Configuration method and device of three-dimensional structured light distance measuring equipment

Abstract

The invention relates to the technical field of optical equipment, in particular to a configuration method and device of three-dimensional structured light distance measuring equipment. Performing characteristic analysis on the predicted dynamic environment characteristic data of each sub-configuration region to obtain a class of candidate sub-configuration regions; screening and analyzing the first-class candidate sub-configuration regions according to the three-dimensional model diagram of the topographic region to obtain second-class candidate sub-configuration regions; screening and analyzing the second-class candidate sub-configuration regions according to the scene three-dimensional model graph to obtain third-class candidate sub-configuration regions; and obtaining the bit error rate of each of the three types of candidate sub-configuration areas, screening and analyzing the three types of candidate sub-configuration areas according to the bit error rates, and obtaining a final configuration area of the three-dimensional structured light distance measurement equipment. According to the method, the adaptive capacity of the three-dimensional structured light distance measurement equipment to environment change can be enhanced, the reliability of a measurement result is improved, and the equipment configuration cost is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of optical equipment, and in particular to a configuration method and device for a three-dimensional structured light ranging device. Background Art

[0002] With the rapid development of industrial automation, intelligent manufacturing and other fields, the demand for three-dimensional structured light ranging equipment is growing. Three-dimensional structured light ranging equipment can quickly and accurately obtain three-dimensional information of target objects and is widely used in industrial inspection, robot vision, reverse engineering and other fields. However, in practical applications, the performance of three-dimensional structured light ranging equipment is greatly affected by environmental factors, such as lighting conditions, temperature, humidity, wireless signal interference, etc. How to choose a suitable equipment configuration area to ensure that the equipment works stably and reliably in a complex and changing environment has become a key issue that needs to be solved urgently.

[0003] Traditional equipment configuration methods usually rely on manual experience and on-site surveys, which are inefficient, costly, and difficult to cope with complex and changing environments. For example, selecting the configuration area based solely on experience may not fully consider the impact of environmental factors on equipment performance, resulting in unstable equipment operation or reduced measurement accuracy. Relying solely on on-site surveys consumes a lot of time and manpower and is difficult to meet the needs of rapid deployment. Therefore, there is an urgent need for a method that can intelligently and automatically select the configuration area of ​​3D structured light ranging equipment to improve the stability and reliability of the equipment and reduce deployment costs. Summary of the invention

[0004] The present invention overcomes the shortcomings of the prior art and provides a configuration method and device for a three-dimensional structured light ranging device.

[0005] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is:

[0006] The present invention discloses a configuration method of a three-dimensional structured light ranging device, comprising the following steps:

[0007] Acquire a target configuration area of ​​a three-dimensional structured light ranging device, and divide the target configuration area into a plurality of sub-configuration areas;

[0008] Acquire dynamic environmental feature data of each sub-configuration area within a preset historical time period, construct a long short-term memory network prediction model, and train the long short-term memory network prediction model based on the dynamic environmental feature data to obtain a trained long short-term memory network prediction model;

[0009] According to the trained long short-term memory network prediction model, the predicted dynamic environment characteristic data of each sub-configuration area within a preset working time period is obtained; the predicted dynamic environment characteristic data of each sub-configuration area is analyzed to obtain a class of candidate sub-configuration areas;

[0010] Obtaining a three-dimensional model map of the terrain area of ​​each first-category candidate sub-configuration area, and screening and analyzing each first-category candidate sub-configuration area according to the three-dimensional model map of the terrain area to obtain a second-category candidate sub-configuration area;

[0011] Obtaining a three-dimensional model diagram of the scene between each of the second-category candidate sub-configuration regions and the position area of ​​the object to be measured, and screening and analyzing the second-category candidate sub-configuration regions according to the three-dimensional model diagram of the scene to obtain three-category candidate sub-configuration regions;

[0012] The bit error rates of the three types of candidate sub-configuration areas are obtained, and the three types of candidate sub-configuration areas are screened and analyzed according to the bit error rates to obtain the final configuration area of ​​the three-dimensional structured light ranging device.

[0013] Preferably, the dynamic environmental feature data of each sub-configuration area within a preset historical time period is obtained, a long short-term memory network prediction model is constructed, and the long short-term memory network prediction model is trained based on the dynamic environmental feature data to obtain a trained long short-term memory network prediction model, specifically:

[0014] Acquire monitoring logs of each environmental monitoring device in the sub-configuration area, and acquire dynamic environmental characteristic data of each sub-configuration area within a preset historical time period according to the monitoring logs;

[0015] Calculate the LOF value of the dynamic environment feature data in each sub-configuration area; filter out the dynamic environment feature data with a LOF value greater than a preset threshold to obtain the denoised dynamic environment feature data;

[0016] The denoised dynamic environment feature data is sorted based on the time series to obtain the dynamic environment feature data of each sub-configuration area based on the time series;

[0017] Constructing a long short-term memory network prediction model, and importing the time-series dynamic environmental feature data of each sub-configuration area into the long short-term memory network prediction model for training, and obtaining the training parameters of the model after each training process is completed;

[0018] When the training parameters of the model meet the preset requirements, the last training parameters are used as the final parameters of the model, and the trained long short-term memory network prediction model is output.

[0019] Preferably, the predicted dynamic environment characteristic data of each sub-configuration area within a preset working time period is obtained according to the trained long short-term memory network prediction model; the predicted dynamic environment characteristic data of each sub-configuration area is analyzed for characteristics to obtain a class of candidate sub-configuration areas, specifically:

[0020] Obtaining a preset working time period of a three-dimensional structured light ranging device, and importing the preset working time period into the trained long short-term memory network prediction model to perform prediction and obtain predicted dynamic environment feature data of each sub-configuration area within the preset working time period;

[0021] Acquiring performance characteristic information of the three-dimensional structured light ranging device, and determining optimal dynamic environment characteristic data of the three-dimensional structured light ranging device during operation according to the performance characteristic information;

[0022] Calculating the Pearson correlation coefficient between the predicted dynamic environment characteristic data of each sub-configuration area within the preset working time period and the optimal dynamic environment characteristic data;

[0023] The sub-configuration regions corresponding to the Pearson correlation coefficient being greater than the preset coefficient threshold are marked as a type of candidate sub-configuration regions; and the sub-configuration regions corresponding to the Pearson correlation coefficient being not greater than the preset coefficient threshold are marked as non-sub-configuration regions.

[0024] Preferably, a three-dimensional model map of the terrain area of ​​each first-class candidate sub-configuration area is obtained, and each first-class candidate sub-configuration area is screened and analyzed according to the three-dimensional model map of the terrain area to obtain the second-class candidate sub-configuration area, specifically:

[0025] Acquire terrain region image information of each first-class candidate sub-configuration region, and construct a terrain region three-dimensional model map of each first-class candidate sub-configuration region according to the terrain region image information;

[0026] Obtain a device instruction manual of the three-dimensional structured light ranging device, and obtain a three-dimensional model map of the standard installation terrain of the three-dimensional structured light ranging device according to the device instruction manual;

[0027] Constructing a plurality of virtual spaces, and importing the three-dimensional model images of the terrain regions of each type of candidate sub-configuration region into the virtual spaces respectively, and importing the three-dimensional model images of the standard installation terrain into the virtual spaces;

[0028] Iteratively registering the three-dimensional model map of the terrain area in each virtual space with the three-dimensional model map of the standard installation terrain based on an iterative closest point matching algorithm until the iterative registration process is completed;

[0029] Obtaining the volume value of the intersection model between the terrain area three-dimensional model map and the standard installation terrain three-dimensional model map in each virtual space, and obtaining the volume value of the non-intersection model between the terrain area three-dimensional model map and the standard installation terrain three-dimensional model map in each virtual space;

[0030] Calculating the degree of intersection between the three-dimensional model diagram of the terrain area in each virtual space and the three-dimensional model diagram of the standard installation terrain according to the volume value of the intersection model and the volume value of the non-intersection model between the three-dimensional model diagram of the terrain area in each virtual space and the three-dimensional model diagram of the standard installation terrain;

[0031] A type of candidate subconfiguration region corresponding to a degree of intersection greater than a preset degree of intersection is marked as a type of candidate subconfiguration region; a type of candidate subconfiguration region corresponding to a degree of intersection not greater than the preset degree of intersection is marked as a non-subconfigurable region.

[0032] Preferably, a scene 3D model diagram between each of the second-category candidate sub-configuration areas and the position area of ​​the object to be detected is obtained, and the second-category candidate sub-configuration areas are screened and analyzed according to the scene 3D model diagram to obtain three-category candidate sub-configuration areas, specifically:

[0033] Acquire the location area of ​​the object to be measured, and acquire scene image information between each of the second-category candidate sub-configuration areas and the location area of ​​the object to be measured;

[0034] Constructing a three-dimensional model diagram of the scene between each of the two types of candidate sub-configuration areas and the position area of ​​the object to be measured according to the corresponding scene image information;

[0035] Acquire the external dimension information of the three-dimensional structured light ranging device, and construct a three-dimensional model diagram of the three-dimensional structured light ranging device according to the external dimension information;

[0036] The three-dimensional model diagram of the device is simulated and installed on the two types of candidate sub-configuration areas of the three-dimensional model diagram of each scene; and the three-dimensional structured light ranging device on the two types of candidate sub-configuration areas is simulated and analyzed in the three-dimensional model diagram of the scene;

[0037] Determine whether there is ranging interference between the three-dimensional structured light ranging device on each of the second-category candidate sub-configuration areas and the object to be measured;

[0038] The second type of candidate sub-configuration area without ranging interference is marked as the third type of candidate sub-configuration area; the second type of candidate sub-configuration area with ranging interference is marked as a non-sub-configurable area.

[0039] Preferably, the bit error rate of each of the three types of candidate sub-configuration areas is obtained, and the three types of candidate sub-configuration areas are screened and analyzed according to the bit error rate to obtain the final configuration area of ​​the three-dimensional structured light ranging device, specifically:

[0040] Obtain the position nodes of the three types of candidate sub-configuration areas, and calibrate the position nodes of the three types of candidate sub-configuration areas as the signal transmission start position nodes; obtain the remote signal receiving node of the three-dimensional structured light ranging device, and calibrate the remote signal receiving node of the three-dimensional structured light ranging device as the signal transmission end position node;

[0041] Acquire predicted dynamic environment feature data of each of the three types of candidate sub-configuration areas within a preset working time period, and determine the signal spectrum distortion rate of the three-dimensional structured light ranging device in each of the three types of candidate sub-configuration areas when working within the preset working time period according to the predicted dynamic environment feature data of each of the three types of candidate sub-configuration areas within the preset working time period;

[0042] And obtaining the distance value between each signal transmission starting position node and the signal transmission end position node, and obtaining the distance value between the three-dimensional structured light ranging device and the remote signal receiving node in each of the three types of candidate sub-configuration areas;

[0043] Determine the bit error rate of the three-dimensional structured light ranging device in each of the three types of candidate sub-configuration areas during operation according to the distance value between the three-dimensional structured light ranging device and the remote signal receiving node in each of the three types of candidate sub-configuration areas and the signal spectrum distortion rate when working in a preset working time period;

[0044] The three types of candidate sub-configuration areas with the lowest bit error rates are selected and calibrated as the final configuration areas of the three-dimensional structured light ranging device.

[0045] The present invention also discloses a configuration device for a three-dimensional structured light ranging device, the configuration device for the three-dimensional structured light ranging device comprising a memory and a processor, the memory storing a configuration method program for the three-dimensional structured light ranging device, and when the configuration method program for the three-dimensional structured light ranging device is executed by the processor, any one of the configuration method steps for the three-dimensional structured light ranging device is implemented.

[0046] The present invention solves the technical defects existing in the background technology, and has the following beneficial effects: preliminary screening is performed based on predicted dynamic environment feature data, and then further screening is performed in combination with factors such as terrain, scene and bit error rate. This hierarchical screening mechanism can reduce unnecessary calculation and analysis workload. The configuration area finally determined is the area most suitable for the device to perform ranging work after comprehensive consideration of various factors. The device can complete the ranging task more efficiently by configuring in this area, enhance the device's adaptability to environmental changes, improve the reliability of measurement results, and effectively reduce the device configuration cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0048] Figure 1 It is an overall method flow chart of a configuration method of a three-dimensional structured light ranging device;

[0049] Figure 2 A partial method flow chart of a configuration method of a three-dimensional structured light ranging device;

[0050] Figure 3 The present invention is a schematic diagram of the structure of a configuration device of a three-dimensional structured light ranging device. DETAILED DESCRIPTION

[0051] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0052] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0053] The present invention discloses a configuration method of a three-dimensional structured light ranging device, such as Figure 1 As shown, the following steps are included:

[0054] S102: Acquire a target configuration area of ​​a three-dimensional structured light ranging device, and divide the target configuration area into a plurality of sub-configuration areas;

[0055] S104: acquiring dynamic environmental feature data of each sub-configuration area within a preset historical time period, constructing a long short-term memory network prediction model, and training the long short-term memory network prediction model based on the dynamic environmental feature data to obtain a trained long short-term memory network prediction model;

[0056] S106: obtaining predicted dynamic environment feature data of each sub-configuration area within a preset working time period according to the trained long short-term memory network prediction model; performing characteristic analysis on the predicted dynamic environment feature data of each sub-configuration area to obtain a class of candidate sub-configuration areas;

[0057] S108: Obtain a three-dimensional model map of the terrain area of ​​each first-category candidate sub-configuration area, and screen and analyze each first-category candidate sub-configuration area according to the three-dimensional model map of the terrain area to obtain a second-category candidate sub-configuration area;

[0058] S110: Obtain a scene 3D model diagram between each of the second-category candidate sub-configuration regions and the location area of ​​the object to be detected, and screen and analyze the second-category candidate sub-configuration regions according to the scene 3D model diagram to obtain three-category candidate sub-configuration regions;

[0059] S112: Obtaining bit error rates of the three types of candidate sub-configuration areas, screening and analyzing the three types of candidate sub-configuration areas according to the bit error rates, and obtaining a final configuration area of ​​the three-dimensional structured light ranging device.

[0060] It should be noted that the target configuration area is the range of concern when the device is configured, which is determined according to the working scenario, task requirements, etc. of the device. For each sub-configuration area, its dynamic environment feature data within a preset historical time period is obtained. The dynamic environment feature data includes changes in light intensity, movement of objects, fluctuations in ambient temperature, changes in magnetic field intensity, etc., all of which may affect structured light ranging. The predicted dynamic environment feature data of each sub-configuration area is analyzed for characteristics. This characteristic analysis may include analysis of data stability, change trends, etc. For example, if the predicted light intensity of a sub-configuration area changes very drastically, it may affect the accuracy of structured light ranging. Through characteristic analysis, a class of candidate sub-configuration areas is obtained, which are relatively stable or meet certain requirements in terms of predicted dynamic environment characteristics. For each class of candidate sub-configuration areas, its terrain area three-dimensional model map is obtained, and this terrain area three-dimensional model map contains information such as terrain undulations and obstacle distribution in the sub-configuration area. Then, the first-class candidate sub-configuration areas are screened and analyzed according to the three-dimensional model map of the terrain area. For example, if there are a large number of towering obstacles in a certain first-class candidate sub-configuration area, the propagation of the structured light may be blocked, affecting the ranging effect. Through this screening, these unfavorable areas can be excluded, thereby obtaining the second-class candidate sub-configuration areas. For each second-class candidate sub-configuration area, the scene three-dimensional model map between it and the position area of ​​the object to be measured is obtained. This scene three-dimensional model map can reflect the spatial relationship between the candidate sub-configuration area and the object to be measured, including whether there is occlusion, the distance between the two, and other information. The second-class candidate sub-configuration areas are screened and analyzed according to the scene three-dimensional model map. For example, if there are many occlusions between a certain second-class candidate sub-configuration area and the object to be measured, the structured light may not accurately reach the object to be measured, thereby affecting the ranging result. Through this screening, the third-class candidate sub-configuration area is obtained. For each third-class candidate sub-configuration area, its bit error rate is obtained. The bit error rate is an indicator to measure the accuracy of data transmission. Here, it may be related to signal transmission, data processing, and other links in the structured light ranging process. If the bit error rate in a certain area is high, it means that there may be more data errors when measuring distance in this area. The three types of candidate sub-configuration areas are screened and analyzed according to the bit error rate. The area with a lower bit error rate is selected as the final configuration area of ​​the 3D structured light ranging device, which can ensure that the device can measure distance more accurately in this area.

[0061] In summary, preliminary screening is performed based on the predicted dynamic environmental feature data, and then further screening is performed in combination with factors such as terrain, scene, and bit error rate. This hierarchical screening mechanism can reduce unnecessary calculation and analysis workload. The configuration area finally determined is the area that is most suitable for the device to perform ranging work after comprehensive consideration of multiple factors. The device can be configured in this area to complete the ranging task more efficiently, enhance the device's ability to adapt to environmental changes, improve the reliability of measurement results, and effectively reduce the device configuration cost.

[0062] Preferably, the dynamic environmental feature data of each sub-configuration area within a preset historical time period is obtained, a long short-term memory network prediction model is constructed, and the long short-term memory network prediction model is trained based on the dynamic environmental feature data to obtain a trained long short-term memory network prediction model, such as Figure 2 As shown, specifically:

[0063] S202: Acquire monitoring logs of each environmental monitoring device in the sub-configuration area, and acquire dynamic environmental characteristic data of each sub-configuration area within a preset historical time period according to the monitoring logs;

[0064] S204: Calculate the LOF value of the dynamic environment feature data in each sub-configuration area; filter out the dynamic environment feature data with a LOF value greater than a preset threshold to obtain denoised dynamic environment feature data;

[0065] S206: performing time series sorting processing on the denoised dynamic environment feature data to obtain time series-based dynamic environment feature data of each sub-configuration area;

[0066] S208: constructing a long short-term memory network prediction model, and importing the time-series dynamic environment feature data of each sub-configuration area into the long short-term memory network prediction model for training, and obtaining the training parameters of the model after each training process is completed;

[0067] S210: When the training parameters of the model meet the preset requirements, the last training parameters are used as the final parameters of the model, and the trained long short-term memory network prediction model is output.

[0068] It should be noted that the monitoring log is used to record various environmental information in the sub-configuration area. According to the monitoring log, the dynamic environmental feature data of each sub-configuration area in the preset historical time period is obtained. For example, the log of the light sensor can provide the data of the change of light intensity over time, and the log of the object motion detector can provide the data of the movement trajectory of the object in the area, etc. These data constitute the dynamic environmental feature data. The local anomaly factor (LOF) value of the dynamic environmental feature data in each sub-configuration area is calculated. LOF is an algorithm for identifying local anomalies in a data set. For each data point, it determines whether it is an anomaly by comparing the density of the point with its neighboring points. In the context of dynamic environmental feature data, there may be some abnormal data points caused by sensor failure, sudden interference, etc. For example, the light sensor may record an extremely high or extremely low light intensity value due to instantaneous electromagnetic interference. This value is inconsistent with the normal light intensity change trend around. Such anomalies can be identified by calculating the LOF value. The dynamic environmental feature data with a LOF value greater than a preset threshold is filtered out. This preset threshold is set according to the characteristics of the data and the tolerance to anomalies. By filtering out these abnormal data, we can obtain the denoised dynamic environmental feature data, making the data used for subsequent model training purer and more able to reflect the real changing rules of environmental features. Construct a long short-term memory network prediction model (LSTM prediction model). The long short-term memory network prediction model consists of an input gate, a forget gate, and an output gate. It can effectively handle long-term dependencies when processing time series data. For example, when processing dynamic environmental feature data, it can predict future environmental states based on past information such as light intensity and object movement. Import the time-series dynamic environmental feature data of each sub-configuration area into the LSTM prediction model for training. During the training process, the model adjusts its own weights and other parameters based on the input data to minimize the error between the predicted results and the actual results. After each training process is completed, the training parameters of the model are obtained. These training parameters include weights, biases, etc., which reflect the performance of the model in the current training state.

[0069] Preferably, the predicted dynamic environment characteristic data of each sub-configuration area within a preset working time period is obtained according to the trained long short-term memory network prediction model; the predicted dynamic environment characteristic data of each sub-configuration area is analyzed for characteristics to obtain a class of candidate sub-configuration areas, specifically:

[0070] Obtaining a preset working time period of a three-dimensional structured light ranging device, and importing the preset working time period into the trained long short-term memory network prediction model to perform prediction and obtain predicted dynamic environment feature data of each sub-configuration area within the preset working time period;

[0071] Acquiring performance characteristic information of the three-dimensional structured light ranging device, and determining optimal dynamic environment characteristic data of the three-dimensional structured light ranging device during operation according to the performance characteristic information;

[0072] Calculating the Pearson correlation coefficient between the predicted dynamic environment characteristic data of each sub-configuration area within the preset working time period and the optimal dynamic environment characteristic data;

[0073] The sub-configuration regions corresponding to the Pearson correlation coefficient being greater than the preset coefficient threshold are marked as a type of candidate sub-configuration regions; and the sub-configuration regions corresponding to the Pearson correlation coefficient being not greater than the preset coefficient threshold are marked as non-sub-configuration regions.

[0074] It should be noted that the performance characteristic information of the three-dimensional structured light ranging device includes information on the ranging accuracy, signal receiving capability, anti-interference capability, etc. of the device under different environmental conditions. For example, the device may be able to achieve the best ranging effect within a specific range of light intensity and temperature range. The optimal dynamic environmental characteristic data of the three-dimensional structured light ranging device during operation is determined based on the performance characteristic information. For example, if the device has the highest ranging accuracy when the light intensity is 500 to 1000 lux and the temperature is 20 to 25 degrees Celsius, then the environmental characteristic data corresponding to this light intensity and temperature range can be determined as the optimal dynamic environmental characteristic data. The Pearson correlation coefficient is a statistical indicator that measures the degree of linear correlation between two variables, and its value range is between -1 and 1. The sub-configuration area corresponding to the Pearson correlation coefficient greater than the preset coefficient threshold is marked as a class of candidate sub-configuration areas, which means that the predicted environmental characteristic data of these sub-configuration areas have a high correlation with the optimal environmental characteristic data of the device, and structured light ranging in these areas may have a better effect. The sub-configuration areas corresponding to the Pearson correlation coefficient not greater than the preset coefficient threshold are marked as non-sub-configuration areas. The environmental characteristics of these areas are quite different from the optimal environmental characteristics of the device and are not suitable for the configuration of the structured light ranging device. Quantitative analysis makes the screening process more scientific and accurate, which helps to find the sub-configuration area that is most conducive to the operation of the device and further optimize the performance of the device.

[0075] Preferably, a three-dimensional model map of the terrain area of ​​each first-class candidate sub-configuration area is obtained, and each first-class candidate sub-configuration area is screened and analyzed according to the three-dimensional model map of the terrain area to obtain the second-class candidate sub-configuration area, specifically:

[0076] Acquire terrain region image information of each first-class candidate sub-configuration region, and construct a terrain region three-dimensional model map of each first-class candidate sub-configuration region according to the terrain region image information;

[0077] Obtain a device instruction manual of the three-dimensional structured light ranging device, and obtain a three-dimensional model map of the standard installation terrain of the three-dimensional structured light ranging device according to the device instruction manual;

[0078] Constructing a plurality of virtual spaces, and importing the three-dimensional model images of the terrain regions of each type of candidate sub-configuration region into the virtual spaces respectively, and importing the three-dimensional model images of the standard installation terrain into the virtual spaces;

[0079] Iteratively registering the three-dimensional model map of the terrain area in each virtual space with the three-dimensional model map of the standard installation terrain based on an iterative closest point matching algorithm until the iterative registration process is completed;

[0080] Obtaining the volume value of the intersection model between the terrain area three-dimensional model map and the standard installation terrain three-dimensional model map in each virtual space, and obtaining the volume value of the non-intersection model between the terrain area three-dimensional model map and the standard installation terrain three-dimensional model map in each virtual space;

[0081] Calculating the degree of intersection between the three-dimensional model diagram of the terrain area in each virtual space and the three-dimensional model diagram of the standard installation terrain according to the volume value of the intersection model and the volume value of the non-intersection model between the three-dimensional model diagram of the terrain area in each virtual space and the three-dimensional model diagram of the standard installation terrain;

[0082] A type of candidate subconfiguration region corresponding to a degree of intersection greater than a preset degree of intersection is marked as a type of candidate subconfiguration region; a type of candidate subconfiguration region corresponding to a degree of intersection not greater than the preset degree of intersection is marked as a non-subconfigurable region.

[0083] It should be noted that the three-dimensional model map of the terrain area in each virtual space is iteratively registered with the three-dimensional model map of the standard installation terrain based on the iterative closest point (ICP) matching algorithm. The ICP algorithm is an algorithm for matching two point clouds (which can be regarded as point sets in a three-dimensional model here). During the iteration process, the position, posture and other parameters of the three-dimensional model map of the terrain area are continuously adjusted to make it match the three-dimensional model map of the standard installation terrain as much as possible. This process will continue until the iterative registration process is completed, that is, a certain convergence condition is reached, such as the matching error between the models is less than a certain threshold. The first type of candidate sub-configuration area corresponding to the intersection degree greater than the preset intersection degree is marked as the second type of candidate sub-configuration area, which means that the terrain conditions in these areas are similar to the standard installation terrain conditions of the equipment, and the installation of the equipment in these areas may have better performance. The first type of candidate sub-configuration area corresponding to the intersection degree not greater than the preset intersection degree is marked as a non-sub-configurable area. The terrain conditions in these areas are quite different from the standard installation terrain conditions of the equipment, and it is not suitable for installing a three-dimensional structured light ranging device.

[0084] In summary, by constructing a 3D model map of the terrain area and accurately matching it with the 3D model map of the standard installation terrain (through the ICP algorithm), it is possible to accurately evaluate whether the terrain of each candidate sub-configuration area is suitable for equipment installation. This precise analysis based on terrain avoids the problem of equipment performance degradation caused by terrain mismatch. For example, if the equipment needs to be installed on relatively flat terrain, this method can screen out those sub-configuration areas with sufficiently flat terrain.

[0085] Screening is performed based on the standard installation terrain 3D model diagram in the equipment manual to ensure that the equipment can work on terrain that meets its installation requirements within the selected Class II candidate sub-configuration area. When the equipment is installed on a suitable terrain, it can better play its ranging function, reduce errors caused by terrain factors, and improve the working accuracy and stability of the equipment.

[0086] A class of candidate sub-configuration areas with a degree of intersection not greater than a preset degree of intersection are marked as non-sub-configuration areas, effectively excluding areas where terrain conditions are not conducive to device operation. This avoids a series of problems that may arise from installing equipment on unsuitable terrain, such as ranging deviations caused by equipment tilt, blocked signal transmission, etc., thereby optimizing the overall working performance of the equipment.

[0087] Preferably, a scene 3D model diagram between each of the second-category candidate sub-configuration areas and the position area of ​​the object to be detected is obtained, and the second-category candidate sub-configuration areas are screened and analyzed according to the scene 3D model diagram to obtain three-category candidate sub-configuration areas, specifically:

[0088] Acquire the location area of ​​the object to be measured, and acquire scene image information between each of the second-category candidate sub-configuration areas and the location area of ​​the object to be measured;

[0089] Constructing a three-dimensional model diagram of the scene between each of the two types of candidate sub-configuration areas and the position area of ​​the object to be measured according to the corresponding scene image information;

[0090] Acquire the external dimension information of the three-dimensional structured light ranging device, and construct a three-dimensional model diagram of the three-dimensional structured light ranging device according to the external dimension information;

[0091] The three-dimensional model diagram of the device is simulated and installed on the two types of candidate sub-configuration areas of the three-dimensional model diagram of each scene; and the three-dimensional structured light ranging device on the two types of candidate sub-configuration areas is simulated and analyzed in the three-dimensional model diagram of the scene;

[0092] Determine whether there is ranging interference between the three-dimensional structured light ranging device on each of the second-category candidate sub-configuration areas and the object to be measured;

[0093] The second type of candidate sub-configuration area without ranging interference is marked as the third type of candidate sub-configuration area; the second type of candidate sub-configuration area with ranging interference is marked as a non-sub-configurable area.

[0094] It should be noted that the scene image information between each second-category candidate sub-configuration area and the location area of ​​the object to be measured is obtained. This image information can come from various sensors, such as cameras, laser radars, etc., which can capture the environmental features between the two location areas, including the terrain and obstacles in the middle. A three-dimensional model diagram of the scene between each second-category candidate sub-configuration area and the location area of ​​the object to be measured is constructed according to the corresponding scene image information. This model diagram can fully reflect the environmental features such as terrain and obstacles between the two location areas. The three-dimensional model diagram of the device is simulated and installed on the second-category candidate sub-configuration area of ​​each scene three-dimensional model diagram. Simulation installation refers to placing the device at a specified position in a virtual environment for subsequent analysis. The simulated ranging analysis includes simulating the device to emit structured light, receive reflected light, calculate ranging results, etc., to evaluate the actual ranging effect of the device in the area. Determine whether there is ranging interference between the three-dimensional structured light ranging device on each second-category candidate sub-configuration area and the object to be measured. Ranging interference may come from obstacles in the middle, terrain undulations, etc. These factors may block the propagation of structured light, resulting in inaccurate ranging results. The second-category candidate sub-configuration areas where there is no ranging interference are marked as third-category candidate sub-configuration areas, which means that the terrain and obstacle distribution in these areas are suitable for the device to measure the distance and will not affect the accuracy of the ranging results. The second-category candidate sub-configuration areas where there is ranging interference are marked as non-sub-configurable areas. The terrain and obstacle distribution in these areas may affect the ranging effect of the device and are therefore not suitable for installing the device. This method not only improves the efficiency of device configuration, but also ensures that the device can work under optimal conditions, improves overall work efficiency, and improves the reliability and stability of the device in actual work.

[0095] Preferably, the bit error rate of each of the three types of candidate sub-configuration areas is obtained, and the three types of candidate sub-configuration areas are screened and analyzed according to the bit error rate to obtain the final configuration area of ​​the three-dimensional structured light ranging device, specifically:

[0096] Obtain the position nodes of the three types of candidate sub-configuration areas, and calibrate the position nodes of the three types of candidate sub-configuration areas as the signal transmission start position nodes; obtain the remote signal receiving node of the three-dimensional structured light ranging device, and calibrate the remote signal receiving node of the three-dimensional structured light ranging device as the signal transmission end position node;

[0097] Acquire predicted dynamic environment feature data of each of the three types of candidate sub-configuration areas within a preset working time period, and determine the signal spectrum distortion rate of the three-dimensional structured light ranging device in each of the three types of candidate sub-configuration areas when working within the preset working time period according to the predicted dynamic environment feature data of each of the three types of candidate sub-configuration areas within the preset working time period;

[0098] And obtaining the distance value between each signal transmission starting position node and the signal transmission end position node, and obtaining the distance value between the three-dimensional structured light ranging device and the remote signal receiving node in each of the three types of candidate sub-configuration areas;

[0099] Determine the bit error rate of the three-dimensional structured light ranging device in each of the three types of candidate sub-configuration areas during operation according to the distance value between the three-dimensional structured light ranging device and the remote signal receiving node in each of the three types of candidate sub-configuration areas and the signal spectrum distortion rate when working in a preset working time period;

[0100] The three types of candidate sub-configuration areas with the lowest bit error rates are selected and calibrated as the final configuration areas of the three-dimensional structured light ranging device.

[0101] It should be noted that environmental factors will affect the spectrum of the signal. For example, temperature changes may cause changes in the characteristics of the signal propagation medium, thereby distorting the signal spectrum. By analyzing and predicting dynamic environmental characteristic data, the degree of this distortion can be estimated. The distance value between each signal transmission starting position node and the signal transmission end position node is obtained. This distance value is an important factor affecting the quality of signal transmission. Longer distances will cause signal attenuation, increased interference and other problems. The bit error rate is an indicator of signal transmission accuracy. The larger the distance value and the higher the signal spectrum distortion rate, the higher the bit error rate may lead to. For example, a longer distance will cause signal attenuation, and spectrum distortion may make it difficult for the signal to be correctly decoded at the receiving end, thereby increasing the possibility of bit errors.

[0102] By considering the signal spectrum distortion rate and the distance between the device and the receiving node to determine the bit error rate, the key factors affecting the accuracy of signal transmission are comprehensively integrated. This comprehensive consideration avoids the problem of inaccurate evaluation of signal transmission accuracy that may occur when focusing on a single factor (such as only considering the distance and ignoring the impact of the environment on the spectrum). Selecting the area with the lowest bit error rate as the final configuration area can minimize the bit errors in the signal transmission process. When configuring the device in this area, the signal can be more accurately transmitted from the device to the remote signal receiving node, thereby improving the accuracy and reliability of signal transmission. In addition, when determining the bit error rate, the predicted dynamic environmental feature data of each of the three candidate sub-configuration areas within the preset working time period is used, which enables the configuration of the device to take into account future environmental changes. Even if the environmental factors change, the device can still maintain a relatively stable working state in the final configuration area because the impact of these factors on signal transmission is considered in advance, thereby enhancing the device's ability to adapt to environmental changes.

[0103] In addition, the configuration method further includes the following steps:

[0104] Obtaining a natural disaster monitoring record book of the final configuration area, and obtaining natural disaster events that have occurred in the final configuration area according to the natural disaster monitoring record book; and obtaining historical environmental characteristic data corresponding to natural disaster events that occurred in the final configuration area;

[0105] According to the probability of occurrence of natural disaster events, the conditional dependency relationship between each natural disaster event and historical environmental characteristic data is obtained;

[0106] Constructing a conditional random field, taking the natural disaster event as the first conditional node of the conditional random field, and taking each historical environmental feature data as the second conditional node of the conditional random field;

[0107] Determine a connection relationship between a first conditional node and a second conditional node according to the conditional dependency relationship, and connect the first conditional node and the second conditional node in a directed manner according to the connection relationship to obtain a network topology diagram;

[0108] Building a probabilistic graph model based on a convolutional neural network, and importing the network topology graph into the embedded probabilistic graph model;

[0109] Obtaining predicted dynamic environmental characteristic data of the final configuration area within a preset working time period, importing the predicted dynamic environmental characteristic data of the final configuration area within the preset working time period into the probability graph model for prediction, and obtaining a probability value of a natural disaster occurring in the final configuration area within the preset working time period;

[0110] And obtain the time node corresponding to the probability value of natural disasters occurring in the final configuration area within the preset working time period that is greater than the preset probability value, and output the time node corresponding to the probability value of natural disasters occurring that is greater than the preset probability value as the early warning time node to remind to take protective measures for the three-dimensional structured light ranging equipment.

[0111] It should be noted that the conditional dependency relationship between each natural disaster event and the historical environmental characteristic data is obtained according to the probability of occurrence of natural disaster events. For example, if the probability of flood occurrence in a certain area increases significantly when the rainfall reaches a certain level, then it can be determined that there is a certain conditional dependency relationship between the natural disaster event of flood and the historical environmental characteristic data of rainfall. This relationship can be determined by statistical analysis, data mining and other methods. Conditional random field is a probabilistic graph model used to model conditional dependencies in sequence data. Here, it is used to describe the complex relationship between natural disaster events and historical environmental characteristic data. The network topology diagram intuitively shows the logical relationship structure between natural disaster events and historical environmental characteristic data. A probabilistic graph model is constructed based on a convolutional neural network, and the network topology diagram is imported into the embedded probabilistic graph model. Convolutional neural networks have powerful feature extraction and pattern recognition capabilities. By embedding the network topology diagram into the probabilistic graph model, the advantages of convolutional neural networks can be used to process the complex relationship between natural disaster events and environmental characteristic data. The probability value of natural disasters occurring in the final configuration area within the preset working time period is greater than the time node corresponding to the preset probability value, and these time nodes are output as warning time nodes. The preset probability value is a threshold value set according to actual needs and risk tolerance. When the predicted probability of natural disasters exceeds this threshold, protective measures need to be taken for the three-dimensional structured light ranging equipment. It can be seen that by analyzing the relationship between historical natural disaster events and historical environmental feature data, constructing a probability graph model and using the predicted dynamic environmental feature data for prediction, the probability value of natural disasters occurring in the final configuration area in the future working time period can be obtained in advance. This enables a quantitative assessment of the risks faced by the equipment before a natural disaster occurs, rather than waiting until a natural disaster occurs to react, avoiding damage to the equipment due to natural disasters and improving the foresight of equipment protection. At the same time, according to the probability value of natural disasters, a more reasonable equipment maintenance strategy can be formulated. If the probability of natural disasters occurring in a certain period of time is high, it may be necessary to strengthen the inspection and maintenance frequency of the equipment to ensure that the equipment can work normally when facing the risk of natural disasters. On the contrary, if the probability is low, the maintenance plan can be adjusted appropriately to save resources.

[0112] The present invention also discloses a configuration device for a three-dimensional structured light ranging device, such as Figure 3As shown, the configuration device of the three-dimensional structured light ranging device includes a memory 20 and a processor 90. The memory 20 stores a configuration method program of the three-dimensional structured light ranging device. When the configuration method program of the three-dimensional structured light ranging device is executed by the processor 90, any one of the configuration method steps of the three-dimensional structured light ranging device is implemented.

[0113] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A configuration method for a three-dimensional structured light ranging device, characterized in that: The following steps are involved: Acquire a target configuration area of ​​a three-dimensional structured light ranging device, and divide the target configuration area into a plurality of sub-configuration areas; Acquire dynamic environmental feature data of each sub-configuration area within a preset historical time period, construct a long short-term memory network prediction model, and train the long short-term memory network prediction model based on the dynamic environmental feature data to obtain a trained long short-term memory network prediction model; According to the trained long short-term memory network prediction model, the predicted dynamic environment characteristic data of each sub-configuration area within a preset working time period is obtained; the predicted dynamic environment characteristic data of each sub-configuration area is analyzed to obtain a class of candidate sub-configuration areas; Obtaining a three-dimensional model map of the terrain area of ​​each first-category candidate sub-configuration area, and screening and analyzing each first-category candidate sub-configuration area according to the three-dimensional model map of the terrain area to obtain a second-category candidate sub-configuration area; Obtaining a three-dimensional model diagram of the scene between each of the second-category candidate sub-configuration regions and the position area of ​​the object to be measured, and screening and analyzing the second-category candidate sub-configuration regions according to the three-dimensional model diagram of the scene to obtain three-category candidate sub-configuration regions; The bit error rates of the three types of candidate sub-configuration areas are obtained, and the three types of candidate sub-configuration areas are screened and analyzed according to the bit error rates to obtain the final configuration area of ​​the three-dimensional structured light ranging device.

2. The configuration method of a three-dimensional structured light ranging device according to claim 1, characterized in that: The dynamic environmental feature data of each sub-configuration area in a preset historical time period is obtained, a long short-term memory network prediction model is constructed, and the long short-term memory network prediction model is trained based on the dynamic environmental feature data to obtain a trained long short-term memory network prediction model, specifically: Acquire monitoring logs of each environmental monitoring device in the sub-configuration area, and acquire dynamic environmental characteristic data of each sub-configuration area within a preset historical time period according to the monitoring logs; Calculate the LOF value of the dynamic environment feature data in each sub-configuration area; filter out the dynamic environment feature data with a LOF value greater than a preset threshold to obtain the denoised dynamic environment feature data; The denoised dynamic environment feature data is sorted based on the time series to obtain the dynamic environment feature data of each sub-configuration area based on the time series; Constructing a long short-term memory network prediction model, and importing the time-series dynamic environmental feature data of each sub-configuration area into the long short-term memory network prediction model for training, and obtaining the training parameters of the model after each training process is completed; When the training parameters of the model meet the preset requirements, the last training parameters are used as the final parameters of the model, and the trained long short-term memory network prediction model is output.

3. The configuration method of a three-dimensional structured light ranging device according to claim 1, characterized in that: The predicted dynamic environment characteristic data of each sub-configuration area within a preset working time period are obtained according to the trained long short-term memory network prediction model; the predicted dynamic environment characteristic data of each sub-configuration area are analyzed to obtain a class of candidate sub-configuration areas, specifically: Obtaining a preset working time period of a three-dimensional structured light ranging device, and importing the preset working time period into the trained long short-term memory network prediction model to perform prediction and obtain predicted dynamic environment feature data of each sub-configuration area within the preset working time period; Acquiring performance characteristic information of the three-dimensional structured light ranging device, and determining optimal dynamic environment characteristic data of the three-dimensional structured light ranging device during operation according to the performance characteristic information; Calculating the Pearson correlation coefficient between the predicted dynamic environment characteristic data of each sub-configuration area within the preset working time period and the optimal dynamic environment characteristic data; The sub-configuration regions corresponding to the Pearson correlation coefficient being greater than the preset coefficient threshold are marked as a type of candidate sub-configuration regions; and the sub-configuration regions corresponding to the Pearson correlation coefficient being not greater than the preset coefficient threshold are marked as non-sub-configuration regions.

4. The configuration method of a three-dimensional structured light ranging device according to claim 1, characterized in that: A three-dimensional model map of the terrain area of ​​each first-class candidate sub-configuration area is obtained, and each first-class candidate sub-configuration area is screened and analyzed according to the three-dimensional model map of the terrain area to obtain a second-class candidate sub-configuration area, specifically: Acquire terrain region image information of each first-class candidate sub-configuration region, and construct a terrain region three-dimensional model map of each first-class candidate sub-configuration region according to the terrain region image information; Obtain a device instruction manual of the three-dimensional structured light ranging device, and obtain a three-dimensional model map of the standard installation terrain of the three-dimensional structured light ranging device according to the device instruction manual; Constructing a plurality of virtual spaces, and importing the three-dimensional model images of the terrain regions of each type of candidate sub-configuration region into the virtual spaces respectively, and importing the three-dimensional model images of the standard installation terrain into the virtual spaces; Iteratively registering the three-dimensional model map of the terrain area in each virtual space with the three-dimensional model map of the standard installation terrain based on an iterative closest point matching algorithm until the iterative registration process is completed; Obtaining the volume value of the intersection model between the terrain area three-dimensional model map and the standard installation terrain three-dimensional model map in each virtual space, and obtaining the volume value of the non-intersection model between the terrain area three-dimensional model map and the standard installation terrain three-dimensional model map in each virtual space; Calculating the degree of intersection between the three-dimensional model diagram of the terrain area in each virtual space and the three-dimensional model diagram of the standard installation terrain according to the volume value of the intersection model and the volume value of the non-intersection model between the three-dimensional model diagram of the terrain area in each virtual space and the three-dimensional model diagram of the standard installation terrain; A type of candidate subconfiguration region corresponding to a degree of intersection greater than a preset degree of intersection is marked as a type of candidate subconfiguration region; a type of candidate subconfiguration region corresponding to a degree of intersection not greater than the preset degree of intersection is marked as a non-subconfigurable region.

5. The configuration method of a three-dimensional structured light ranging device according to claim 1, characterized in that: Obtain a scene 3D model diagram between each of the second-category candidate sub-configuration areas and the location area of ​​the object to be measured, and screen and analyze the second-category candidate sub-configuration areas according to the scene 3D model diagram to obtain three categories of candidate sub-configuration areas, specifically: Acquire the location area of ​​the object to be measured, and acquire scene image information between each of the second-category candidate sub-configuration areas and the location area of ​​the object to be measured; Constructing a three-dimensional model diagram of the scene between each of the two types of candidate sub-configuration areas and the position area of ​​the object to be measured according to the corresponding scene image information; Acquire the external dimension information of the three-dimensional structured light ranging device, and construct a three-dimensional model diagram of the three-dimensional structured light ranging device according to the external dimension information; The three-dimensional model diagram of the device is simulated and installed on the two types of candidate sub-configuration areas of the three-dimensional model diagram of each scene; and the three-dimensional structured light ranging device on the two types of candidate sub-configuration areas is simulated and analyzed in the three-dimensional model diagram of the scene; Determine whether there is ranging interference between the three-dimensional structured light ranging device on each of the second-category candidate sub-configuration areas and the object to be measured; The second type of candidate sub-configuration area without ranging interference is marked as the third type of candidate sub-configuration area; the second type of candidate sub-configuration area with ranging interference is marked as a non-sub-configurable area.

6. The configuration method of a three-dimensional structured light ranging device according to claim 1, characterized in that: Obtain the bit error rate of each of the three types of candidate sub-configuration areas, screen and analyze the three types of candidate sub-configuration areas according to the bit error rate, and obtain the final configuration area of ​​the three-dimensional structured light ranging device, specifically: Obtain the position nodes of the three types of candidate sub-configuration areas, and calibrate the position nodes of the three types of candidate sub-configuration areas as the signal transmission start position nodes; obtain the remote signal receiving node of the three-dimensional structured light ranging device, and calibrate the remote signal receiving node of the three-dimensional structured light ranging device as the signal transmission end position node; Acquire predicted dynamic environment feature data of each of the three types of candidate sub-configuration areas within a preset working time period, and determine the signal spectrum distortion rate of the three-dimensional structured light ranging device in each of the three types of candidate sub-configuration areas when working within the preset working time period according to the predicted dynamic environment feature data of each of the three types of candidate sub-configuration areas within the preset working time period; And obtaining the distance value between each signal transmission starting position node and the signal transmission end position node, and obtaining the distance value between the three-dimensional structured light ranging device and the remote signal receiving node in each of the three types of candidate sub-configuration areas; Determine the bit error rate of the three-dimensional structured light ranging device in each of the three types of candidate sub-configuration areas during operation according to the distance value between the three-dimensional structured light ranging device and the remote signal receiving node in each of the three types of candidate sub-configuration areas and the signal spectrum distortion rate when working in a preset working time period; The three types of candidate sub-configuration areas with the lowest bit error rates are selected and calibrated as the final configuration areas of the three-dimensional structured light ranging device.

7. A configuration device for a three-dimensional structured light ranging device, characterized in that: The configuration device of the three-dimensional structured light ranging device includes a memory and a processor, wherein the memory stores a configuration method program of the three-dimensional structured light ranging device, and when the configuration method program of the three-dimensional structured light ranging device is executed by the processor, the configuration method steps of the three-dimensional structured light ranging device according to any one of claims 1 to 6 are implemented.

Images