Intelligent material table robot system based on AI vision technology

Abstract

The invention relates to the technical field of material table robots, in particular to an intelligent material table robot system based on an AI vision technology, which is provided with a material table control module, a feature acquisition module, a material analysis module, a material identification module and a grabbing regulation and control module, a material analysis module determines the gray fluctuation condition of to-be-sorted materials according to gray characterization parameters in a grabbed recognition area at different illumination angles so as to mark the feature grabbed recognition area, a material recognition module determines a shielding trend sub-area, a shielding trend sub-vector is constructed, and the shielding trend sub-vector is calculated. And a grabbing compensation mode of the to-be-sorted materials is determined through a grabbing regulation and control module according to the comparison condition of the conveying direction vector and the shielding trend vector. According to the method, the grabbing recognition area with shielding interference is marked according to the gray features of the materials, the grabbing compensation mode is adaptively adjusted based on the change trend of dynamic shielding, and the efficiency and reliability of an intelligent material table robot system are improved.

Description

Technical Field

[0001] The present invention relates to the field of material platform robot technology, and in particular to an intelligent material platform robot system based on AI vision technology. Background Art

[0002] In industrial automation and intelligent manufacturing scenarios, vision-based pallet robot systems serve as the core link between material storage and production execution. They are widely used in scenarios such as electronics manufacturing, automotive parts assembly, and logistics sorting. They undertake the automated tasks of material identification, positioning, grasping, and conveying. AI vision systems collect image information from the pallet area and combine it with deep learning algorithms to accurately identify and estimate the position of materials. Actuators such as robotic arms then complete grasping and sorting, replacing manual operations and improving production efficiency and stability. However, the lighting environment in industrial workshops is complex and changeable. The superposition of natural and artificial lighting, the dynamic changes in equipment shadows, and the reflective properties of the material itself can affect the accuracy of visual recognition. Furthermore, the stacking of materials on the pallet, occlusion caused by the motion of the robotic arm itself, and the temporary intervention of peripheral equipment can cause material obstruction. This makes it difficult to adaptively adjust the grasping method according to the actual situation in the dynamic pallet scene, affecting the recognition accuracy and response efficiency of the intelligent pallet robot system based on AI vision technology. Therefore, improving the efficiency and reliability of intelligent pallet robot systems is a technical problem that needs to be solved urgently.

[0003] For example, China Patent Authorization Announcement No.: CN115366152B, the invention discloses a robot vision automated grasping system, including a transport mechanism and a material dividing mechanism; the transport mechanism includes a workbench, a transport device and a grasping assembly for grasping materials on the transport device, the transport device and the grasping assembly are both installed on the workbench, and a capture camera is provided on the grasping assembly; the material dividing mechanism includes a material dividing room, a magnetic suction device and a discharge conveyor belt and a recovery conveyor belt respectively located on both sides of the transport device, the material dividing room cover is located above the discharge end of the transport device, the magnetic suction device is installed in the material dividing room and is used to transfer materials from the transport device to the discharge conveyor belt or the recovery conveyor belt, and the material dividing room is provided with a discharge camera and a recovery camera.

[0004] The following problems also exist in the prior art: The existing technology does not take into account the dynamic interference of light and occlusion in the industrial environment, which affects the recognition and grasping of materials by the material table robot. The existing technology cannot mark the grasping and recognition areas where occlusion interference exists according to the grayscale characteristics of the material, and cannot adaptively adjust the grasping compensation method according to the changing trend of dynamic occlusion, affecting the efficiency and reliability of the intelligent material table robot system. Summary of the Invention

[0005] To this end, the present invention provides an intelligent material table robot system based on AI vision technology to overcome the problems that the existing technology cannot mark the grasping and identification area with occlusion interference according to the grayscale characteristics of the material, and cannot adaptively adjust the grasping compensation method according to the changing trend of dynamic occlusion, which affects the efficiency and reliability of the intelligent material table robot system.

[0006] To achieve the above objectives, the present invention provides an intelligent platform robot system based on AI vision technology, comprising: A material platform control module, comprising a grabbing unit for grabbing the materials to be sorted on the material platform and a conveying unit for conveying the materials to be sorted; A feature acquisition module, connected to the material platform control module, for acquiring the grayscale characterization parameters and conveying direction vector of the material to be sorted; A material analysis module, connected to the feature acquisition module, is used to divide the material platform into several grabbing and identification areas, and determine the grayscale fluctuation of the material to be sorted based on the grayscale characterization parameters within the grabbing and identification areas under different lighting angles to mark the feature grabbing and identification areas; a material recognition module, connected to the feature acquisition module and the material analysis module, respectively, determining an occlusion trend sub-region based on the grayscale characterization parameters of the feature capture and recognition region under a preset illumination angle, and constructing an occlusion trend sub-vector based on the obtained occlusion trend sub-regions; a grabbing control module, which is respectively connected to the material platform control module, the feature acquisition module, and the material identification module, and is used to determine a grabbing compensation method for the material to be sorted by adjusting the grabbing time for grabbing the material to be sorted in the feature grabbing identification area according to the comparison between the conveying direction vector and the occlusion trend vector; or determining whether there is a grasping abnormality risk in the feature grasping and identifying area according to the occlusion trend vector, and adjusting a grasping compensation parameter of the feature grasping and identifying area; The occlusion tendency vector is determined according to a plurality of occlusion tendency sub-vectors.

[0007] Furthermore, the material analysis module is used to determine the grayscale fluctuation characterization amount and the occlusion tendency coefficient based on the grayscale characterization parameters within the grasping and identifying area, wherein: The grayscale fluctuation characterization value is the difference between the maximum value and the minimum value of the grayscale characterization parameter in the grasping and recognition area under the same illumination angle; The occlusion tendency coefficient is the variance of the grayscale characterization parameter under different illumination angles.

[0008] Furthermore, the material analysis module is used to mark the grabbing and identifying area as a characteristic grabbing and identifying area based on the determination result that the grayscale fluctuation characterization quantity and the occlusion tendency coefficient in the grabbing and identifying area under different illumination angles meet the characteristic grabbing and identifying area conditions, wherein, The condition for feature capture and identification area is that the grayscale fluctuation characterization value exceeds a preset grayscale fluctuation characterization value threshold, and the occlusion tendency coefficient does not exceed a preset occlusion tendency coefficient threshold.

[0009] Furthermore, the material identification module is used to determine the characteristic sub-region as the occlusion tendency sub-region based on the determination result that the grayscale characterization parameter of the characteristic sub-region of the characteristic capture and identification region meets the occlusion tendency sub-region condition, wherein, The material recognition module divides the feature capture and recognition area into several sub-areas and obtains grayscale characterization parameters at several positions of the material to be sorted in the sub-areas; The occlusion tendency sub-region condition is that the grayscale characterization parameter of the characteristic sub-region does not exceed a preset grayscale characterization parameter threshold, and the characteristic sub-region is the sub-region where the grayscale characterization parameter has the minimum value in the feature capture and recognition region.

[0010] Furthermore, the material identification module is used to construct an occlusion trend sub-vector of the latter of the adjacent collection moments based on the occlusion trend sub-regions at adjacent collection moments within a preset monitoring period, wherein: The material recognition module obtains the occlusion trend sub-area at several acquisition moments within the feature capture and recognition area; The occlusion trend sub-vector is constructed with the area center point of the occlusion trend sub-region at the previous acquisition moment in adjacent acquisition moments as the vector starting point of the occlusion trend sub-vector, and with the area center point of the occlusion trend sub-region at the next acquisition moment in adjacent acquisition moments as the vector end point of the occlusion trend sub-vector.

[0011] Furthermore, the grabbing control module is used to determine a grabbing compensation method for the materials to be sorted in the characteristic grabbing identification area, wherein: If the conveying direction vector and the occlusion tendency vector of the material to be sorted in the characteristic grasping and identifying area meet the first grasping compensation condition, the grasping control module determines that the grasping compensation method is to adjust the grasping time for grasping the material to be sorted in the characteristic grasping and identifying area; If the conveying direction vector and the occlusion tendency vector of the material to be sorted in the feature grasping and identification area do not meet the first grasping compensation condition, the grasping control module determines the grasping compensation method by determining whether there is a grasping abnormality risk in the feature grasping and identification area, and adjusting the grasping compensation parameters.

[0012] Furthermore, the first grasping compensation condition is that the relative offset tendency parameter exceeds a preset relative offset tendency parameter threshold, the relative offset tendency parameter is the vector angle between the transmission direction vector and the occlusion tendency vector, and the occlusion tendency vector is the vector obtained by adding the occlusion tendency sub-vectors at several monitoring moments.

[0013] Furthermore, the capture control module is used to adjust the capture time, wherein: The delay duration of the grabbing moment is negatively correlated with the relative deviation tendency parameter.

[0014] Furthermore, the capture control module is used to determine whether there is a capture abnormality risk in the feature capture identification area, wherein: The grasping control module determines that there is a grasping abnormality risk in the feature grasping identification area based on a determination result that the occlusion trend vector of the feature grasping identification area meets the grasping abnormality condition; Based on the result of determining that the occlusion trend vector of the characteristic grasping and identifying area does not meet the grasping abnormality condition, it is determined that there is no grasping abnormality risk in the characteristic grasping and identifying area, and the material to be sorted in the characteristic grasping and identifying area is grasped based on the grasping compensation parameter; The abnormal capture condition is that the risk tendency parameter exceeds a preset risk tendency parameter threshold, and the risk tendency parameter is the absolute value of the difference between the module lengths of the occlusion trend vectors at adjacent acquisition moments.

[0015] Furthermore, the grasping control module is used to determine grasping compensation parameters, and the grasping compensation parameters include grasping direction and grasping speed, wherein: The grabbing direction is the vector direction of the material tendency vector, and the material tendency vector is the vector obtained by adding the blocking tendency vector and the conveying direction vector; The reduction in the grabbing speed is positively correlated with the vector size of the occlusion trend vector. Compared with the prior art, the beneficial effect of the present invention is that the present invention sets a material platform control module, a feature acquisition module, a material analysis module, a material identification module, and a grabbing control module, obtains the grayscale characterization parameters and the transmission direction vector of the material to be sorted by the feature acquisition module, determines the grayscale fluctuation of the material to be sorted according to the grayscale characterization parameters in the grabbing and identification area under different illumination angles by the material analysis module, and marks the feature grabbing and identification area, determines the occlusion trend sub-area according to the grayscale characterization parameters of the feature grabbing and identification area by the material identification module, constructs the occlusion trend sub-vector based on the obtained several occlusion trend sub-areas, determines the grabbing compensation method for the material to be sorted according to the comparison between the transmission direction vector and the occlusion trend vector by the grabbing control module, and then realizes the grabbing and identification area with occlusion interference marked according to the grayscale feature of the material, and adaptively adjusts the grabbing compensation method based on the changing trend of dynamic occlusion, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0016] In particular, the present invention determines the grayscale fluctuation of the material to be sorted according to the grayscale characterization parameters in the grasping and identification area under different lighting angles through the material analysis module, so as to mark the feature grasping and identification area. It can be understood that the grayscale fluctuation characterization quantity is used to screen out areas with obvious characteristics to avoid invalid analysis of areas with no difference. At the same time, the occlusion tendency coefficient is used to exclude areas affected by lighting interference, thereby reducing misjudgments caused by lighting changes, such as shadows being misjudged as material occlusions. Even in scenes with unstable lighting angles, such as changes in natural light in the workshop and jitter of equipment light sources, the material feature areas can still be stably identified, the system's ability to resist environmental interference is enhanced, and reliable analysis objects are provided for subsequent grasping and regulation. By screening out key feature grasping and identification areas, the full analysis of the entire material platform area is avoided, invalid data processing is reduced, and the system operation efficiency is improved. Furthermore, the grasping and identification areas with occlusion interference are marked according to the grayscale characteristics of the material, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0017] In particular, the present invention determines the occlusion trend sub-region according to the grayscale characterization parameters of the feature grasping and identifying area through the material recognition module, and constructs the occlusion trend sub-vector based on the obtained several occlusion trend sub-regions. It can be understood that the occlusion trend sub-region is determined by the grayscale value of the sub-region within the feature grasping and identifying area, and the area where the occlusion shadow exists in the feature grasping and identifying area is locked. By quantifying the moving direction and amplitude of the occlusion shadow, its dynamic changes are reflected in real time, providing predictive information for the intelligent material platform robot, capturing dynamic occlusion trends, and improving the system's adaptability to dynamic environments. The present invention determines the occlusion trend sub-region according to the grayscale characterization parameters of the feature grasping and identifying area through the material recognition module, and constructs the occlusion trend sub-vector based on the obtained several occlusion trend sub-regions. Furthermore, the occlusion trend sub-region is screened, and the occlusion trend sub-vector is constructed, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0018] In particular, the present invention determines the gripping compensation method for the material to be sorted according to the comparison between the transmission direction vector and the occlusion trend vector through the gripping control module. It can be understood that the duration and risk level of the occlusion interference can be predicted through the relative movement trend of the occlusion object and the material, so as to select a targeted response strategy. The transmission direction vector represents the movement direction of the material to be sorted, such as the moving direction of the conveyor belt and the pushing direction of the material table, reflecting the spatial position change law of the material itself. The occlusion trend vector is determined by the occlusion trend sub-vectors at several monitoring moments, representing the occlusion objects such as other stacked materials and the overall moving robot arm. The motion trend reflects the spatial position change law of the occlusion source. The angle between the two vectors can characterize the relative motion relationship between the occlusion and the material. The larger the angle, the more significant the difference in the motion directions of the two. The smaller the angle, the closer the motion directions of the two are. Different grasping compensation methods are adaptively adjusted for different comparison situations to improve the accuracy of grasping timing, reduce the risk of collision under continuous occlusion, balance efficiency and safety, and enhance the system's adaptability to dynamic scenes. Furthermore, the grasping compensation method is adaptively adjusted based on the changing trend of dynamic occlusion, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0019] In particular, the present invention adjusts the grabbing time of grabbing the material to be sorted in the feature grabbing identification area when the feature grabbing identification area meets the first grabbing compensation condition through the grabbing control module. It can be understood that in the actual material platform sorting, the movement speed and direction of the obstruction may change dynamically. By adjusting the delay in real time through the vector angle, the system can adapt to the obstruction rhythm under different working conditions. The feature grabbing identification area meets the first grabbing compensation condition, that is, the relative movement trend of the obstruction and the material shows an inconsistent trend. The larger the relative offset tendency parameter, the faster the obstruction is released. Shortening the delay time can reduce The residence time of materials on the conveyor belt can avoid subsequent material accumulation due to excessive waiting, increase the sorting volume per unit time, avoid ineffective waiting, and improve grasping efficiency. The smaller the relative offset tendency parameter, the slower the speed of occlusion removal. Extending the delay time can avoid the robot arm accidentally touching the obstruction due to premature grasping, reduce the risk of material damage and equipment failure, ensure the removal of obstructions, improve grasping accuracy, reduce energy consumption and mechanical loss, and dynamically adapt to the occlusion rhythm. Furthermore, it realizes the adaptive adjustment of the grasping compensation method based on the changing trend of dynamic occlusion, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0020] In particular, the present invention determines whether there is a risk of abnormal grasping in the feature grasping identification area when the feature grasping identification area does not meet the first grasping compensation condition through the grasping control module, and adjusts the grasping compensation parameters. It can be understood that the feature grasping identification area does not meet the first grasping compensation condition, that is, the relative movement trend of the occlusion and the material presents a relatively consistent trend, and the shadow cast by the occlusion on the material will not change due to the material transmission. By monitoring the rate of change of the occlusion trend vector modulus, that is, the risk tendency parameter, the risk of changing the occlusion state can be quickly identified, such as sudden large-scale occlusion, material position offset caused by rapid movement of the occlusion, and in the case of abnormal risks, it can be paused or warned in advance to avoid grasping when the occlusion is unstable. The grasping direction is determined based on the material tendency vector, which is equivalent to incorporating the indirect impact of the occlusion on the material into the trajectory calculation, so that the grasping can be carried out smoothly. The action tracks the actual position of the material more accurately, and the grasping speed adjustment amount is determined according to the module length of the occlusion trend vector, achieving a dynamic balance of giving priority to precision when the occlusion influence is strong and giving priority to efficiency when the influence is weak. The larger the vector size of the occlusion trend vector, the more a large speed adjustment is required, such as deceleration to improve positioning accuracy, or acceleration to leave the occlusion area, to ensure that the grasping action can still be stably executed under interference. The smaller the vector size of the occlusion trend vector, the smaller the speed adjustment amplitude, avoiding the decrease in sorting efficiency caused by excessive deceleration. Through the combined strategy of risk judgment and dynamic parameter adjustment, the intelligent material platform robot system can maintain efficient grasping under dynamic occlusion conditions, taking into account both grasping accuracy and efficiency while effectively avoiding potential abnormal risks. Furthermore, it realizes the adaptive adjustment of the grasping compensation method based on the changing trend of dynamic occlusion, thereby improving the efficiency and reliability of the intelligent material platform robot system. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a functional block diagram of an intelligent platform robot system based on AI vision technology according to an embodiment of the present invention; Figure 2 This is a logic flow chart of the marking feature capture and identification area of ​​the material analysis module according to an embodiment of the present invention; Figure 3 A logic flow chart of determining an occlusion trend sub-area by a material recognition module according to an embodiment of the present invention; Figure 4 This is a logic flow chart of the grasping control module in an embodiment of the present invention determining the grasping compensation method for the materials to be sorted within the feature grasping recognition area. DETAILED DESCRIPTION

[0022] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0023] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0024] It should be noted that, in the description of the present invention, terms such as "upper", "lower", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0025] Furthermore, it should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the terms "mounted" and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; or internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0026] See also Figure 1 As shown in FIG, it is a functional block diagram of an intelligent platform robot system based on AI vision technology according to an embodiment of the present invention. An intelligent platform robot system based on AI vision technology according to the present invention includes: A material platform control module, comprising a grabbing unit for grabbing the materials to be sorted on the material platform and a conveying unit for conveying the materials to be sorted; Specifically, the embodiments of the present invention do not limit the specific structures of the grasping unit and the conveying unit. Preferably, the grasping unit can be a multi-degree-of-freedom robotic arm with an end effector, such as a pneumatic gripper, a vacuum suction cup, or a magnetic gripper, for grasping the materials to be sorted on the material table, and the conveying unit can be a belt conveyor for unidirectionally conveying the materials to be sorted, which will not be repeated here.

[0027] A feature acquisition module, connected to the material platform control module, for acquiring the grayscale characterization parameters and conveying direction vector of the material to be sorted; Specifically, the embodiment of the present invention does not limit the specific structure of the feature acquisition module. Preferably, it can be an industrial camera combined with a microprocessor to obtain the grayscale characterization parameters and conveying direction vector of the material to be sorted. The vector direction of the conveying direction vector is the conveying direction of the conveying mechanism where the material to be sorted is located, such as a belt conveyor. This will not be repeated.

[0028] A material analysis module, connected to the feature acquisition module, is used to divide the material platform into several grabbing and identification areas, and determine the grayscale fluctuation of the material to be sorted based on the grayscale characterization parameters within the grabbing and identification areas under different lighting angles to mark the feature grabbing and identification areas; Specifically, the area of ​​the grasping and identification area is the product of the material platform area and the first area division factor. The first area division factor can be set by technical personnel in this field according to the accuracy requirements of the intelligent material platform robot system. The higher the accuracy requirement, the smaller the first area division factor is set. The value range of the first area division factor can be [0.1, 0.3]. Preferably, it can be 0.2.

[0029] Specifically, the single change amount and number of changes of the illumination angle can be set by technical personnel in this field according to the accuracy requirements of the intelligent material table robot system. The higher the accuracy requirement, the smaller the single change amount and the greater the number of changes. The value range of the single change amount can be [5, 10], and the interval unit is °. The value range of the number of changes can be [3, 8], and the interval unit is times. Preferably, the single change amount can be 6° and the number of changes can be 5 times.

[0030] Specifically, the embodiment of the present invention does not limit the specific structure of the material analysis module. Preferably, it can be a processor used in a computer to determine the grayscale fluctuation of the material to be sorted and mark the feature capture and identification area, which will not be repeated here.

[0031] a material recognition module, connected to the feature acquisition module and the material analysis module, respectively, determining an occlusion trend sub-region based on the grayscale characterization parameters of the feature capture and recognition region under a preset illumination angle, and constructing an occlusion trend sub-vector based on the obtained occlusion trend sub-regions; Specifically, the embodiment of the present invention does not limit the specific structure of the material identification module. Preferably, it can be a microprocessor for determining the occlusion trend sub-region and constructing the occlusion trend sub-vector, which will not be repeated here.

[0032] Specifically, the preset illumination angle can be determined by technicians in this field based on the visual recognition accuracy of materials to be grasped of the same material in historical data. The value range of the preset illumination angle can be [30, 60], and the interval unit is °. Preferably, it can be 45°.

[0033] a grabbing control module, which is respectively connected to the material platform control module, the feature acquisition module, and the material identification module, and is used to determine a grabbing compensation method for the material to be sorted by adjusting the grabbing time for grabbing the material to be sorted in the feature grabbing identification area according to the comparison between the conveying direction vector and the occlusion trend vector; or determining whether there is a grasping abnormality risk in the feature grasping and identifying area according to the occlusion trend vector, and adjusting a grasping compensation parameter of the feature grasping and identifying area; The occlusion tendency vector is determined according to a plurality of occlusion tendency sub-vectors.

[0034] Specifically, the embodiment of the present invention does not limit the specific structure of the grasping control module. Preferably, it can be a microprocessor to determine the grasping compensation method for the sorted material, adjust the grasping time, and grasping compensation parameters, which will not be repeated here.

[0035] Specifically, the material analysis module is used to determine the grayscale fluctuation characterization amount and the occlusion tendency coefficient based on the grayscale characterization parameters within the grasping and identifying area, wherein: The grayscale fluctuation characterization value is the difference between the maximum value and the minimum value of the grayscale characterization parameter in the grasping and recognition area under the same illumination angle; The occlusion tendency coefficient is the variance of the grayscale characterization parameter under different illumination angles.

[0036] See also Figure 2 As shown, it is a logical flow chart of the material analysis module marking the feature grabbing and identifying area in an embodiment of the present invention. The material analysis module is used to mark the grabbing and identifying area as the feature grabbing and identifying area based on the determination result that the grayscale fluctuation characterization quantity and the occlusion tendency coefficient in the grabbing and identifying area under different illumination angles meet the feature grabbing and identifying area conditions, wherein, If the grayscale fluctuation characterization quantity and the occlusion tendency coefficient within the grasping and identifying area under different illumination angles do not meet the characteristic grasping and identifying area conditions, the grasping and identifying area is not marked; The condition for feature capture and identification area is that the grayscale fluctuation characterization value exceeds a preset grayscale fluctuation characterization value threshold, and the occlusion tendency coefficient does not exceed a preset occlusion tendency coefficient threshold.

[0037] Specifically, the preset grayscale fluctuation characterization value threshold is the product of the grayscale fluctuation characterization value reference value and the grayscale fluctuation coefficient, the preset occlusion tendency coefficient threshold is the product of the occlusion tendency coefficient reference value and the occlusion tendency factor, the grayscale fluctuation characterization value reference value is the average value of the grayscale fluctuation characterization value under the same working conditions in the historical data, and the occlusion tendency coefficient reference value is the average value of the occlusion tendency coefficient under the same working conditions in the historical data. The grayscale fluctuation coefficient and the occlusion tendency factor can be set by technical personnel in this field according to the accuracy requirements of the intelligent material table robot system. The higher the accuracy requirement, the larger the grayscale fluctuation coefficient is set, and the smaller the occlusion tendency factor is set. The value range of the grayscale fluctuation coefficient can be [1.2, 1.3], and the value range of the occlusion tendency factor can be [1.1, 1.25]. Preferably, the grayscale fluctuation coefficient can be 1.25, and the occlusion tendency factor can be 1.15.

[0038] Specifically, the embodiment of the present invention determines the grayscale fluctuation of the material to be sorted according to the grayscale characterization parameters in the grasping and identification area under different lighting angles through the material analysis module, so as to mark the feature grasping and identification area. It can be understood that the grayscale fluctuation characterization quantity is used to screen out areas with obvious characteristics to avoid invalid analysis of areas with no difference. At the same time, the occlusion tendency coefficient is used to exclude areas affected by lighting interference, thereby reducing misjudgments caused by lighting changes, such as shadows being misjudged as material occlusions. Even in scenes with unstable lighting angles, such as changes in natural light in the workshop and jitter of the equipment light source, the material feature area can still be stably identified, the system's ability to resist environmental interference is enhanced, and reliable analysis objects are provided for subsequent grasping and regulation. By screening out key feature grasping and identification areas, the full analysis of the entire material platform area is avoided, invalid data processing is reduced, and the system operation efficiency is improved. Furthermore, the grasping and identification areas with occlusion interference are marked according to the grayscale characteristics of the material, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0039] Specifically, it can be understood that based on the differentiated performance of the material's own characteristics and environmental interference (light, occlusion), key areas are screened out through quantitative analysis to provide reliable analysis objects for subsequent accurate identification and grasping. The grayscale fluctuation characterization quantity is the difference between the maximum and minimum values ​​of the grayscale characterization parameters in the grasping and recognition area under the same lighting angle, which can characterize whether there is a shadow phenomenon in the area. The larger the grayscale fluctuation characterization quantity, the more likely there is a shadow phenomenon in the area, and the area with shadow phenomenon is screened out. The occlusion tendency coefficient is the variance of the grayscale characterization parameters under different lighting angles, which can characterize the sensitivity of the regional grayscale characteristics to changes in lighting angle. If the shadow is caused by occlusion interference, its position and grayscale characteristics are relatively stable, and are less affected by changes in lighting angle, and the occlusion tendency coefficient is small. If the shadow is caused by changes in lighting angle, the grayscale characteristics will fluctuate with the lighting angle, and the occlusion tendency coefficient is large. In turn, the grasping and recognition areas with occlusion interference are marked according to the grayscale characteristics of the material, thereby improving the efficiency and reliability of the intelligent material table robot system.

[0040] See also Figure 3 As shown, it is a logic flow chart of the material recognition module determining the occlusion trend sub-region according to an embodiment of the present invention. The material recognition module is used to determine the characteristic sub-region as the occlusion trend sub-region based on the determination result that the grayscale representation parameter of the characteristic sub-region of the feature capture recognition region meets the occlusion trend sub-region condition, wherein, The material recognition module divides the feature capture and recognition area into several sub-areas and obtains grayscale characterization parameters at several positions of the material to be sorted in the sub-areas; If the grayscale representation parameter of the feature sub-region of the feature capture and recognition region does not meet the occlusion trend sub-region condition, the feature sub-region is not screened; The occlusion tendency sub-region condition is that the grayscale characterization parameter of the characteristic sub-region does not exceed a preset grayscale characterization parameter threshold, and the characteristic sub-region is the sub-region where the grayscale characterization parameter has the minimum value in the feature capture and recognition region.

[0041] Specifically, the preset grayscale characterization parameter threshold is the product of the grayscale characterization parameter reference value and the grayscale characterization factor. The grayscale characterization parameter reference value is the average value of the grayscale characterization parameters of the same working conditions in the historical data. The grayscale characterization factor can be set by technical personnel in this field according to the accuracy requirements of the intelligent material platform robot system. The higher the accuracy requirement, the smaller the grayscale characterization factor is set. The value range of the grayscale characterization factor can be [1.12, 1.25]. Preferably, it can be 1.15.

[0042] Specifically, the divided area of ​​the sub-area is the product of the area of ​​the feature grasping and identification area and the second area division factor. The second area division factor can be set by technical personnel in this field according to the accuracy requirements of the intelligent material platform robot system. The higher the accuracy requirement, the smaller the second area division factor is set. The value range of the second area division factor can be [0.1, 0.3]. Preferably, it can be 0.2.

[0043] Specifically, the material identification module is used to construct an occlusion trend sub-vector of the latter of the adjacent acquisition moments based on the occlusion trend sub-regions at adjacent acquisition moments within a preset monitoring period, wherein: The material recognition module obtains the occlusion trend sub-area at several acquisition moments within the feature capture and recognition area; The occlusion trend sub-vector is constructed with the area center point of the occlusion trend sub-region at the previous acquisition moment in adjacent acquisition moments as the vector starting point of the occlusion trend sub-vector, and with the area center point of the occlusion trend sub-region at the next acquisition moment in adjacent acquisition moments as the vector end point of the occlusion trend sub-vector.

[0044] Specifically, the preset monitoring period and the interval between adjacent collection moments in the preset monitoring period can be set by technical personnel in this field according to the accuracy requirements of the intelligent material platform robot system. The higher the accuracy requirement, the shorter the preset monitoring period and the shorter the interval. The value range of the preset monitoring period can be [20, 40], and the interval unit is min. The value range of the interval length can be [30, 50], and the interval unit is s. Preferably, the preset monitoring period can be 30 minutes and the interval length can be 45 seconds.

[0045] Specifically, an embodiment of the present invention determines the occlusion trend sub-region according to the grayscale characterization parameters of the feature grasping and identifying area through a material recognition module, and constructs an occlusion trend sub-vector based on the obtained several occlusion trend sub-regions. It can be understood that the occlusion trend sub-region is determined by the grayscale value of the sub-region within the feature grasping and identifying area, and the area where the occlusion shadow exists in the feature grasping and identifying area is locked. By quantifying the moving direction and amplitude of the occlusion shadow, its dynamic changes are reflected in real time, providing predictive information for the intelligent material platform robot, capturing dynamic occlusion trends, and improving the system's adaptability to dynamic environments. An embodiment of the present invention determines the occlusion trend sub-region according to the grayscale characterization parameters of the feature grasping and identifying area through a material recognition module, and constructs an occlusion trend sub-vector based on the obtained several occlusion trend sub-regions. Furthermore, the occlusion trend sub-region is screened, and the occlusion trend sub-vector is constructed, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0046] Specifically, it can be understood that the occlusion tendency sub-area is the shadow area projected by the occlusion source onto the surface of the material to be grasped. The shadow area is the light-deficient area formed on the surface of the material to be grasped after the occlusion blocks the light of the light source. When the surface of the material is blocked by an occlusion object, such as other materials or robotic arm parts, the light of the light source is blocked, and this area can only receive weak scattered light in the environment, and the intensity of reflected light is significantly reduced. In the material table environment, the shadow area projected by the occlusion source onto the surface of the material to be grasped may be caused by slight shaking of other materials or external force, or position change of robotic arm parts. The center points of the occlusion trend sub-areas at adjacent acquisition moments respectively represent the shadow position of the occlusion source projected onto the surface of the material to be grasped at the previous moment and the shadow position of the occlusion source projected onto the surface of the material to be grasped at the current moment. The vector constructed with the two points as the starting point and the end point is the occlusion trend sub-vector, which can intuitively quantify the movement direction and movement amplitude of the occlusion source relative to the material to be grasped, thereby capturing the dynamic trend of the occlusion source relative to the material to be grasped. Furthermore, the occlusion trend sub-areas are screened, the occlusion trend sub-vector is constructed, and the efficiency and reliability of the intelligent material platform robot system are improved.

[0047] See also Figure 4 As shown, it is a logic flow chart of the grasping control module of an embodiment of the present invention determining the grasping compensation method for the material to be sorted in the characteristic grasping identification area. The grasping control module is used to determine the grasping compensation method for the material to be sorted in the characteristic grasping identification area, wherein: If the conveying direction vector and the occlusion tendency vector of the material to be sorted in the characteristic grasping and identifying area meet the first grasping compensation condition, the grasping control module determines that the grasping compensation method is to adjust the grasping time for grasping the material to be sorted in the characteristic grasping and identifying area; If the conveying direction vector and the occlusion tendency vector of the material to be sorted in the feature grasping and identification area do not meet the first grasping compensation condition, the grasping control module determines the grasping compensation method by determining whether there is a grasping abnormality risk in the feature grasping and identification area, and adjusting the grasping compensation parameters.

[0048] Specifically, the embodiment of the present invention determines the grabbing compensation method for the material to be sorted according to the comparison between the transmission direction vector and the occlusion trend vector through the grasping control module. It can be understood that the duration and risk level of the occlusion interference can be predicted through the relative movement trend of the occlusion object and the material, so as to select a targeted response strategy. The transmission direction vector represents the movement direction of the material to be sorted, such as the moving direction of the conveyor belt and the pushing direction of the material table, reflecting the spatial position change law of the material itself. The occlusion trend vector is determined by the occlusion trend sub-vectors at several monitoring moments, representing the occlusion objects such as other stacked materials and moving robotic arms. The overall movement trend of reflects the spatial position change law of the occlusion source. The angle between the two vectors can characterize the relative movement relationship between the occlusion and the material. The larger the angle, the more significant the difference in the movement directions of the two. The smaller the angle, the closer the movement directions of the two. Different grasping compensation methods are adaptively adjusted for different comparison situations to improve the accuracy of grasping timing, reduce the risk of collision under continuous occlusion, balance efficiency and safety, and enhance the adaptability of the system to dynamic scenes. Furthermore, the grasping compensation method is adaptively adjusted based on the changing trend of dynamic occlusion to improve the efficiency and reliability of the intelligent material platform robot system.

[0049] Specifically, the first grasping compensation condition is that the relative offset tendency parameter exceeds a preset relative offset tendency parameter threshold, the relative offset tendency parameter is the vector angle between the transmission direction vector and the occlusion tendency vector, and the occlusion tendency vector is the vector obtained by adding the occlusion tendency sub-vectors at several monitoring moments.

[0050] Specifically, the preset relative offset tendency parameter threshold is the product of the relative offset tendency parameter reference value and the relative offset coefficient. The relative offset tendency parameter reference value is the average value of the relative offset tendency parameter under the same working conditions in the historical data. The relative offset coefficient can be set by technical personnel in this field according to the accuracy requirements of the intelligent material platform robot system. The higher the accuracy requirement, the larger the relative offset coefficient is set. The value range of the relative offset coefficient can be [1.15, 1.3]. Preferably, it can be 1.2.

[0051] Specifically, the capture control module is used to adjust the capture time, wherein: The delay duration of the grabbing moment is negatively correlated with the relative deviation tendency parameter.

[0052] Specifically, the delay duration is the relative offset tendency parameter reference value / relative offset tendency parameter × delay factor. The relative offset tendency parameter reference value is the average value of the relative offset tendency parameter under the same working conditions in the historical data. The delay factor can be set by a technician in this field according to the accuracy requirements of the intelligent material platform robot system. The value range of the delay factor can be [0.3, 0.7] to avoid the delay duration being set too large or too small. Preferably, the delay factor can be 0.4.

[0053] Specifically, the embodiment of the present invention adjusts the grabbing time of grabbing the material to be sorted in the feature grabbing identification area through the grabbing control module when the feature grabbing identification area meets the first grabbing compensation condition. It can be understood that in the actual material platform sorting, the movement speed and direction of the obstruction may change dynamically. By adjusting the delay in real time through the vector angle, the system can adapt to the occlusion rhythm under different working conditions. The feature grabbing identification area meets the first grabbing compensation condition, that is, the relative movement trend of the obstruction and the material shows an inconsistent trend. The larger the relative offset tendency parameter, the faster the occlusion is released, and the shorter the delay time can be Reduce the residence time of materials on the conveyor belt, avoid subsequent material accumulation due to excessive waiting, increase the sorting volume per unit time, avoid ineffective waiting, and improve grasping efficiency. The smaller the relative offset tendency parameter, the slower the speed of occlusion removal. Extending the delay time can avoid the robot arm accidentally touching the obstruction due to premature grasping, reduce the risk of material damage and equipment failure, ensure the removal of obstructions, improve grasping accuracy, reduce energy consumption and mechanical loss, and dynamically adapt to the occlusion rhythm. In addition, it realizes the adaptive adjustment of the grasping compensation method based on the changing trend of dynamic occlusion, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0054] Specifically, it can be understood that the vector angle between the transmission direction vector and the occlusion tendency vector can characterize the separation efficiency of the relative motion between the material and the occlusion. The larger the relative offset tendency parameter, that is, the larger the angle, the more significant the difference in the motion direction of the occlusion and the material to be grasped, the faster the relative separation speed, and the shorter the duration of the occlusion state such as shadow coverage. The smaller the relative offset tendency parameter, that is, the smaller the angle, the slower the separation speed and the longer the occlusion duration. Based on the relative motion speed of the occlusion and the material, the time window for occlusion release is dynamically matched to achieve accurate quantification of the delay time. Furthermore, the grasping compensation method is adaptively adjusted based on the changing trend of dynamic occlusion, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0055] Specifically, the capture control module is used to determine whether there is a capture abnormality risk in the feature capture identification area, wherein: The grasping control module determines that there is a grasping abnormality risk in the feature grasping identification area based on the result of determining that the occlusion trend vector of the feature grasping identification area meets the grasping abnormality condition, and issues an abnormality warning signal; Based on the result of determining that the occlusion trend vector of the characteristic grasping and identifying area does not meet the grasping abnormality condition, it is determined that there is no grasping abnormality risk in the characteristic grasping and identifying area, and the material to be sorted in the characteristic grasping and identifying area is grasped based on the grasping compensation parameter; The abnormal capture condition is that the risk tendency parameter exceeds a preset risk tendency parameter threshold, and the risk tendency parameter is the absolute value of the difference between the module lengths of the occlusion trend vectors at adjacent acquisition moments.

[0056] Specifically, the preset risk tendency parameter threshold is the product of the risk tendency parameter reference value and the risk tendency factor. The risk tendency parameter reference value is the average value of the risk tendency parameter under the same working conditions in the historical data. The risk tendency factor can be set by technical personnel in this field according to the accuracy requirements of the intelligent material platform robot system. The higher the accuracy requirement, the smaller the risk tendency factor is set. The value range of the risk tendency factor can be [1.2, 1.35]. Preferably, it can be 1.25.

[0057] Specifically, the grasping control module is used to determine grasping compensation parameters, which include grasping direction and grasping speed, wherein: The grabbing direction is the vector direction of the material tendency vector, and the material tendency vector is the vector obtained by adding the blocking tendency vector and the conveying direction vector; The reduction amount of the grasping speed is positively correlated with the vector size of the occlusion tendency vector.

[0058] Specifically, the reduction in grasping speed is the current grasping speed × the vector size of the occlusion tendency vector × the speed control factor. The speed control factor can be set by technical personnel in this field according to the accuracy requirements of the intelligent material table robot system. The value range of the speed control factor can be [0.2, 0.4] to avoid the speed reduction being too large or too small. Preferably, the speed control factor can be 0.3.

[0059] Specifically, the embodiment of the present invention determines whether there is a grasping abnormality risk in the feature grasping identification area when the feature grasping identification area does not meet the first grasping compensation condition through the grasping control module, and adjusts the grasping compensation parameters. It can be understood that the feature grasping identification area does not meet the first grasping compensation condition, that is, the relative movement trend of the occlusion and the material presents a relatively consistent trend, and the shadow cast by the occlusion on the material will not change due to the material transmission. By monitoring the rate of change of the occlusion trend vector modulus, that is, the risk tendency parameter, the risk of change in the occlusion state can be quickly identified, such as sudden large-scale occlusion, material position offset caused by rapid movement of the occlusion, and in the case of abnormal risks, it can be paused or warned in advance to avoid grasping when the occlusion is unstable. The grasping direction is determined based on the material tendency vector, which is equivalent to incorporating the indirect impact of the occlusion on the material into the trajectory calculation. The grasping action tracks the actual position of the material more accurately, and the grasping speed adjustment amount is determined according to the module length of the occlusion trend vector, achieving a dynamic balance in which precision is prioritized when the occlusion impact is strong, and efficiency is prioritized when the impact is weak. The larger the vector size of the occlusion trend vector, the more a large speed adjustment is required, such as deceleration to improve positioning accuracy, or acceleration to leave the occlusion area, to ensure that the grasping action can still be stably executed under interference. The smaller the vector size of the occlusion trend vector, the smaller the speed adjustment amplitude, avoiding the decrease in sorting efficiency caused by excessive deceleration. Through the combined strategy of risk judgment and dynamic parameter adjustment, the intelligent material platform robot system can maintain efficient grasping under dynamic occlusion conditions, taking into account both grasping accuracy and efficiency while effectively avoiding potential abnormal risks. Furthermore, it realizes the adaptive adjustment of the grasping compensation method based on the changing trend of dynamic occlusion, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0060] Specifically, it can be understood that the risk tendency parameter represents the rate of change of the occlusion tendency, and the modulus of the occlusion tendency vector represents the intensity of the influence of the occlusion on the material. The larger the modulus, the more significant the change in the occlusion shadow concentration. The absolute value of the difference in modulus can represent the rate of change of this influence intensity. The larger the risk tendency parameter, the more drastic the occlusion state changes in a short period of time, which may cause the surface features of the material, such as the grasping point, to change rapidly. At this time, the grasping is prone to abnormal risks such as positioning deviation and collision with obstructions. By adjusting the grasping direction and speed, the implicit interference of stable occlusion on the material movement trajectory can be offset. The actual movement trend of the material is the indirect influence of the transmission power (transmission direction vector) and the occlusion. For example, the obstruction may produce a slight thrust or friction on the material. It is reflected in the superposition of occlusion tendency vectors. The material tendency vector can represent the real movement direction of the material under occlusion interference. The larger the modulus of the occlusion tendency vector, the stronger the impact of the occlusion on the material. For example, the occlusion range is large, the visual positioning error caused by the shadow is large, or the interaction between the occlusion and the material is more significant. At this time, the speed adjustment amount increases with the increase of the modulus. The interference is offset by dynamically adapting the speed. The larger the vector size of the occlusion tendency vector, the more it is necessary to slow down the speed to improve the grasping positioning accuracy. The smaller the vector size of the occlusion tendency vector, the smaller the speed reduction, avoiding efficiency loss caused by excessive adjustment. Furthermore, the grasping compensation method is adaptively adjusted based on the changing trend of dynamic occlusion, thereby improving the efficiency and reliability of the intelligent material platform robot system.

[0061] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

[0062] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that the present invention is susceptible to various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. An intelligent material platform robot system based on AI vision technology, characterized in that: include: A material platform control module, comprising a grabbing unit for grabbing the materials to be sorted on the material platform and a conveying unit for conveying the materials to be sorted; A feature acquisition module, connected to the material platform control module, for acquiring the grayscale characterization parameters and conveying direction vector of the material to be sorted; A material analysis module, connected to the feature acquisition module, is used to divide the material platform into several grabbing and identification areas, and determine the grayscale fluctuation of the material to be sorted based on the grayscale characterization parameters within the grabbing and identification areas under different lighting angles to mark the feature grabbing and identification areas; a material recognition module, connected to the feature acquisition module and the material analysis module, respectively, determining an occlusion trend sub-region based on the grayscale characterization parameters of the feature capture and recognition region under a preset illumination angle, and constructing an occlusion trend sub-vector based on the obtained occlusion trend sub-regions; a grabbing control module, which is respectively connected to the material platform control module, the feature acquisition module, and the material identification module, and is used to determine a grabbing compensation method for the material to be sorted by adjusting the grabbing time for grabbing the material to be sorted in the feature grabbing identification area according to the comparison between the conveying direction vector and the occlusion trend vector; or determining whether there is a grasping abnormality risk in the feature grasping and identifying area according to the occlusion trend vector, and adjusting a grasping compensation parameter of the feature grasping and identifying area; The occlusion tendency vector is determined according to a plurality of occlusion tendency sub-vectors.

2. The intelligent material platform robot system based on AI vision technology according to claim 1 is characterized in that: The material analysis module is used to determine the grayscale fluctuation characterization amount and the occlusion tendency coefficient based on the grayscale characterization parameters in the grasping and identifying area, wherein: The grayscale fluctuation characterization value is the difference between the maximum value and the minimum value of the grayscale characterization parameter in the grasping and recognition area under the same illumination angle; The occlusion tendency coefficient is the variance of the grayscale characterization parameter under different illumination angles.

3. The intelligent material platform robot system based on AI vision technology according to claim 2 is characterized in that: The material analysis module is used to mark the grasping and identifying area as a characteristic grasping and identifying area based on the determination result that the grayscale fluctuation characterization quantity and the occlusion tendency coefficient in the grasping and identifying area under different illumination angles meet the characteristic grasping and identifying area conditions, wherein: The condition for feature capture and identification area is that the grayscale fluctuation characterization value exceeds a preset grayscale fluctuation characterization value threshold, and the occlusion tendency coefficient does not exceed a preset occlusion tendency coefficient threshold.

4. The intelligent material platform robot system based on AI vision technology according to claim 3 is characterized in that: The material recognition module is used to determine the characteristic sub-region as the occlusion tendency sub-region based on the determination result that the grayscale characterization parameter of the characteristic sub-region of the characteristic capture and recognition region meets the occlusion tendency sub-region condition, wherein, The material recognition module divides the feature capture and recognition area into several sub-areas and obtains grayscale characterization parameters at several positions of the material to be sorted in the sub-areas; The occlusion tendency sub-region condition is that the grayscale characterization parameter of the characteristic sub-region does not exceed a preset grayscale characterization parameter threshold, and the characteristic sub-region is the sub-region where the grayscale characterization parameter has the minimum value in the feature capture and recognition region.

5. The intelligent material platform robot system based on AI vision technology according to claim 4 is characterized in that: The material identification module is used to construct an occlusion trend sub-vector of the latter of the adjacent acquisition moments based on the occlusion trend sub-regions at adjacent acquisition moments within a preset monitoring period, wherein: The material recognition module obtains the occlusion trend sub-area at several acquisition moments within the feature capture and recognition area; The occlusion trend sub-vector is constructed with the area center point of the occlusion trend sub-region at the previous acquisition moment in adjacent acquisition moments as the vector starting point of the occlusion trend sub-vector, and with the area center point of the occlusion trend sub-region at the next acquisition moment in adjacent acquisition moments as the vector end point of the occlusion trend sub-vector.

6. The intelligent platform robot system based on AI vision technology according to claim 5 is characterized in that: The grabbing control module is used to determine the grabbing compensation method for the materials to be sorted in the characteristic grabbing identification area, wherein: If the conveying direction vector and the occlusion tendency vector of the material to be sorted in the characteristic grasping and identifying area meet the first grasping compensation condition, the grasping control module determines that the grasping compensation method is to adjust the grasping time for grasping the material to be sorted in the characteristic grasping and identifying area; If the conveying direction vector and the occlusion tendency vector of the material to be sorted in the feature grasping and identification area do not meet the first grasping compensation condition, the grasping control module determines the grasping compensation method by determining whether there is a grasping abnormality risk in the feature grasping and identification area, and adjusting the grasping compensation parameters.

7. The intelligent platform robot system based on AI vision technology according to claim 6 is characterized in that: The first grasping compensation condition is that the relative offset tendency parameter exceeds a preset relative offset tendency parameter threshold, the relative offset tendency parameter is the vector angle between the transmission direction vector and the occlusion tendency vector, and the occlusion tendency vector is the vector obtained by adding the occlusion tendency sub-vectors at several monitoring moments.

8. The intelligent material platform robot system based on AI vision technology according to claim 7 is characterized in that: The grabbing control module is used to adjust the grabbing time, wherein: The delay duration of the grabbing moment is negatively correlated with the relative deviation tendency parameter.

9. The intelligent platform robot system based on AI vision technology according to claim 7 is characterized in that: The grab control module is used to determine whether there is a grab abnormality risk in the feature grab identification area, wherein: The grasping control module determines that there is a grasping abnormality risk in the feature grasping identification area based on a determination result that the occlusion trend vector of the feature grasping identification area meets the grasping abnormality condition; Based on the result of determining that the occlusion trend vector of the characteristic grasping and identifying area does not meet the grasping abnormality condition, it is determined that there is no grasping abnormality risk in the characteristic grasping and identifying area, and the material to be sorted in the characteristic grasping and identifying area is grasped based on the grasping compensation parameter; The abnormal capture condition is that the risk tendency parameter exceeds a preset risk tendency parameter threshold, and the risk tendency parameter is the absolute value of the difference between the module lengths of the occlusion trend vectors at adjacent acquisition moments.

10. The intelligent platform robot system based on AI vision technology according to claim 9 is characterized in that: The grasping control module is used to determine grasping compensation parameters, which include grasping direction and grasping speed. The grabbing direction is the vector direction of the material tendency vector, and the material tendency vector is the vector obtained by adding the blocking tendency vector and the conveying direction vector; The reduction amount of the grasping speed is positively correlated with the vector size of the occlusion tendency vector.

Images