the conditions from the error compensation calculation by a simple data screening or marking method. Then, the error values ​​and rising edge widths of these edge scatter points at a close distance are extracted, and these close distances are usually multiple distance values ​​near the assumed distance (for example, 900mm, 950mm, 1050mm, etc.). These error values ​​and rising edge width data measured at close distances are relatively more reliable because they usually meet the preset scatter point conditions and are less affected by measurement errors and signal noise. By extracting N-1 error values ​​and N-1 rising edge widths of edge scatter points at a close distance, a set of data for calculating the fitting model is formed. These data will be used as input to generate a fitting formula through a mathematical model, usually through polynomial fitting or linear fitting. The purpose of this formula is to derive the error compensation coefficient of the edge scatter at a long distance based on the error value and the rising edge width. The process of generating the fitting formula usually includes: firstly, fitting the data of N-1 error values ​​and the rising edge width, and calculating the fitting coefficients (such as a, b, etc.), which characterize the relationship between the error and the rising edge width. The generated fitting formula will be applied to the measurement data of the edge scatter, and finally the error compensation coefficient of the edge scatter will be obtained.

[0066] In one embodiment, see Figure 4 , the selection methods of the central scatter point include: S1212: Taking the current scattered point as the center, select a 6×6 area as the scattered point area to be measured.

[0067] Specifically, firstly, the current scattered point is selected as the center point in the measurement area, and then a 6×6 area is selected around the center point, which contains 6 rows and 6 columns, a total of 36 scattered points. The selection of the area is achieved by determining the coordinates of the current scattered point in the two-dimensional coordinate system and extending it by a unit length in the upper, lower, left and right directions to determine the boundary of the area.

[0068] S1213: Obtain third depth values ​​of m scattered points in the scattered point area to be measured.

[0069] Specifically, in the selected 6×6 area, a depth sensor or measuring tool is used to measure each scattered point, and the depth information of these scattered points during the measurement process is obtained by obtaining the depth value of each scattered point in the area. These depth values ​​are calculated based on the round-trip time of the reflected signal. The reflected signal is sent and received by laser or other sensing technology. The sensor measures the flight time of the light signal and calculates the distance from each scattered point to the target object based on the flight time. The depth value calculation formula for each scattered point is depth value = (speed of light × round-trip time) / 2, and the depth value obtained is the third depth value of the scattered point.

[0070] S1214: Calculate m differences between the third depth values ​​of the m scattered points and the depth values ​​corresponding to the true distances.

[0071] Specifically, after obtaining the third depth value of each scattered point in the 6×6 area, it is first necessary to obtain the true distance of each scattered point at the position, that is, the predetermined distance between these scattered points and the real reference target. The real distance can be obtained by manual measurement or a pre-calibrated standard distance. For example, the real standard distance may be obtained by the known position of the reference target and the data of the known calibration instrument. Then, for each scattered point, the error value of the scattered point is obtained using the calculation formula "error = third depth value - true distance". This process is performed once for each scattered point in the area to obtain m difference values.

[0072] S1215: Count the number of m scattered points whose difference is less than 40 mm. If the number of scattered points is greater than 6, the current scattered point is considered to be the central scattered point.

[0073] Specifically, for the m error values ​​obtained, a conditional judgment method is used to filter out all scatter points with error values ​​less than 40mm. Specifically, all error values ​​are checked one by one to determine whether they meet the condition of "error value <40mm". If the error is less than 40mm, the scatter point is considered to be a central scatter point. This process is achieved by looping through each difference and comparing them. Finally, the number of central scatter points that meet the conditions is counted. If the number of central scatter points is greater than 6, it means that the current scatter point belongs to the central scatter point and meets the subsequent processing conditions. If the number of central scatter points is less than 6, it is considered that the current scatter point does not meet the central scatter point condition and needs to be excluded.

[0074] In one embodiment, the present application also provides a scattered point DTOF chip, see Figure 6 , including a transmitter, a scatter array, a histogram module, and a controller, wherein the controller executes any of the methods described above.

[0075] A transmitter, used for emitting laser light; A scattered point array is used to receive laser signals reflected from the target object; A histogram module, used for performing histogram processing on the received signal to obtain histogram data; The controller is used to obtain a second histogram of each scattered point during real-time distance measurement, and determine the second rising edge width and the second depth value of the highest peak according to the second histogram; input the error compensation coefficient of each scattered point and the width of the second rising edge into the fitting formula to obtain the second error value of each scattered point at the current distance; and add the second error value to the second depth value to obtain the final measured depth.

[0076] The controller is further used to obtain N first depth values ​​and N first histograms of each scatter point at N different preset distances; for each scatter point, according to one of the first histograms, calculate the first rising edge width of the highest peak, obtain the first preset depth value according to the first preset distance, thereby calculating the first error value between the first depth value and the first preset depth value, thereby obtaining the N first error values ​​and the N first rising edge widths of each scatter point; input the N first error values ​​and the N first rising edge widths of each scatter point into a mathematical model to generate a fitting formula, calculate the error compensation coefficient of each scatter point through the fitting formula, and save the error compensation coefficient in the chip.

[0077] The present application also provides an electronic device, including the scattered-point DTOF chip as described above.

[0078] In summary, the present application includes at least one of the following beneficial technical effects: 1. By adopting the above technical solution, the error compensation coefficient of each scattered point can be obtained by fitting the depth error value of each scattered point and the rising edge width of the highest peak of the histogram, thereby effectively correcting the errors caused by signal waveform distortion or environmental changes in the ranging process. By combining the error compensation coefficient with real-time measurement data, the system can accurately correct the depth value of each scattered point and achieve the final high-precision measurement, significantly improving the ranging accuracy and system stability, especially when facing different ranging distances, environmental changes and other factors, it can still maintain excellent performance and ensure the accuracy of the measurement results. In addition, the error compensation coefficient storage method in the solution enables the system to have better adaptability, and can automatically adjust to different ranging conditions, improve the robustness and adaptability of the system, and reduce the need for manual adjustment.

[0079] 2. At long distances, due to the long distance, environmental factors and noise may have a greater impact, and some scattered points may not receive the signal of the reflector, so it is necessary to distinguish between central scattered points and edge scattered points. For edge scattered points, the linear fitting formula is slightly different. The central scattered points that meet the conditions will be used for subsequent error compensation calculations, while the scattered points that do not meet the conditions are judged as edge scattered points. The calculation method of the error compensation coefficient of edge scattered points is slightly different from that of the central scattered points, but in the ranging process, the edge scattered points also need to output depth values, and depth error compensation is required.

[0080] 3. The central scattered points can be effectively screened out during the measurement process to further improve the measurement accuracy. First, with the current scattered point as the center, a 6×6 area is selected as the scattered point area to be measured, so that more data can be collected in a smaller range to more accurately evaluate the effectiveness of the current scattered point. Then, by obtaining the third depth value of the m scattered points in the area, the specific measurement data of the scattered points in the area can be obtained. Subsequently, the difference between the third depth value of these scattered points and the depth value corresponding to the true distance is calculated to obtain m differences, which are used to evaluate the measurement error of each scattered point. Finally, by counting the number of scattered points with a difference less than 40 mm, if the number of central scattered points that meet the conditions exceeds 6, the current scattered point is considered to be the central scattered point, thereby ensuring that the screened scattered points can meet the requirements of high-precision measurement.

[0081] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A scattered point DTOF chip error compensation method, characterized in that: include: Under N different preset distances, obtain N first depth values ​​and N first histograms of each scattered point; For each of the scattered points, according to one of the first histograms, a first rising edge width of the highest peak is calculated, and a first preset depth value is obtained according to a first preset distance, thereby calculating a first error value between the first depth value and the first preset depth value, thereby obtaining the N first error values ​​and the N first rising edge widths of each of the scattered points; Inputting the N first error values ​​and the N first rising edge widths of each scattered point into a mathematical model to generate a fitting formula, calculating the error compensation coefficient of each scattered point by the fitting formula, and storing the error compensation coefficient in the chip; In the real-time distance measurement process, a second histogram of each scattered point is obtained, and a second rising edge width and a second depth value of the highest peak are determined according to the second histogram; Inputting the error compensation coefficient of each scattered point and the width of the second rising edge into the fitting formula to obtain a second error value of each scattered point at the current distance; The second error value is added to the second depth value to obtain a final measured depth.

2. The scattered point DTOF chip error compensation method according to claim 1, characterized in that: The obtaining of N first depth values ​​and N first histograms of each scattered point at N different preset distances includes: At a preset distance, obtaining a first histogram of each scatter point; Based on the first histogram, a highest peak is obtained, and according to the highest peak, a first depth value is obtained; After obtaining the one first depth value, the preset distance is changed to obtain the N first depth values ​​and the N first histograms of each scattered point.

3. The scattered point DTOF chip error compensation method according to claim 1, characterized in that: The first rising edge width and the second rising edge width include: The first rising edge width is obtained by calculating the time bin difference between the highest point of the highest peak in the first histogram and the starting point of the highest peak; The second rising edge width is obtained by calculating the time bin difference between the highest point of the highest peak in the second histogram and the starting point of the highest peak.

4. The scattered point DTOF chip error compensation method according to claim 1, characterized in that: The first depth value and the second depth value include: The first depth value is a first time bin corresponding to the highest point of the highest peak in the first histogram; The second depth value is a second time bin corresponding to the highest point of the highest peak in the second histogram.

5. The scattered point DTOF chip error compensation method according to claim 1, characterized in that: The mathematical model includes: the mathematical model is a linear mathematical model.

6. The scattered point DTOF chip error compensation method according to claim 1, characterized in that: The fitting formula includes: the fitting formula is y=a·x+b, wherein y is a characterizing error value, x is a rising edge width characterizing the highest peak, and a and b are the error compensation coefficients.

7. The scattered point DTOF chip error compensation method according to claim 1, characterized in that: The N different preset distances also include: N-1 different preset distances under the assumed distance are close distances; A preset distance outside the assumed distance is a long distance.

8. The scattered point DTOF chip error compensation method according to claim 7, characterized in that: The method at the long distance includes: at the long distance, selecting the scattered points in the scattered point array that meet the preset scattered point conditions as the central scattered points, and the remaining scattered points as the edge scattered points.

9. The scattered point DTOF chip error compensation method according to claim 8, characterized in that: The edge scatter includes: for the edge scatter, first remove the first error value and the first rising edge width at the long distance, then input N-1 first error values ​​and N-1 first rising edge widths of the edge scatter at the close distance into the mathematical model, generate a fitting formula, and calculate the error compensation coefficient of the edge scatter by the fitting formula.

10. The scattered point DTOF chip error compensation method according to claim 8, characterized in that: The method for selecting the central scattered point includes: Taking the current scattered point as the center, select a 6×6 area as the scattered point area to be measured; Acquire third depth values ​​of m scattered points in the scattered point area to be measured; Calculating m differences between the third depth values ​​of the m scattered points and the depth values ​​corresponding to the true distances; The number of the m scattered points whose difference values ​​are less than 40 mm is counted, and if the number of the scattered points is greater than 6, the current scattered point is considered to be the central scattered point.

11. A scattered point DTOF chip, characterized in that: It comprises a transmitter, a scatter array, a histogram module, and a controller, and the controller executes the method as claimed in any one of claims 1 to 10.

12. An electronic device, characterized in that: Comprising the scattered-point DTOF chip as described in claim 11.

Images