A high-precision stacker crane intelligent obstacle avoidance system and method based on hybrid guidance

Abstract

The present invention relates to the field of obstacle avoidance control technology, and specifically discloses a high-precision intelligent obstacle avoidance system and method for stackers based on hybrid guidance. The system significantly improves operational safety and efficiency through a triple intelligent adjustment mechanism. First, laser reflection parameters are collected in real time, and the reflection process is dynamically optimized to ensure environmental perception accuracy. Then, high-density point cloud mapping is generated to accurately restore the spatial obstacle distribution, and the mapping algorithm is automatically calibrated through parameter analysis to improve scene restoration. Finally, multi-source data is integrated, and a hybrid guidance strategy is adopted to generate an obstacle avoidance path. Abnormal fluctuations in decision parameters are simultaneously monitored to achieve closed-loop optimization of path planning. This technology breaks through the static response limitations of traditional obstacle avoidance systems. Through the three-level adaptive adjustment of laser-point cloud-decision, it achieves centimeter-level obstacle avoidance accuracy in complex warehousing environments, effectively reducing collision risks and improving the intelligence and reliability of warehousing logistics.

Description

Technical Field

[0001] The present invention relates to the field of obstacle avoidance control technology, and in particular to a high-precision intelligent obstacle avoidance system and method for a stacker based on hybrid guidance. Background Art

[0002] In the field of automated warehousing, high-precision stackers need to balance operational efficiency, safe obstacle avoidance, and positioning accuracy in complex dynamic environments. Therefore, an intelligent obstacle avoidance system integrating multimodal perception and dynamic prediction is constructed to overcome the limitations of traditional technologies in environmental adaptability, decision-making coordination, and execution accuracy.

[0003] For example, the invention patent with announcement number CN112650225B announces an AGV obstacle avoidance method, which includes the following steps: setting a positioning point A on the AGV; the AGV's obstacle avoidance radar detects the position of the obstacle B around the positioning point A, and obtains the distance Lab between the positioning point A and the obstacle B; selecting the target point T according to the remaining path planning, and calculating the distance Lat between the positioning point A and the target point T; calculating the reference distance Ltb based on the distance Lab between the positioning point A and the obstacle B, and the distance Lat between the positioning point A and the target point T, where ; Calculate the expected stopping distance Ls and the expected stopping acceleration a based on the reference distance Ltb; AGV adjusts the motion state based on the expected stopping distance Ls and the expected stopping acceleration a.

[0004] For example, patent application publication number CN115562282A discloses a method for AGV dynamic obstacle avoidance based on improved speed barriers. The method includes the following steps: 1. Using the Kalman filter algorithm to predict the next obstacle position based on the speed and direction of the dynamic obstacle; 2. Constructing a speed barrier buffer based on the predicted obstacle position; and 3. Performing multi-objective optimization on efficiency and safety to select the optimal speed. This method uses a Kalman filter algorithm to predict the position of dynamic obstacles and constructs a speed barrier model based on this prediction.

[0005] However, in the process of implementing the embodiments of the present application, it was found that the above-mentioned technology has at least the following technical problems: the existing technology only relies on distance signals (such as infrared / ultrasonic waves) and does not integrate multimodal data (such as visual texture, laser point cloud). It is easy to misdetect in complex reflectivity scenes (such as black light-absorbing objects), resulting in perception failure in complex reflectivity environments (such as low-reflectivity black light-absorbing objects, high-reflectivity metal surfaces), resulting in obstacle avoidance anomalies. Summary of the Invention

[0006] In response to the deficiencies in the prior art, the present invention provides a high-precision stacker crane intelligent obstacle avoidance system and method based on hybrid guidance, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: In the first aspect, the present invention provides a high-precision stacker intelligent obstacle avoidance system based on hybrid guidance, including: a laser reflection adjustment module, which is used to collect and analyze the laser reflection parameters of the high-precision stacker, so as to determine whether to intelligently adjust the laser reflection process of the high-precision stacker; a point cloud mapping adjustment module, which is used to collect the laser reflection results of the high-precision stacker, and perform point cloud mapping, obtain and analyze the point cloud mapping parameters, so as to determine whether to intelligently adjust the point cloud mapping process of the high-precision stacker; an obstacle avoidance process adjustment module, which is used to obtain and, based on the point cloud mapping results of the high-precision stacker, hybrid guide the high-precision stacker to avoid obstacles, collect and analyze the obstacle avoidance decision parameters of the high-precision stacker, so as to determine whether to adjust the obstacle avoidance process of the high-precision stacker.

[0008] As a further solution, it is determined whether to perform intelligent adjustment on the laser reflection process of the high-precision stacker. The specific determination process is: analyzing the laser reflection parameters of the high-precision stacker, obtaining the laser reflection index of the high-precision stacker, and comparing it with the laser reflection reference interval. If the laser reflection index of the high-precision stacker belongs to the laser reflection reference interval, it is determined that the laser reflection process of the high-precision stacker is not to be intelligently adjusted. At the same time, it is determined whether the first condition exists. If the first condition exists, it is determined that the laser reflection process of the high-precision stacker is not to be intelligently restored. If the first condition does not exist, it is determined that the laser reflection process of the high-precision stacker is to be intelligently restored. Specifically, according to the reflection recovery speed preset in the intelligent database, the amplifier gain of the high-precision stacker is restored to the basic amplifier gain, and the exposure time of the high-precision stacker is restored to the basic exposure time; if the high-precision If the laser reflection index of the high-precision stacker does not belong to the laser reflection reference interval, it is determined that the laser reflection process of the high-precision stacker is intelligently adjusted, specifically: if the laser reflection index of the high-precision stacker is greater than the maximum value of the laser reflection reference interval, the amplifier gain and exposure time of the high-precision stacker are obtained and reduced based on the first laser reflectivity of the high-precision stacker, and the area to which the high-precision stacker belongs is marked as a high-reflection area; if the laser reflection index of the high-precision stacker is less than the minimum value of the laser reflection reference interval, the amplifier gain and exposure time of the high-precision stacker are obtained and increased based on the second laser reflectivity of the high-precision stacker, and the area to which the high-precision stacker belongs is marked as a low-reflection area; the first condition refers to that the amplifier gain of the high-precision stacker is the basic amplifier gain and the exposure time of the high-precision stacker is the basic exposure time.

[0009] As a further solution, the point cloud mapping process of the high-precision stacker is intelligently adjusted. The specific adjustment process is: obtain the reflection type of the area to which the high-precision stacker belongs and the accurate deviation value of the laser point cloud mapping of the high-precision stacker. If the area to which the high-precision stacker belongs is a high-reflection area, the number of plane fitting iterations and the reflectivity mutation response speed of the high-precision stacker are increased based on the accurate deviation value of the laser point cloud mapping; if the area to which the high-precision stacker belongs is a low-reflection area, the number of plane fitting iterations and the reflectivity mutation response speed of the high-precision stacker are decreased based on the accurate deviation value of the laser point cloud mapping.

[0010] As a further solution, the obstacle avoidance process of the high-precision stacker is adjusted. The specific adjustment process is: determine whether the fourth condition exists. If the fourth condition exists, make a decision warning for the obstacle avoidance process of the high-precision stacker; if the fourth condition does not exist, obtain and based on the obstacle avoidance decision efficiency deviation value of the high-precision stacker, reduce the operating speed of the high-precision stacker and increase the safety margin of the high-precision stacker; the fourth condition is that the operating speed of the high-precision stacker is the minimum operating speed and the safety margin of the high-precision stacker is the maximum safety margin.

[0011] The second aspect of the present invention provides an intelligent obstacle avoidance method for a high-precision stacker based on hybrid guidance, including: step 1, collecting and analyzing the laser reflection parameters of the high-precision stacker, so as to determine whether to intelligently adjust the laser reflection process of the high-precision stacker; step 2, collecting the laser reflection results of the high-precision stacker, and performing point cloud mapping, obtaining and analyzing the point cloud mapping parameters, so as to determine whether to intelligently adjust the point cloud mapping process of the high-precision stacker; step 3, obtaining and based on the point cloud mapping results of the high-precision stacker, hybrid guiding the high-precision stacker to avoid obstacles, collecting and analyzing the obstacle avoidance decision parameters of the high-precision stacker, so as to determine whether to adjust the obstacle avoidance process of the high-precision stacker.

[0012] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0013] (1) The present invention provides a high-precision intelligent obstacle avoidance system and method for stackers based on hybrid guidance, which significantly improves the safety and efficiency of operations through a triple intelligent adjustment mechanism. First, the laser reflection parameters are collected in real time, and the reflection process is dynamically optimized to ensure the accuracy of environmental perception. Then, a high-density point cloud mapping is generated to accurately restore the spatial obstacle distribution, and the mapping algorithm is automatically calibrated through parameter analysis to improve the scene restoration degree. Finally, multi-source data is integrated, and a hybrid guidance strategy is used to generate an obstacle avoidance path. Abnormal fluctuations in decision parameters are simultaneously monitored to achieve closed-loop optimization of path planning. This technology breaks through the static response limitations of traditional obstacle avoidance systems. Through the three-level adaptive adjustment of laser-point cloud-decision, centimeter-level obstacle avoidance accuracy is achieved in complex warehousing environments, effectively reducing the risk of collision and improving the intelligence and reliability of warehousing logistics.

[0014] (2) The present invention achieves dual optimization of environmental perception accuracy and equipment stability through dynamic analysis of laser reflection parameters through an intelligent adjustment mechanism. Its adaptive adjustment strategy can respond to high / low reflection scenarios in real time to ensure data acquisition reliability; the intelligent recovery function prevents parameter offset and guarantees long-term operational stability; and the area marking provides environmental prior information for subsequent obstacle avoidance. The three mechanisms work together to significantly improve the robustness of the system, enabling the stacker crane to achieve more efficient obstacle avoidance operations in complex storage environments.

[0015] (3) The present invention analyzes the intelligent adjustment mechanism of point cloud mapping, identifies the environmental reflection characteristics, and dynamically optimizes the algorithm parameters, thereby significantly improving the three-dimensional reconstruction accuracy in complex scenes. For high-reflection areas, the number of iterations and response speed are increased, effectively suppressing noise interference; for low-reflection areas, computational redundancy is reduced and processing efficiency is improved. This adaptive strategy enables the system to accurately restore the spatial form of objects of different materials, provide a more reliable environmental model for obstacle avoidance planning, and overall improve the intelligent level of warehousing operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention is further described with reference to the accompanying drawings. However, the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of system module connections of the present invention.

[0018] Figure 2 Schematic diagram of the method steps of the present invention.

[0019] Figure 3 It is a schematic diagram of the laser reflection adjustment process of the present invention.

[0020] Figure 4 Schematic diagram of the point cloud mapping adjustment process of the present invention.

[0021] Figure 5 It is a schematic diagram of the obstacle avoidance process adjustment flow of the present invention. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0023] Reference Figure 1 As shown, the first aspect of the present invention provides a high-precision stacker intelligent obstacle avoidance system based on hybrid guidance, including: a laser reflection adjustment module, a point cloud mapping adjustment module, an obstacle avoidance process adjustment module and an intelligent database.

[0024] The intelligent database is used to store data contained in a high-precision stacker crane intelligent obstacle avoidance system based on hybrid guidance.

[0025] The laser reflection adjustment module is connected to the point cloud mapping adjustment module, the point cloud mapping adjustment module is connected to the obstacle avoidance process adjustment module, and the laser reflection adjustment module, the point cloud mapping adjustment module and the obstacle avoidance process adjustment module are all connected to the intelligent database.

[0026] The laser reflection adjustment module is used to collect and analyze the laser reflection parameters of the high-precision stacker, so as to determine whether to perform intelligent adjustment on the laser reflection process of the high-precision stacker.

[0027] Specifically, it is determined whether to perform intelligent adjustment on the laser reflection process of the high-precision stacker. The specific determination process is: analyzing the laser reflection parameters of the high-precision stacker, obtaining the laser reflection index of the high-precision stacker, and comparing it with the laser reflection reference interval. If the laser reflection index of the high-precision stacker belongs to the laser reflection reference interval, it is determined that the laser reflection process of the high-precision stacker is not to be intelligently adjusted. At the same time, it is determined whether the first condition exists. If the first condition exists, it is determined that the laser reflection process of the high-precision stacker is not to be intelligently restored. If the first condition does not exist, it is determined that the laser reflection process of the high-precision stacker is The process is intelligently restored, specifically, the amplifier gain of the high-precision stacker is restored to the basic amplifier gain, and the exposure time of the high-precision stacker is restored to the basic exposure time according to the reflection recovery speed preset in the intelligent database; the above-mentioned laser reflection reference interval is extracted from the intelligent database and is used to distinguish the reflection conditions of the area to which the high-precision stacker belongs; it should be explained that when restoring the amplifier gain, the amplifier gain recovery speed is used, and when restoring the exposure time, the exposure time recovery speed is used, and the amplifier gain recovery speed and the exposure time recovery speed are uniformly marked as the reflection recovery speed.

[0028] If the laser reflection index of the high-precision stacker does not belong to the laser reflection reference interval, it is determined that the laser reflection process of the high-precision stacker is intelligently adjusted, specifically: if the laser reflection index of the high-precision stacker is greater than the maximum value of the laser reflection reference interval, the amplifier gain and exposure time of the high-precision stacker are obtained and reduced based on the first laser reflectivity of the high-precision stacker, and the area to which the high-precision stacker belongs is marked as a high-reflection area; the above-mentioned first laser reflectivity refers to the laser reflection index of the high-precision stacker minus the maximum value of the laser reflection reference interval, and the processing result is divided by the maximum value of the laser reflection reference interval to obtain the first laser reflectivity, and the first laser reflectivity-amplifier gain is stored in the intelligent database. The reduction coefficient mapping table and the first laser reflectivity-exposure time reduction coefficient mapping table can directly query the first laser reflectivity of the high-precision stacker in the intelligent database to obtain the corresponding amplifier gain reduction coefficient and exposure time reduction coefficient. The amplifier gain reduction coefficient is multiplied by the amplifier gain, and the exposure time reduction coefficient is multiplied by the exposure time. The product result is the reduced amplifier gain and exposure time. The amplifier gain reduction coefficient represents the proportional value of the amplifier gain reduction, and the exposure time reduction coefficient represents the proportional value of the exposure time reduction. The area to which the above-mentioned high-precision stacker belongs refers to the spatial area where an effective echo signal is formed by reflection from the surface of an object after laser emission.

[0029] If the laser reflection index of the high-precision stacker is less than the minimum value of the laser reflection reference interval, the amplifier gain and exposure time of the high-precision stacker are obtained and increased based on the second laser reflectivity of the high-precision stacker, and the area to which the high-precision stacker belongs is marked as a low-reflection area; the above-mentioned second laser reflectivity refers to the minimum value of the laser reflection reference interval minus the laser reflection index of the high-precision stacker, and the processing result is divided by the minimum value of the laser reflection reference interval to obtain the second laser reflectivity, and the second laser reflectivity-amplifier gain increase coefficient mapping table and the first laser reflectivity-exposure time increase coefficient mapping table are stored in the intelligent database The second laser reflectivity of the high-precision stacker can be directly queried in the intelligent database to obtain the corresponding amplifier gain increase coefficient and exposure time increase coefficient. The amplifier gain increase coefficient is multiplied by the amplifier gain, and the exposure time increase coefficient is multiplied by the exposure time. The product result is the increased and adjusted amplifier gain and exposure time, among which the amplifier gain increase coefficient represents the proportional value of the amplifier gain increase, and the exposure time increase coefficient represents the proportional value of the exposure time increase. The area to which the above-mentioned high-precision stacker belongs refers to the spatial area where the effective echo signal is formed by reflection from the surface of the object after the laser is emitted.

[0030] The first condition refers to the amplifier gain of the high-precision stacker as the basic amplifier gain and the exposure time of the high-precision stacker as the basic exposure time, where the basic amplifier gain refers to the initial amplifier gain set in the high-precision stacker, and the basic exposure time refers to the initial exposure time set in the high-precision stacker.

[0031] It should be explained that the amplifier gain and exposure time follow the principle of synchronous adjustment. Therefore, when the amplifier gain is the basic amplifier gain, the exposure time is also the basic exposure time.

[0032] Furthermore, the laser reflection index of the high-precision stacker is specifically analyzed as follows: the laser reflection parameters of the high-precision stacker include the laser reflection frequency of the high-precision stacker, the laser reflectivity of the high-precision stacker, and the speckle contrast of the high-precision stacker; the above-mentioned laser reflection frequency refers to the number of effective reflections of the laser beam per unit time, which can be measured by a photoelectric detector; the above-mentioned laser reflectivity refers to the ratio of the laser power emitted by the high-precision stacker to the laser power received back, usually expressed as a percentage, and can be measured by a laser radar; the above-mentioned speckle contrast refers to the statistical measure of the light intensity fluctuations in the bright and dark areas in the speckle pattern formed when the laser irradiates a rough surface, and the calculation formula is , where σ is the standard deviation of light intensity, μ is the mean light intensity, and the speckle contrast reflects the surface roughness information. It is an important indicator of the quality of laser reflection signal and can be measured by laser speckle velocimeter.

[0033] Importance coefficients are introduced to quantify the influence of the deviation between the laser reflection frequency and the reference laser reflection frequency on the laser reflection index, the influence of the deviation between the laser reflectivity and the reference laser reflectivity on the laser reflection index, and the influence of the deviation between the speckle contrast and the reference speckle contrast on the laser reflection index. The influence degrees are summarized to obtain the laser reflection index of the high-precision stacker.

[0034] The laser reflection index of a high-precision stacker is a numerical value used to measure the comprehensive laser reflection characteristics of the stacker. The specific expression is:

[0035] ;

[0036] Where, It is the laser reflection index of high-precision stacker. is the laser reflection frequency component of the high-precision stacker, indicating the degree of deviation between the laser reflection frequency of the high-precision stacker and the reference laser reflection frequency, specifically: ; is the laser reflectivity component of the high-precision stacker, indicating the degree of deviation between the laser reflectivity of the high-precision stacker and the reference laser reflectivity, specifically: , is the speckle contrast component of the high-precision stacker, which indicates the degree of deviation between the speckle contrast of the high-precision stacker and the reference speckle contrast. Specifically: .

[0037] is the importance coefficient of the laser reflection frequency component preset in the intelligent database, is the importance coefficient of the laser reflectivity component preset in the intelligent database, is the importance coefficient of the speckle contrast component preset in the intelligent database, is the laser reflection frequency of the high-precision stacker, The reference laser reflection frequency preset in the intelligent database, The laser reflectivity of the high-precision stacker, is the reference laser reflectivity preset in the intelligent database, For the speckle contrast of high-precision stacker, It is the reference speckle contrast preset in the intelligent database.

[0038] The reference laser reflection frequency refers to a reference value of the laser reflection frequency; the reference laser reflectivity refers to a reference value of the laser reflectivity; and the reference speckle contrast refers to a reference value of the speckle contrast.

[0039] It should be explained that high-reflection areas typically refer to smooth, highly reflective surfaces, such as metal and glass. On such surfaces, the laser reflection frequency will be higher than the corresponding reference value. This is because the reflected light is strong, resulting in a high signal frequency received by the detector. The laser reflectivity will also be higher than the corresponding reference value because more of the incident light is reflected. Speckle contrast is a parameter in laser speckle interferometry. High speckle contrast indicates a clear speckle pattern, which typically occurs when the surface roughness is low (i.e., smooth). Rough surfaces blur the speckle pattern and reduce contrast. Conversely, low-reflection areas typically refer to rough, low-reflectivity surfaces, such as dark fabrics and rough plastics. On such surfaces, the laser reflection frequency is low because the reflected light is weak, resulting in a low signal frequency received by the detector. The laser reflectivity is lower than the reference value because most of the incident light is absorbed or scattered. The speckle contrast is low because the surface roughness blurs the speckle pattern, reducing contrast. Therefore, these parameters are interrelated and together reflect the reflection conditions in the area belonging to the high-precision stacker.

[0040] The above-mentioned laser reflection frequency component importance coefficient represents the proportion of the laser reflection frequency component in the laser reflection index; the above-mentioned laser reflectivity component importance coefficient represents the proportion of the laser reflectivity component in the laser reflection index; the above-mentioned speckle contrast component importance coefficient represents the proportion of the speckle contrast component in the laser reflection index. The intelligent database stores the correspondence between the laser reflection frequency component, the laser reflectivity component, and the speckle contrast component and their corresponding importance coefficients. For example, when the laser reflection frequency component, the laser reflectivity component, and the speckle contrast component are input into the intelligent database, the intelligent database can retrieve the laser reflection frequency component importance coefficient, the laser reflectivity component importance coefficient, and the speckle contrast component importance coefficient, and the value range of each is between 0 and 1.

[0041] In a specific embodiment, the present invention achieves dual optimization of environmental perception accuracy and equipment stability through dynamic analysis of laser reflection parameters through an intelligent adjustment mechanism. Its adaptive adjustment strategy can respond to high / low reflection scenarios in real time to ensure data acquisition reliability; the intelligent recovery function prevents parameter offset and guarantees long-term operational stability; and area marking provides environmental prior information for subsequent obstacle avoidance. The three mechanisms work together to significantly improve the robustness of the system, enabling the stacker to achieve more efficient obstacle avoidance operations in complex warehousing environments.

[0042] The point cloud mapping adjustment module is used to collect the laser reflection results of the high-precision stacker, perform point cloud mapping, obtain and analyze the point cloud mapping parameters, and thus determine whether to perform intelligent adjustment on the point cloud mapping process of the high-precision stacker.

[0043] Specifically, it is determined whether to perform intelligent adjustment on the point cloud mapping process of the high-precision stacker. The specific analysis process is: analyzing the point cloud mapping parameters, obtaining the laser point cloud mapping accuracy coefficient of the high-precision stacker, and comparing it with the laser point cloud mapping accuracy threshold. If the laser point cloud mapping accuracy coefficient of the high-precision stacker is greater than or equal to the laser point cloud mapping accuracy threshold, it is determined not to perform intelligent adjustment on the point cloud mapping process of the high-precision stacker, and at the same time, it is determined whether there is a second condition. If the second condition exists, it is determined not to perform intelligent recovery on the point cloud mapping process of the high-precision stacker. If the second condition does not exist, it is determined to perform intelligent recovery on the point cloud mapping process of the high-precision stacker. Specifically, according to the intelligent number The mapping recovery speed preset in the database is used to restore the plane fitting iteration number of the high-precision stacker to the basic plane fitting iteration number, and the reflectivity mutation response speed of the high-precision stacker to the basic reflectivity mutation response speed; the above-mentioned laser point cloud mapping accuracy threshold represents the minimum value allowed by the laser point cloud mapping accuracy coefficient stored in the intelligent database; it should be explained that when restoring the plane fitting iteration number, the plane fitting iteration number recovery speed is used, and when restoring the reflectivity mutation response speed, the reflectivity mutation response speed recovery speed is used, and the plane fitting iteration number recovery speed and the reflectivity mutation response speed recovery speed are uniformly marked as mapping recovery speed.

[0044] If the laser point cloud mapping accuracy coefficient of the high-precision stacker is less than the laser point cloud mapping accuracy threshold, it is determined that the point cloud mapping process of the high-precision stacker is to be intelligently adjusted.

[0045] The second condition refers to that the laser point cloud mapping accuracy margin value of the high-precision stacker is greater than or equal to the defined laser point cloud mapping accuracy margin value, and at the same time, the plane fitting iteration number of the high-precision stacker is the basic plane fitting iteration number and the reflectivity mutation response speed of the high-precision stacker is the basic reflectivity mutation response speed; the above-mentioned laser point cloud mapping accuracy margin value refers to the degree to which the laser point cloud mapping accuracy coefficient is greater than the laser point cloud mapping accuracy threshold, specifically, the laser point cloud mapping accuracy threshold is subtracted from the laser point cloud mapping accuracy coefficient, and the processed result is divided by the laser point cloud mapping accuracy threshold, and the final result is the laser point cloud mapping accuracy margin value; the above-mentioned defined laser point cloud mapping accuracy margin value represents the minimum value allowed for the laser point cloud mapping accuracy margin value, which is extracted from the intelligent database, wherein the basic plane fitting iteration number refers to the initial plane fitting iteration number set in the high-precision stacker, and the basic reflectivity mutation response speed refers to the initial reflectivity mutation response speed set in the high-precision stacker.

[0046] It should be explained that the reflectivity mutation response speed and the number of plane fitting iterations follow the principle of synchronous regulation. Therefore, when the reflectivity mutation response speed is the basic reflectivity mutation response speed, the number of plane fitting iterations is also the basic plane fitting iteration number.

[0047] Specifically, the laser point cloud mapping accuracy coefficient of the high-precision stacker, the specific analysis process is as follows: the point cloud mapping parameters include the laser point cloud distribution density of the high-precision stacker, the laser point cloud mapping time deviation of the high-precision stacker and the laser point cloud angular resolution of the high-precision stacker; the above-mentioned laser point cloud distribution density refers to the number of laser points in a unit space volume, which can be obtained through laser radar monitoring; the above-mentioned laser point cloud mapping time deviation refers to the deviation ratio between the actual point cloud mapping time and the defined consumption time of the high-precision stacker, the actual point cloud mapping time is subtracted from the defined consumption time, and the difference result is divided by the defined consumption time to finally obtain the laser point cloud mapping time deviation, which can be obtained through embedded timer monitoring, where the defined consumption time represents the maximum value allowed for the point cloud mapping time, which is extracted from the intelligent database; the above-mentioned laser point cloud angular resolution refers to the minimum angular interval that the laser scanning system can distinguish between two adjacent point clouds in the same ranging unit, which can be obtained through laser radar monitoring.

[0048] According to the laser reflection index of the high-precision stacker, a correction coefficient is matched from the intelligent database; the correction coefficient represents the proportional value of the data correction of the laser point cloud mapping accuracy coefficient and the obstacle avoidance decision efficiency index.

[0049] It should be explained that the core value of matching the correction coefficient corresponding to the laser reflection index from the intelligent database based on the laser reflection index of the high-precision stacker lies in the construction of an intelligent closed-loop system of "environmental perception-parameter adaptation-efficiency improvement". The corrected data is closer to the real environment, reducing the processing burden of subsequent algorithms. Accurate data enables the obstacle avoidance algorithm to adopt a simpler model (such as reducing redundant collision detection loops), thereby improving response speed. This positive cycle of "improving data quality → improving algorithm efficiency" is the core mechanism by which the correction coefficient can correct data.

[0050] By introducing weighted quantitative analysis to determine the influence of the deviation between the laser point cloud distribution density and the reference laser point cloud distribution density on the laser point cloud mapping accuracy coefficient, the influence of the proportional relationship between the laser point cloud mapping time deviation and the defined laser point cloud mapping time deviation on the laser point cloud mapping accuracy coefficient, and the influence of the proportional relationship between the laser point cloud angular resolution and the defined laser point cloud angular resolution on the laser point cloud mapping accuracy coefficient, each influence degree is aggregated and a correction coefficient is introduced to correct the aggregation result, thereby obtaining the laser point cloud mapping accuracy coefficient of the high-precision stacker.

[0051] The accuracy coefficient of the laser point cloud mapping of the high-precision stacker is used to quantify the accuracy of the laser point cloud mapping of the high-precision stacker. The specific expression is:

[0052] ;

[0053] Where, is the accuracy coefficient of laser point cloud mapping of high-precision stacker, is the correction factor, The laser point cloud distribution density of the high-precision stacker crane is The reference laser point cloud distribution density preset in the intelligent database, It is the time deviation of laser point cloud mapping of high-precision stacker. It is the deviation of the laser point cloud mapping duration preset in the intelligent database. The laser point cloud angle resolution of the high-precision stacker crane is The angle resolution of the laser point cloud is defined in the intelligent database. It is the laser point cloud distribution density weighted number preset in the intelligent database. It is the weighted deviation of the laser point cloud mapping duration preset in the intelligent database. It is the weighted number of the laser point cloud angle resolution preset in the intelligent database. k is a constant and is not 0. It is used to ensure the validity of the laser point cloud mapping accuracy coefficient.

[0054] The above-mentioned reference laser point cloud distribution density represents the reference value of the laser point cloud distribution density; the above-mentioned definition of laser point cloud mapping time deviation represents the maximum value allowed by the laser point cloud mapping time deviation; the above-mentioned definition of laser point cloud angular resolution represents the minimum value allowed by the laser point cloud angular resolution.

[0055] It needs to be explained that when the distribution density of the laser point cloud deviates significantly from the preset reference value, this directly indicates that the laser mapping system has abnormal fluctuations, which leads to a decrease in angular resolution, resulting in the loss of spatial details and the formation of "information blind spots" - for example, in complex curved surfaces or edge areas, low angular resolution will prevent the point cloud from fitting closely to the actual contours of the object, thereby reducing the mapping accuracy coefficient. The deviation of the laser point cloud mapping time directly reflects the degree of matching between the system's real-time performance and computing resources: when the distribution density or angular resolution increases, the data collection volume and processing complexity increase simultaneously. If the hardware performance is not upgraded synchronously, the mapping time deviation will increase significantly. At this time, the distribution density or angular resolution may be automatically reduced to sacrifice local accuracy in exchange for overall operating efficiency. , to prevent the stacker from reducing its operating speed due to waiting for data, the laser point cloud distribution density indirectly determines the required threshold of angular resolution by affecting the amount of data collected; and insufficient angular resolution will in turn limit the effectiveness of the distribution density, forming an accuracy bottleneck. The mapping time deviation is used as a real-time feedback signal for the efficiency of the high-precision stacker, and the set values ​​of the distribution density and angular resolution are dynamically adjusted; at the same time, the reflectivity mutation response speed is based on the real-time deviation value. Under the constraints of distribution density and angular resolution, the point cloud fitting strategy is optimized to compensate for environmental interference. The three work together through the closed-loop collaboration of "data acquisition-processing efficiency-environmental adaptability" to affect the laser point cloud mapping accuracy coefficient, so that the stacker can dynamically balance accuracy and efficiency in different reflective characteristic areas.

[0056] The above-mentioned laser point cloud distribution density weighted number is used to quantify the influence of the unit value of the laser point cloud distribution density on the laser point cloud mapping accuracy coefficient; the above-mentioned laser point cloud mapping time deviation weighted number is used to quantify the influence of the unit value of the laser point cloud mapping time deviation on the laser point cloud mapping accuracy coefficient; the above-mentioned laser point cloud angular resolution weighted number is used to quantify the influence of the unit value of the laser point cloud angular resolution on the laser point cloud mapping accuracy coefficient. The intelligent database stores the correspondence between the laser point cloud distribution density, the laser point cloud mapping time deviation and the laser point cloud angular resolution and their corresponding weighted numbers. For example, the laser point cloud distribution density, the laser point cloud mapping time deviation and the laser point cloud angular resolution are input into the intelligent database, and the intelligent database can retrieve the laser point cloud distribution density weighted number, the laser point cloud mapping time deviation weighted number and the laser point cloud angular resolution weighted number, and the value range is between 0 and 1.

[0057] Furthermore, the point cloud mapping process of the high-precision stacker is intelligently adjusted, and the specific adjustment process is: obtaining the reflection type of the area to which the high-precision stacker belongs and the accurate deviation value of the laser point cloud mapping of the high-precision stacker; if the area to which the high-precision stacker belongs is a high-reflection area, then the plane fitting iteration number and the reflectivity mutation response speed of the high-precision stacker are increased and adjusted based on the accurate deviation value of the laser point cloud mapping; the above-mentioned accurate deviation value of the laser point cloud mapping refers to the deviation value between the laser point cloud mapping accuracy coefficient of the high-precision stacker and the laser point cloud mapping accuracy threshold, specifically, the laser point cloud mapping accuracy coefficient of the high-precision stacker is subtracted from the laser point cloud mapping accuracy threshold, and the result is the laser point cloud mapping accurate deviation value; under the condition that the area to which the high-precision stacker belongs is a high-reflection area, the corresponding plane fitting iteration number increase coefficient and reflectivity mutation response speed increase coefficient are queried. Specifically, a mapping table of accurate deviation value of laser point cloud mapping-increase coefficient of plane fitting iteration number and a mapping table of accurate deviation value of laser point cloud mapping-increase coefficient of reflectivity mutation response speed are stored in the intelligent database. The accurate deviation value of laser point cloud mapping of the high-precision stacker can be directly queried in the intelligent database to obtain the corresponding increase coefficient of plane fitting iteration number and increase coefficient of reflectivity mutation response speed. The increase coefficient of plane fitting iteration number is multiplied by the number of plane fitting iterations, and the increase coefficient of reflectivity mutation response speed is multiplied by the reflectivity mutation response speed. The product result is the increased and adjusted number of plane fitting iterations and reflectivity mutation response speed. The increase coefficient of plane fitting iteration number represents the proportional value of increasing the number of plane fitting iterations, and the increase coefficient of reflectivity mutation response speed represents the proportional value of increasing the reflectivity mutation response speed.

[0058] If the area to which the high-precision stacker belongs is a low-reflection area, the plane fitting iteration number and the reflectivity mutation response speed of the high-precision stacker are reduced and adjusted based on the accurate deviation value of the laser point cloud mapping. Under the condition that the area to which the high-precision stacker belongs is a low-reflection area, the corresponding plane fitting iteration number reduction coefficient and the reflectivity mutation response speed reduction coefficient are queried. Specifically, a laser point cloud mapping accurate deviation value-plane fitting iteration number reduction coefficient mapping table and a laser point cloud mapping accurate deviation value-reflectivity mutation response speed reduction coefficient mapping table are stored in the intelligent database. The laser point cloud mapping accurate deviation value of the high-precision stacker can be directly queried in the intelligent database to obtain the corresponding plane fitting iteration number reduction coefficient and the reflectivity mutation response speed reduction coefficient. The plane fitting iteration number reduction coefficient is multiplied by the plane fitting iteration number, and the reflectivity mutation response speed reduction coefficient is multiplied by the reflectivity mutation response speed. The product result is the reduced plane fitting iteration number and the reflectivity mutation response speed. The plane fitting iteration number reduction coefficient represents the proportional value for reducing the plane fitting iteration number, and the reflectivity mutation response speed reduction coefficient represents the proportional value for reducing the reflectivity mutation response speed.

[0059] In a specific embodiment, the present invention analyzes the point cloud mapping intelligent adjustment mechanism, identifies the environmental reflection characteristics, and dynamically optimizes the algorithm parameters, thereby significantly improving the three-dimensional reconstruction accuracy in complex scenes. For high-reflection areas, the number of iterations and response speed are increased to effectively suppress noise interference; for low-reflection areas, computational redundancy is reduced and processing efficiency is improved. This adaptive strategy enables the system to accurately restore the spatial form of objects of different materials, provide a more reliable environmental model for obstacle avoidance planning, and overall improve the intelligence level of warehousing operations.

[0060] The obstacle avoidance process adjustment module is used to obtain and hybrid guide the high-precision stacker to avoid obstacles based on the point cloud mapping results of the high-precision stacker, collect and analyze the obstacle avoidance decision parameters of the high-precision stacker, and thus determine whether to adjust the obstacle avoidance process of the high-precision stacker.

[0061] Specifically, it is determined whether to adjust the obstacle avoidance process of the high-precision stacker. The specific determination process is: analyzing the obstacle avoidance decision parameters of the high-precision stacker, obtaining the obstacle avoidance decision efficiency index of the high-precision stacker, and comparing it with the obstacle avoidance decision efficiency threshold. If the obstacle avoidance decision efficiency index of the high-precision stacker is greater than or equal to the obstacle avoidance decision efficiency threshold, it is determined not to adjust the obstacle avoidance process of the high-precision stacker, and at the same time, it is determined whether there is a third condition. If the third condition exists, it is determined not to intelligently restore the obstacle avoidance process of the high-precision stacker. If the third condition does not exist, it is determined to intelligently restore the obstacle avoidance process of the high-precision stacker, specifically according to the decision recovery preset in the intelligent database. Restoring speed, restoring the operating speed of the high-precision stacker to the basic operating speed, and restoring the safety margin of the high-precision stacker to the basic safety margin; the above-mentioned obstacle avoidance decision effectiveness threshold represents the minimum value allowed by the obstacle avoidance decision effectiveness index, which is extracted from the intelligent database; it should be explained that when restoring the operating speed, the operating speed recovery speed is used, and when restoring the safety margin, the safety margin recovery speed is used, and the operating speed recovery speed and the safety margin recovery speed are uniformly marked as the decision recovery speed, the basic operating speed represents the initial operating speed set in the high-precision stacker, and the basic safety margin represents the initial safety margin set in the high-precision stacker.

[0062] If the obstacle avoidance decision efficiency index of the high-precision stacker is less than the obstacle avoidance decision efficiency threshold, it is determined that the obstacle avoidance process of the high-precision stacker is adjusted; the third condition refers to that the obstacle avoidance decision efficiency margin value of the high-precision stacker is greater than or equal to the defined obstacle avoidance decision efficiency margin value, and at the same time, the operating speed of the high-precision stacker is the basic operating speed and the safety margin of the high-precision stacker is the basic safety margin.

[0063] It needs to be explained that the operating speed and safety margin follow the principle of synchronous adjustment. Therefore, when the operating speed is the basic operating speed, the safety margin is also the basic safety margin. The operating speed is the driving speed, and the safety margin is the safety distance.

[0064] Specifically, the obstacle avoidance decision-making efficiency index of the high-precision stacker is analyzed as follows: the obstacle avoidance decision parameters of the high-precision stacker include the command response time of the high-precision stacker and the obstacle avoidance action accuracy factor of the high-precision stacker; the command response time refers to the physical time interval from the high-precision stacker receiving the command signal (such as an emergency stop command or a path adjustment command) to the actuator (such as a servo motor) starting to respond, and the unit is usually milliseconds and can be monitored by a signal generator; the obstacle avoidance action accuracy factor represents the degree of consistency between the actual motion trajectory and the planned trajectory when the stacker performs the obstacle avoidance action, and its calculation formula is: , which can be captured by a high-precision motion capture system (such as Optik).

[0065] The importance values ​​are used to quantify the influence of the laser point cloud mapping accuracy coefficient, the proportional relationship between the command response time and the defined command response time, and the proportional relationship between the obstacle avoidance action accuracy factor and the defined obstacle avoidance action accuracy factor on the obstacle avoidance decision effectiveness index. The influence degrees are summarized and the summary results are corrected by the correction coefficient to obtain the obstacle avoidance decision effectiveness index of the high-precision stacker.

[0066] The obstacle avoidance decision-making efficiency index of the high-precision stacker is used to quantify the obstacle avoidance decision-making efficiency of the high-precision stacker. The specific expression is:

[0067] ;

[0068] Where, is the obstacle avoidance decision efficiency index of the high-precision stacker, is the correction factor, is the accuracy coefficient of laser point cloud mapping of high-precision stacker, The command response time of the high-precision stacker. It is the defined instruction response time preset in the intelligent database. is the accuracy factor of the obstacle avoidance action of the high-precision stacker. It is the accuracy factor of obstacle avoidance action preset in the intelligent database. is the importance value of the laser point cloud mapping accuracy coefficient preset in the intelligent database, It is the importance value of the instruction response time preset in the intelligent database. It is the importance value of the obstacle avoidance action accuracy factor preset in the intelligent database.

[0069] The above-mentioned definition of the command response time indicates the maximum value allowed for the command response time; the above-mentioned definition of the obstacle avoidance action accuracy factor indicates the minimum value allowed for the obstacle avoidance action accuracy factor.

[0070] It should be explained that the laser point cloud mapping accuracy coefficient, command response time, and obstacle avoidance action accuracy factor form a strong coupling relationship through the "perception-decision-execution" closed loop, and jointly affect the obstacle avoidance decision-making efficiency index. The specific correlation mechanism is as follows: the laser point cloud mapping accuracy coefficient directly drives the obstacle avoidance action accuracy factor through the environmental perception accuracy. When the coefficient increases, the obstacle outline resolution is improved and the spatial positioning reference is more accurate, which increases the obstacle avoidance action accuracy factor. The command response time inversely constrains the obstacle avoidance action accuracy factor through real-time constraints. As the response time increases, the risk of path deviation caused by the dynamic displacement of obstacles (such as cargo movement) increases, which causes the obstacle avoidance action accuracy factor to decrease. The obstacle avoidance action accuracy factor, in turn, optimizes the laser point cloud mapping accuracy coefficient through execution effect feedback. When the obstacle avoidance action accuracy factor decreases, local point cloud encryption resampling or adjustment of the reflectivity compensation algorithm is automatically triggered to increase the accuracy of the obstacle avoidance decision, thereby increasing the mapping accuracy coefficient. Through the interactive mechanism of "accuracy transmission → real-time constraint → feedback optimization", the three factors form a positive cycle of "perception optimization → improved decision reliability → improved execution accuracy" in a dynamic environment.

[0071] The importance value of the above-mentioned laser point cloud mapping accuracy coefficient is used to quantify the influence of the unit value of the laser point cloud mapping accuracy coefficient on the obstacle avoidance decision efficiency index; the importance value of the above-mentioned command response time is used to quantify the influence of the unit value of the command response time on the obstacle avoidance decision efficiency index; the importance value of the above-mentioned obstacle avoidance action accuracy factor is used to quantify the influence of the unit value of the obstacle avoidance action accuracy factor on the obstacle avoidance decision efficiency index; the intelligent database stores the correspondence between the laser point cloud mapping accuracy coefficient, the command response time and the obstacle avoidance action accuracy factor and their corresponding importance values. For example, the laser point cloud mapping accuracy coefficient, the command response time and the obstacle avoidance action accuracy factor are input into the intelligent database, and the intelligent database can retrieve the importance value of the laser point cloud mapping accuracy coefficient, the importance value of the command response time and the importance value of the obstacle avoidance action accuracy factor, and the value range is between 0 and 1.

[0072] Furthermore, the obstacle avoidance process of the high-precision stacker is adjusted. The specific adjustment process is: determine whether the fourth condition exists. If the fourth condition exists, make a decision and early warning on the obstacle avoidance process of the high-precision stacker, specifically generate an early warning instruction and send it to the technical staff in the form of email or the like.

[0073] If the fourth condition does not exist, the obstacle avoidance decision efficiency deviation value of the high-precision stacker is obtained and the operating speed of the high-precision stacker is reduced and the safety margin of the high-precision stacker is increased. The above-mentioned obstacle avoidance decision efficiency deviation value represents the difference between the obstacle avoidance decision efficiency threshold and the obstacle avoidance decision efficiency index of the high-precision stacker. Specifically, the obstacle avoidance decision efficiency threshold is subtracted from the obstacle avoidance decision efficiency index, and the result is the obstacle avoidance decision efficiency deviation value. The intelligent database stores the obstacle avoidance decision efficiency deviation value-operating speed reduction coefficient mapping table and the obstacle avoidance decision efficiency deviation value. The difference-safety margin increase coefficient mapping table can directly query the obstacle avoidance decision efficiency deviation value of the high-precision stacker in the intelligent database to obtain the corresponding operating speed reduction coefficient and safety margin increase coefficient. The operating speed reduction coefficient is multiplied by the operating speed, and the product result is the operating speed after the reduction adjustment. The safety margin increase coefficient is multiplied by the safety margin, and the product result is the safety margin after the increase adjustment. The operating speed reduction coefficient represents the proportional value of the operating speed reduction, and the safety margin increase coefficient represents the proportional value of the safety margin increase.

[0074] The fourth condition is that the operating speed of the high-precision stacker is the minimum operating speed and the safety margin of the high-precision stacker is the maximum safety margin. The minimum operating speed and the maximum safety margin are determined by technical personnel.

[0075] In a specific embodiment, the present invention provides a high-precision stacker intelligent obstacle avoidance system based on hybrid guidance, which significantly improves operation safety and efficiency through a triple intelligent adjustment mechanism. First, laser reflection parameters are collected in real time, and the reflection process is dynamically optimized to ensure environmental perception accuracy; then high-density point cloud mapping is generated to accurately restore the spatial obstacle distribution, and the mapping algorithm is automatically calibrated through parameter analysis to improve scene restoration; finally, multi-source data is integrated, and a hybrid guidance strategy is adopted to generate an obstacle avoidance path, and abnormal fluctuations in decision parameters are simultaneously monitored to achieve closed-loop optimization of path planning. This technology breaks through the static response limitations of traditional obstacle avoidance systems, and through three-level adaptive adjustment of laser-point cloud-decision, centimeter-level obstacle avoidance accuracy is achieved in complex warehousing environments, effectively reducing collision risks and improving the intelligence and reliability of warehousing logistics.

[0076] Reference Figure 2As shown, the second aspect of the present invention provides a high-precision stacker intelligent obstacle avoidance method based on hybrid guidance, including: step 1, collecting and analyzing the laser reflection parameters of the high-precision stacker, so as to determine whether to intelligently adjust the laser reflection process of the high-precision stacker; step 2, collecting the laser reflection results of the high-precision stacker, and performing point cloud mapping, obtaining and analyzing the point cloud mapping parameters, so as to determine whether to intelligently adjust the point cloud mapping process of the high-precision stacker; step 3, obtaining and based on the point cloud mapping results of the high-precision stacker, hybrid guiding the high-precision stacker to avoid obstacles, collecting and analyzing the obstacle avoidance decision parameters of the high-precision stacker, so as to determine whether to adjust the obstacle avoidance process of the high-precision stacker.

[0077] Figure 3 This is a schematic diagram of the laser reflection adjustment process of the present invention. After the system is started, it first collects laser reflection parameters, including reflection frequency, reflectivity, and speckle contrast. Reflection index calculation is then performed by introducing an importance factor to comprehensively quantify the deviation of each parameter from a reference value, resulting in a laser reflection index. This index is used to comprehensively measure the current laser reflection characteristics. Judgment and adjustment: If the reflection index is within the reference range, the first condition (gain / exposure at baseline values) is checked. If satisfied, the current parameters are maintained. If not, the gain and exposure are restored to baseline values ​​according to a preset reflection recovery speed. If the reflection index exceeds the range: if it is greater than the maximum value, the current area is marked as a high-reflection area, and the gain and exposure are reduced based on the first laser reflectivity. If it is less than the minimum value, the current area is marked as a low-reflection area, and the gain and exposure are increased based on the second laser reflectivity.

[0078] Figure 4 This is a schematic diagram of the point cloud mapping adjustment process of the present invention. The process receives data after laser reflection adjustment and analyzes point cloud mapping parameters, including point cloud distribution density, mapping duration deviation, and angular resolution. The accuracy coefficient is calculated by comprehensively quantifying the deviations of each mapping parameter from the reference value based on the laser reflection index matching correction coefficient. This coefficient is used to quantify the accuracy of the point cloud mapping. Determination and adjustment: If the accuracy coefficient ≥ the threshold, the second condition (whether the mapping margin meets the standard) is checked. If so, the current mapping parameters are maintained. If not, the number of plane fitting iterations and the reflectivity mutation response speed are restored to their baseline values ​​according to the preset mapping recovery speed. If the accuracy coefficient is less than the threshold, the number of plane fitting iterations and the reflectivity mutation response speed are adjusted based on the accuracy deviation value according to the regional reflectivity type (high reflectivity / low reflectivity). The adjustment amount is increased in high-reflectivity areas and decreased in low-reflectivity areas.

[0079] Figure 5This is a schematic diagram of the obstacle avoidance process adjustment process of the present invention. Based on the point cloud mapping results, obstacle avoidance decision parameters, including command response time and obstacle avoidance action accuracy factor, are analyzed. The performance index is calculated by quantifying the deviation of each decision parameter from the reference value using importance values, and then a comprehensive obstacle avoidance decision performance index is calculated. This index is used to quantify the effectiveness of the obstacle avoidance decision. The following describes the following steps: If the performance index ≥ the threshold, the third condition (whether the performance margin meets the standard) is checked. If so, the current obstacle avoidance parameters are maintained. If not, the operating speed and obstacle avoidance safety margin are restored to their baseline values ​​according to the preset decision recovery speed. If the performance index < the threshold, the fourth condition (whether the minimum speed / maximum margin have been reached) is determined. If so, a decision warning is triggered. If not, the operating speed is reduced based on the performance deviation value and the safety margin is increased.

[0080] The above content is merely an example and explanation of the structure of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the scope of protection of the present invention.

Claims

1. A high-precision stacker crane intelligent obstacle avoidance system based on hybrid guidance, characterized in that: include: Laser reflection adjustment module, used to collect and analyze the laser reflection parameters of the high-precision stacker, so as to determine whether to perform intelligent adjustment on the laser reflection process of the high-precision stacker; The point cloud mapping adjustment module is used to collect the laser reflection results of the high-precision stacker, perform point cloud mapping, obtain and analyze the point cloud mapping parameters, and thus determine whether to perform intelligent adjustments to the point cloud mapping process of the high-precision stacker; The obstacle avoidance process adjustment module is used to obtain and hybrid guide the high-precision stacker to avoid obstacles based on the point cloud mapping results of the high-precision stacker, collect and analyze the obstacle avoidance decision parameters of the high-precision stacker, and thus determine whether to adjust the obstacle avoidance process of the high-precision stacker; The specific determination process of whether to adjust the obstacle avoidance process of the high-precision stacker is as follows: Analyze the obstacle avoidance decision parameters of the high-precision stacker, obtain the obstacle avoidance decision efficiency index of the high-precision stacker, and compare it with the obstacle avoidance decision efficiency threshold. If the obstacle avoidance decision efficiency index of the high-precision stacker is greater than or equal to the obstacle avoidance decision efficiency threshold, determine not to adjust the obstacle avoidance process of the high-precision stacker, and at the same time determine whether a third condition exists. If the third condition exists, determine not to intelligently restore the obstacle avoidance process of the high-precision stacker. If the third condition does not exist, determine to intelligently restore the obstacle avoidance process of the high-precision stacker, specifically, according to the decision recovery speed preset in the intelligent database, restore the operating speed of the high-precision stacker to the basic operating speed, and restore the safety margin of the high-precision stacker to the basic safety margin; If the obstacle avoidance decision efficiency index of the high-precision stacker is less than the obstacle avoidance decision efficiency threshold, it is determined that the obstacle avoidance process of the high-precision stacker should be adjusted; The third condition means that the obstacle avoidance decision effectiveness margin value of the high-precision stacker is greater than or equal to the defined obstacle avoidance decision effectiveness margin value, and at the same time, the operating speed of the high-precision stacker is the basic operating speed and the safety margin of the high-precision stacker is the basic safety margin; The obstacle avoidance process of the high-precision stacker is adjusted, and the specific adjustment process is as follows: Determine whether the fourth condition exists. If so, make a decision and give a warning on the obstacle avoidance process of the high-precision stacker. If the fourth condition does not exist, obtaining and adjusting the operating speed of the high-precision stacker to decrease and the safety margin of the high-precision stacker to increase based on the obstacle avoidance decision efficiency deviation value of the high-precision stacker; The fourth condition is that the operating speed of the high-precision stacker is the minimum operating speed and the safety margin of the high-precision stacker is the maximum safety margin.

2. The high-precision stacker crane intelligent obstacle avoidance system based on hybrid guidance according to claim 1, characterized in that: The specific determination process of whether to intelligently adjust the laser reflection process of the high-precision stacker is as follows: Analyze the laser reflection parameters of the high-precision stacker to obtain the laser reflection index of the high-precision stacker, and compare the obtained index with the laser reflection reference interval. If the laser reflection index of the high-precision stacker falls within the laser reflection reference interval, determine not to perform intelligent adjustment on the laser reflection process of the high-precision stacker, and determine whether a first condition exists. If the first condition exists, determine not to perform intelligent restoration on the laser reflection process of the high-precision stacker. If the first condition does not exist, determine to perform intelligent restoration on the laser reflection process of the high-precision stacker, specifically, restore the amplifier gain of the high-precision stacker to the basic amplifier gain and restore the exposure time of the high-precision stacker to the basic exposure time according to the reflection recovery speed preset in the intelligent database. If the laser reflection index of the high-precision stacker does not fall within the laser reflection reference range, it is determined that the laser reflection process of the high-precision stacker is intelligently adjusted, specifically: If the laser reflection index of the high-precision stacker is greater than the maximum value of the laser reflection reference range, the amplifier gain and exposure time of the high-precision stacker are reduced and adjusted based on the first laser reflectivity of the high-precision stacker, and the area to which the high-precision stacker belongs is marked as a high-reflection area; If the laser reflection index of the high-precision stacker is less than the minimum value of the laser reflection reference interval, the amplifier gain and exposure time of the high-precision stacker are increased and adjusted based on the second laser reflectivity of the high-precision stacker, and the area to which the high-precision stacker belongs is marked as a low-reflection area. The first condition refers to the amplifier gain of the high-precision stacker as the basic amplifier gain and the exposure time of the high-precision stacker as the basic exposure time.

3. The high-precision stacker crane intelligent obstacle avoidance system based on hybrid guidance according to claim 2, characterized in that: The specific analysis process of the laser reflection index of the high-precision stacker is as follows: The laser reflection parameters of the high-precision stacker include the laser reflection frequency of the high-precision stacker, the laser reflectivity of the high-precision stacker, and the speckle contrast of the high-precision stacker; Importance coefficients are introduced to quantify the influence of the deviation between the laser reflection frequency and the reference laser reflection frequency on the laser reflection index, the deviation between the laser reflectivity and the reference laser reflectivity on the laser reflection index, and the deviation between the speckle contrast and the reference speckle contrast on the laser reflection index. The influence of each degree is summarized to obtain the laser reflection index of the high-precision stacker. The laser reflection index of the high-precision stacker is a numerical value used to measure the comprehensive situation of the laser reflection characteristics of the stacker.

4. The high-precision stacker crane intelligent obstacle avoidance system based on hybrid guidance according to claim 1, characterized in that: The determination of whether to perform intelligent adjustment on the point cloud mapping process of the high-precision stacker is carried out by the following specific analysis process: Analyze the point cloud mapping parameters to obtain the laser point cloud mapping accuracy coefficient of the high-precision stacker, and compare it with the laser point cloud mapping accuracy threshold. If the laser point cloud mapping accuracy coefficient of the high-precision stacker is greater than or equal to the laser point cloud mapping accuracy threshold, determine not to perform intelligent adjustment on the point cloud mapping process of the high-precision stacker, and at the same time determine whether a second condition exists. If the second condition exists, determine not to perform intelligent recovery on the point cloud mapping process of the high-precision stacker. If the second condition does not exist, determine to perform intelligent recovery on the point cloud mapping process of the high-precision stacker, specifically, according to the mapping recovery speed preset in the intelligent database, restore the number of plane fitting iterations of the high-precision stacker to the basic plane fitting iteration number, and restore the reflectivity mutation response speed of the high-precision stacker to the basic reflectivity mutation response speed; If the laser point cloud mapping accuracy coefficient of the high-precision stacker is less than the laser point cloud mapping accuracy threshold, it is determined that the point cloud mapping process of the high-precision stacker is intelligently adjusted; The second condition means that the laser point cloud mapping accuracy margin value of the high-precision stacker is greater than or equal to the defined laser point cloud mapping accuracy margin value, and at the same time, the plane fitting iteration number of the high-precision stacker is the basic plane fitting iteration number and the reflectivity mutation response speed of the high-precision stacker is the basic reflectivity mutation response speed.

5. The high-precision stacker crane intelligent obstacle avoidance system based on hybrid guidance according to claim 4, characterized in that: The point cloud mapping process of the high-precision stacker is intelligently adjusted, and the specific adjustment process is as follows: Obtain the reflection type of the area to which the high-precision stacker belongs and the accurate deviation value of the laser point cloud mapping of the high-precision stacker. If the area to which the high-precision stacker belongs is a high-reflection area, increase and adjust the number of plane fitting iterations and the reflectivity mutation response speed of the high-precision stacker based on the accurate deviation value of the laser point cloud mapping; If the area to which the high-precision stacker belongs is a low-reflection area, the number of plane fitting iterations and the reflectivity mutation response speed of the high-precision stacker are reduced and adjusted based on the accurate deviation value of the laser point cloud mapping.

6. The high-precision stacker crane intelligent obstacle avoidance system based on hybrid guidance according to claim 4, characterized in that: The accuracy coefficient of the laser point cloud mapping of the high-precision stacker is analyzed as follows: The point cloud mapping parameters include the laser point cloud distribution density of the high-precision stacker, the laser point cloud mapping time deviation of the high-precision stacker, and the laser point cloud angular resolution of the high-precision stacker; According to the laser reflection index of the high-precision stacker, the correction coefficient is matched from the intelligent database; By introducing weighted quantification to determine the influence of the deviation between the laser point cloud distribution density and the reference laser point cloud distribution density on the laser point cloud mapping accuracy coefficient, the influence of the proportional relationship between the laser point cloud mapping time deviation and the defined laser point cloud mapping time deviation on the laser point cloud mapping accuracy coefficient, and the influence of the proportional relationship between the laser point cloud angular resolution and the defined laser point cloud angular resolution on the laser point cloud mapping accuracy coefficient, each influence degree is aggregated, and a correction coefficient is introduced to correct the aggregation result, thereby obtaining the laser point cloud mapping accuracy coefficient of the high-precision stacker; The laser point cloud mapping accuracy coefficient of the high-precision stacker is used to quantify the accuracy of the laser point cloud mapping of the high-precision stacker.

7. The high-precision stacker crane intelligent obstacle avoidance system based on hybrid guidance according to claim 1, characterized in that: The specific analysis process of the obstacle avoidance decision-making efficiency index of the high-precision stacker is as follows: The obstacle avoidance decision parameters of the high-precision stacker include the instruction response time of the high-precision stacker and the obstacle avoidance action accuracy factor of the high-precision stacker; The importance values ​​are used to quantify the influence of the laser point cloud mapping accuracy coefficient, the proportional relationship between the command response time and the defined command response time, and the proportional relationship between the obstacle avoidance action accuracy factor and the defined obstacle avoidance action accuracy factor on the obstacle avoidance decision effectiveness index. The influence degrees are summarized and the summary results are corrected by the correction coefficient to obtain the obstacle avoidance decision effectiveness index of the high-precision stacker. The obstacle avoidance decision-making efficiency index of the high-precision stacker is used to quantify the obstacle avoidance decision-making efficiency of the high-precision stacker.

8. A method for the high-precision stacker crane intelligent obstacle avoidance system based on hybrid guidance according to any one of claims 1 to 7, characterized in that: include: Step 1: Collect and analyze the laser reflection parameters of the high-precision stacker to determine whether to perform intelligent adjustment on the laser reflection process of the high-precision stacker; Step 2: Collect the laser reflection results of the high-precision stacker, perform point cloud mapping, obtain and analyze the point cloud mapping parameters, and determine whether to intelligently adjust the point cloud mapping process of the high-precision stacker; Step 3: Obtain and hybrid guide the high-precision stacker to avoid obstacles based on the point cloud mapping results of the high-precision stacker, collect and analyze the obstacle avoidance decision parameters of the high-precision stacker, and determine whether to adjust the obstacle avoidance process of the high-precision stacker.

Images