State control method and system for AGV robot in traditional Chinese medicine warehouse and storage medium

Abstract

The invention relates to the technical field of storage and transportation, and discloses a state control method and system for a traditional Chinese medicine warehouse AGV robot and a storage medium, and the method achieves the precise regulation and control of the process of transporting traditional Chinese medicine dry medicinal materials through an AGV robot through a multi-dimensional technical means. Cargo images are collected at two positions within a preset radian range, a calculation edge is extracted and matched with a temporary edge, a reference edge is obtained through extension, transverse upheaval data are calculated, early warning is conducted when the transverse upheaval data exceed the limit, and the transportation speed is adjusted in an inverse correlation mode according to the comprehensive upheaval data; the vertical edge of the tray is used for assisting verification, so that the accuracy of the reference edge is improved; calculating a raised discrete data optimization speed regulation strategy; cargo abnormity is identified through calculating density by volume and weight; the acceleration is dynamically adjusted by combining gravity center shift and the movement direction; position sensor data verification and correction are introduced, and the detection precision is ensured. According to the scheme, the safety, reliability and efficiency of AGV transportation of the traditional Chinese medicine warehouse are remarkably improved.

Description

Technical Field

[0001] This application relates to the technical field of warehousing and transportation, and in particular to a method, system and storage medium for controlling the state of an AGV robot in a traditional Chinese medicine warehouse. Background Art

[0002] As an important raw material in the traditional Chinese medicine industry, the warehousing and transportation of dried traditional Chinese medicine materials is crucial for quality control. In actual operation, for the convenience of storage and transportation, dried traditional Chinese medicine materials are usually packed into packaging units of specific specifications. Common packing methods include using woven bags, gunny bags or special sealed bags to encapsulate different types and batches of traditional Chinese medicine materials according to certain weight or volume standards. These packed traditional Chinese medicine material packages are transported through equipment such as conveyor lines and forklifts.

[0003] With the development of intelligent manufacturing technology, AGV robots have gradually emerged in the field of warehouse logistics, and traditional Chinese medicine warehouses have also begun to introduce AGV robots to undertake transportation work. An AGV (Automated Guided Vehicle) robot is a battery-powered driverless mobile robot that can automatically travel along a preset path through electromagnetic, optical or visual guidance methods and can complete tasks such as material handling and transportation. With its advantages of high automation, flexible path planning and strong operation stability, an AGV robot can accurately carry out goods handling in the warehouse according to preset programs or dispatching system instructions. In a traditional Chinese medicine warehouse, an AGV robot can replace manual labor to complete the transportation tasks of traditional Chinese medicine material packages from the storage area to the sorting area and the shipping area, greatly improving the transportation efficiency.

[0004] However, due to the relatively fragile texture of dried traditional Chinese medicine materials themselves, in order to avoid the fragmentation of the materials, their packaging usually cannot be overly compacted, resulting in certain voids inside the traditional Chinese medicine material packages and a relatively loose overall shape. When these loosely packed traditional Chinese medicine material packages are stacked and transported on a pallet, after multiple loading, unloading, handling and jolting during transportation, the traditional Chinese medicine material packages stacked on the side are extremely prone to bulging and protruding parts. During the transportation process of the AGV robot, these bulging and protruding parts are very likely to rub and collide with the logistics racks, shelf corners or other obstacles in the warehouse, which may cause minor damage to the packaging of traditional Chinese medicine materials and the scattering of traditional Chinese medicine materials, or serious problems such as the operation failure of the AGV robot or even transportation accidents, not only causing losses to traditional Chinese medicine materials, but also affecting the normal logistics order of the warehouse. Therefore, it is urgent to propose an effective state control scheme for the AGV robot in a traditional Chinese medicine warehouse to ensure the safety of traditional Chinese medicine material transportation. Summary of the Invention

[0005] In order to improve the safety of the AGV robot during the transportation of dried traditional Chinese medicine materials, this application provides a method, system and storage medium for controlling the state of an AGV robot in a traditional Chinese medicine warehouse.

[0006] In a first aspect, the present application provides a method for controlling the state of an AGV robot in a traditional Chinese medicine warehouse, adopting the following technical solutions: A method for controlling the state of an AGV robot in a traditional Chinese medicine warehouse includes the following steps: Based on the transportation goods instruction, call the AGV robot; Obtain first image data at a first position and second image data at a second position of the AGV robot beside the transported goods, wherein, with the center of the transported goods as the center of the circle, the radian between the first position and the second position is within a preset first radian range; Extract multiple calculated edges of the transported goods from the first image data and the second image data; Match corresponding temporary edges based on the lowest points of the calculated edges, associate the temporary edges with the calculated edges one by one based on the adjacent position relationship, and extend the temporary edges to the highest points of the corresponding calculated edges to obtain reference edges; Calculate the lateral bulge data of the calculated edges relative to the corresponding reference edges, and the lateral bulge data corresponds to the protrusion degree of the calculated edges on the side away from the center of the circle of the reference edges; If the lateral bulge data is greater than a preset bulge reference data, issue a transportation collision warning; According to multiple pieces of lateral bulge data corresponding to the calculated edges one by one, calculate a comprehensive bulge data, and inversely adjust the transportation speed of the AGV robot according to the comprehensive bulge data.

[0007] By adopting the above technical solutions, traditional Chinese medicine packages are relatively loose and fragmented and cannot be pressed too tightly, so they are prone to looseness and protrusion; during the transportation process of the robot, it is necessary to detect the goods. By obtaining image data of the AGV robot at different positions, extracting calculated edges, matching temporary edges, and extending to obtain reference edges, and then calculating the lateral bulge data, the bulging and protruding conditions of the transported goods (traditional Chinese medicine dry medicinal material packages) can be accurately detected. When the lateral bulge data is greater than the preset reference data, a warning is issued, which can detect in advance the risk of rubbing and collision with obstacles due to the bulging and protruding of the goods during transportation, and timely take measures to avoid problems such as damage and scattering of the medicinal material packages and operation failures of the AGV robot, effectively ensuring the safety of medicinal material transportation and reducing medicinal material losses. Calculate the comprehensive bulge data according to the lateral bulge data corresponding to multiple calculated edges, and inversely adjust the transportation speed of the AGV robot according to the comprehensive bulge data. In this way, the AGV robot can dynamically adjust the speed according to the actual bulging situation of the goods, reduce the speed when the goods bulge severely, and reduce the collision risk caused by too high a speed; when the goods bulge slightly, maintain a relatively high speed to ensure the transportation efficiency.

[0008] Optionally, in the step of obtaining the reference edge, the following sub-steps are further included: Extract the vertical edges of the transport pallet from the first image data and the second image data; Verify the one-to-one correspondence between the vertical edge and the calculated edge based on the relationship of adjacent positions; If the verification is passed, extend the vertical edge to the highest point of the corresponding calculated edge; Calculate the similarity data between the vertical edge and the reference edge; If the similarity data is greater than the preset similarity reference value, use the vertical edge as the new reference edge; otherwise, prompt that the temporary edge matching is incorrect.

[0009] By adopting the above technical solution, with the assistance of the vertical edge of the transport pallet to determine the reference edge, through verifying the correspondence relationship and calculating the similarity data, the temporary edge matching errors caused by factors such as irregular cargo shapes and image acquisition errors can be effectively excluded, improving the accuracy of determining the reference edge; if the similarity data meets the standard, the vertical edge is used to replace the reference edge, making the reference edge more conform to the true state of the cargo and the pallet in the actual transportation scenario, further improving the accuracy of calculating the lateral bulge data of the cargo, thereby enhancing the reliability of transport collision warning and transport speed adjustment, and ensuring the safety and stability during the transportation of traditional Chinese medicine dry herbs by the AGV robot.

[0010] Optionally, the method further includes the following steps: If the lateral bulge data is less than the preset bulge reference data, calculate the bulge discrete data between multiple lateral bulge data; If the bulge discrete data is greater than the preset reference discrete data, calculate the absolute value of the sum of multiple lateral bulge data; Calculate the comprehensive absolute value according to multiple absolute values, and inversely adjust the transport speed of the AGV robot according to the comprehensive absolute value; Calculate the comprehensive discrete data according to multiple bulge discrete data, and positively adjust the adjustment step size of the transport speed of the AGV robot according to the comprehensive discrete data.

[0011] By adopting the above technical solution, by calculating the discrete data of the bulge, the uneven degree of the bulge distribution on the surface of the goods can be identified. When the discrete data exceeds the limit, the comprehensive absolute value is calculated by combining the lateral bulge data and the absolute value of the value, and the transportation speed is adjusted in an inverse correlation manner to avoid the collision risk caused by the local slight bulge of the goods but uneven distribution. At the same time, the speed adjustment step size is adjusted in a positive correlation manner according to the comprehensive discrete data. When the discrete degree of the goods bulge is high, the transportation speed adjustment is more refined. When the discrete degree is low, the adjustment process is accelerated, realizing the dynamic and precise control of the transportation speed. On the premise of ensuring transportation safety, the transportation efficiency and the safety of the goods are balanced, effectively adapting to the transportation characteristics of loose packaging and irregular shapes of traditional Chinese medicine dried medicinal materials, and improving the overall stability and reliability of the AGV robot in transporting traditional Chinese medicine dried medicinal materials.

[0012] Optionally, the method further includes the following steps: Calculate the position data of the calculated edge on the transport tray; Calculate the volume value of the transported goods according to the position data and size data of multiple calculated edges; Collect the weight value of the transported goods; Calculate the real-time density value according to the volume value and the weight value; Obtain the standard density value corresponding to the transported goods; Calculate the difference between the real-time density value and the standard density value as the density difference; If the absolute value of the density difference is greater than the preset reference difference, a warning prompt for the type of goods is given.

[0013] By adopting the above technical solution, the actual density state of the transported goods can be accurately identified. By comparing the real-time density with the standard density, it is possible to timely discover whether abnormal situations such as packaging damage and leakage of the goods have occurred, and avoid a series of subsequent transportation problems caused by the leakage of the goods. Through the density difference warning, the accuracy of the logistics link in the traditional Chinese medicine warehouse can be effectively guaranteed, the transportation accidents caused by the damage and leakage of the goods can be reduced, and the standardization and safety of the transportation and management of traditional Chinese medicine dried medicinal materials in the warehouse can be ensured.

[0014] Optionally, in the step of calculating the position data of the calculated edge on the transport tray, the following sub-steps are further included: Obtain the size data of the transport tray; Identify the height value of the smallest packaging unit in the transported goods; Calculate the shooting height value of the smallest packaging unit in contact with the transport tray; Calculate the position data of the calculated edge on the transport tray according to the shooting height value and the height value.

[0015] By adopting the above technical solution, using multi-dimensional data such as the size of the transport pallet and the height of the minimum packaging unit of the goods, fully considering the actual shape and spatial relationship of the goods stacked on the pallet, and calculating the shooting height value, the relative position information of the goods and the pallet is accurately quantified, effectively avoiding position data deviation caused by not considering the details of goods stacking and the influence of pallet size, thereby significantly improving the accuracy of the volume calculation of the goods based on the calculated edges.

[0016] Optionally, the method further includes the following steps: Based on the elimination instruction for eliminating the warning of the goods type, obtain the weight data at multiple positions on the transport pallet; Calculate the center of gravity position according to the multiple weight data; Obtain the center position of the transport pallet; Calculate the position distance between the center of gravity position and the center position; According to the position distance, inversely adjust the loading and unloading speed of the AGV robot for loading and unloading the transport pallet, and inversely adjust the turning speed of the AGV robot when turning.

[0017] By adopting the above technical solution, after manually eliminating the warning of the goods type, calculate the center of gravity position by obtaining the weight data at multiple positions on the transport pallet, accurately locate the deviation of the center of gravity of the goods on the transport pallet, and dynamically adjust the movement speed of the robot based on the position distance. Reducing the speed during loading and unloading can avoid the risks of dropping and scattering caused by the deviation of the center of gravity of the goods; slowing down the speed during turning can effectively control the inertial effect and prevent heavy goods from flying off due to inertia, causing transportation accidents. It not only ensures the integrity and safety of the goods during transportation, but also reduces the mechanical wear of the robot caused by the deviation of the goods, and improves the operation stability and reliability of the AGV robot during the transportation of traditional Chinese medicinal materials.

[0018] Optionally, the method further includes the following steps: Obtain the motion direction vector of the AGV robot; Calculate the center of gravity vector of the goods according to the center position and the center of gravity position; Calculate the vector angle between the motion direction vector and the center of gravity vector of the goods; According to the vector angle, inversely adjust the acceleration for adjusting the speed increase of the AGV robot; According to the vector angle, positively adjust the deceleration for adjusting the speed reduction of the AGV robot.

[0019] By adopting the above technical solution, the spatial relationship between the moving direction of the robot and the offset of the center of gravity of the goods is accurately analyzed, and the motion acceleration is flexibly adjusted according to the vector angle. When the included angle is large and the center of gravity of the goods is severely offset, reducing the acceleration during speed increase can avoid the goods from tipping over due to too fast acceleration; increasing the deceleration during speed reduction can enable the robot to brake in time and prevent the goods from being thrown off under the action of inertia. The dynamic stability during the transportation of traditional Chinese medicine dry medicinal materials by the AGV robot is improved, the probability of transportation accidents is reduced, and the safety of goods transportation is guaranteed.

[0020] Optionally, the method further includes the following data: Calculating the height data of the edge bulge of the transported goods based on the position sensor arranged at the upper edge of the transport tray; Calculating the real-time bulge data according to the height data; Calculating the edge difference according to the lateral bulge data and the real-time bulge data; If the edge difference is greater than a preset reference edge value, a data error warning is given; Correcting the lateral bulge data according to the edge difference.

[0021] By adopting the above technical solution, using the high-precision data of the position sensor, a double verification mechanism is constructed to effectively identify the possible errors in calculating the lateral bulge data by image analysis; through the edge difference warning, the data deviation caused by system anomalies or environmental interference can be detected in time, ensuring the reliability of the state control decision-making; using the gain value to correct the lateral bulge data improves the accuracy of the detection of the goods bulge state, provides a more accurate basis for transportation collision warning and speed adjustment, further enhances the stability and safety of the state control of the AGV robot in the traditional Chinese medicine warehouse, and reduces the misjudgment and transportation accidents caused by data errors.

[0022] In a second aspect, the present application provides a state control system for an AGV robot in a traditional Chinese medicine warehouse, adopting the following technical solution: A state control system for an AGV robot in a traditional Chinese medicine warehouse, including a processor, and the processor executes the steps of the state control method for the AGV robot in a traditional Chinese medicine warehouse as described in any one of the above.

[0023] In a third aspect, the present application provides a storage medium, adopting the following technical solution: A storage medium, in which a program is stored, and when the program is executed by a processor, the steps of the state control method for the AGV robot in a traditional Chinese medicine warehouse as described in any one of the above are realized.

[0024] In summary, the present application includes at least one of the following beneficial technical effects: By combining multi-position image acquisition with pallet edge verification, the protruding state of the goods is quantitatively detected. When the limit is exceeded, an early warning is issued and the transportation speed is adjusted according to the degree of bulge; The edge position is calibrated using multi-dimensional data of the pallet and the goods, and the detection results are cross-checked to ensure reliable data; Abnormal goods are identified through density calculation, triggering a type warning; Based on the relationship between the center of gravity offset and the movement direction, the loading, unloading, steering, acceleration, and deceleration speeds of the robot are dynamically adjusted; For slight bulge situations, the speed is finely adjusted according to the discrete bulge data, taking into account both safety and efficiency, and comprehensively improving the safety, stability, and operation efficiency of the transportation of traditional Chinese medicine dry herbs. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 is a step diagram of the state control method of the AGV robot in the traditional Chinese medicine warehouse.

[0026] Figure 2 is a schematic diagram of the scene in the traditional Chinese medicine warehouse.

[0027] Figure 3 is a schematic diagram of the AGV robot at the first position and the second position beside the transported goods.

[0028] Figure 4 is a schematic diagram of the reference edge, calculated edge, and temporary edge in the image data.

[0029] Figure 5 is a schematic diagram of the extended vertical edge in the image data.

[0030] Figure 6 is a step diagram of obtaining the reference edge.

[0031] Figure 7 is a step diagram of adjusting the motion parameters of the AGV robot according to the lateral bulge data.

[0032] Reference numerals: 1, goods rack; 2, metal support frame; 3, transport pallet; 4, AGV robot; 5, traditional Chinese medicine package. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] The following details the embodiments of the present application, and examples of the embodiments are shown in the drawings.

[0034] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "schematic embodiments", "examples", "specific examples" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0035] Referring to Figure 2 , in the traditional Chinese medicine warehouse, the storage locations are separated by the vertical goods shelves 1, and the square metal support frames 2 laid on the ground (the net height of the bottom is adapted to the passage of AGVs) are used to carry the pallets, and the traditional Chinese medicine packages 5 stacked on the pallets form approximate cubic goods. The AGV robot 4 adopts a lifting transportation mode and realizes the overall handling of the goods by lifting the metal support frame 2. In this scenario, the loose stacking of the goods easily causes the sides to bulge, and the obstacles such as the corners of the shelves and the corners of the passage in the narrow storage passage pose a collision risk during AGV transportation.

[0036] An embodiment of the present application discloses a method for controlling the state of an AGV robot in a traditional Chinese medicine warehouse. Referring to Figure 1 and Figure 2 , it includes the following steps: After the system receives the transportation instruction, the scheduling program plans the shortest path based on the warehouse map and drives the AGV robot 4 towards the target storage location.

[0037] Obtain the first image data at the first position and the second image data at the second position of the AGV robot 4 beside the transported goods, where the radian between the first position and the second position is within a preset first radian range with the center of the transported goods as the center of the circle. The first radian range is between 90 degrees and 180 degrees, excluding 90 degrees. For the convenience of calculation, referring to Figure 3 , in this embodiment, 180 degrees is taken as an example. After arriving beside the goods, the robot performs 180° symmetric dual-view image acquisition: with the geometric center of the goods as the center of the circle, first collect the left view of the goods on one side of the metal support frame 2 (the first position), and then drive straight through the bottom of the support frame to the symmetric position (the second position) to collect the right view. The dual-view covers the complete side contour of the goods and eliminates the detection blind area of the single view, which is especially suitable for the comparison of the opposite side features of cubic goods.

[0038] Referring to Figure 4 and Figure 5, a plurality of computational edges of the transported goods are extracted from the first image data and the second image data; after grayscaling and noise reduction processing of the left and right views, the Canny edge detection algorithm is used to extract the computational edges of the outer contour of the goods (i.e., the irregular boundary segments actually formed by stacking).

[0039] Taking the lowest point of the computational edge as the anchor point, retrieve the temporary edges at adjacent positions from the pre-stored "standard cube edge database" (simulating the vertical edges in the ideal stacking state), and establish a one-to-one correspondence through coordinate mapping. Extend the temporary edge upward from the lowest point to the highest point of the computational edge to form a virtual vertical reference line (reference edge). For example, if the lowest point coordinates of a computational edge are (x1, y1) and the highest point is (x2, y2), then the corresponding reference edge is the line segment vertically extending from (x1, y1) to (x1, y2), representing the ideal boundary when the goods are not bulging.

[0040] Calculate the lateral bulge data of the computational edge relative to the corresponding reference edge. The lateral bulge data corresponds to the degree of bulge of the computational edge on the side away from the center of the circle of the reference edge; the simplest way to calculate the degree of bulge is, for example, to calculate the maximum bulge size. For each computational edge, measure the horizontal distance from each point to the reference edge and take the maximum value as the lateral bulge data (unit: cm). For example, if the midpoint of a computational edge deviates from the reference edge by 3 cm, then this data is 3 cm.

[0041] If the lateral bulge data is greater than the preset bulge reference data, a transportation collision warning is issued. Preset a bulge reference threshold (such as 5 cm). When the lateral bulge data of any computational edge exceeds the limit, the AGV issues a warning through the on-vehicle sound and light alarm and pushes an exception prompt to the warehouse management system to remind manual intervention to sort out the goods.

[0042] According to the plurality of lateral bulge data corresponding one-to-one to the computational edges, calculate the comprehensive bulge data, and inversely adjust the transportation speed of the AGV robot 4 according to the comprehensive bulge data. Take the average value or the mean square deviation of the lateral bulge data of all computational edges to obtain the comprehensive bulge data. Dynamically adjust the transportation speed through an inverse correlation function, such as v = v0×e -k×S , where v0 is the reference speed, S is the comprehensive bulge value, and k is the adjustment coefficient. For example, when the comprehensive bulge value is 0, the robot travels at the maximum speed of 2 m / s; if the bulge value reaches 10 cm, the speed drops to 0.5 m / s to reduce the risk of inertial impact.

[0043] Due to the loose and fragmented texture of traditional Chinese medicine packages and the inadvisability of excessive compaction, the phenomenon of package bulging is extremely likely to occur during transportation. In this solution, the AGV robot 4 is controlled to collect cargo images at different positions, extract and calculate the edge lines and match them with the pre-stored temporary edge lines. After extension processing, a reference edge line is generated, and then the lateral bulge data is quantitatively calculated, which can accurately identify the protrusion degree of traditional Chinese medicine dry material packages. When the detected lateral bulge data exceeds the preset threshold, the system immediately triggers a collision warning to prevent in advance the scraping and collision of the bulging cargo with storage shelves, channel obstacles, etc., effectively avoiding the damage, scattering of the medicine packages and the operation failure of the AGV robot 4, and reducing the medicine loss rate. At the same time, by comprehensively analyzing multiple groups of lateral bulge data, a comprehensive bulge index is generated, and the robot transportation speed is adjusted in an inverse correlation manner - when the cargo bulge is serious, the traveling speed is automatically reduced to reduce the collision risk caused by high-speed movement; when the bulge degree is small, a higher transportation speed is maintained to ensure that the warehousing logistics efficiency is not affected, realizing the dynamic balance between transportation safety and operation efficiency. The 180° symmetric image covers the entire side of the cargo. Combining with the geometric symmetry of the metal support frame 2, the symmetric bulge on both sides or the unilateral protrusion can be accurately identified, avoiding missed detection caused by limited viewing angles. Based on the method of matching the temporary edge line at the lowest point and extending it, it is suitable for generating the benchmark of goods with different heights (such as when the number of stacked pallets changes), without the need for manual presetting of the goods height parameters, improving the generalization ability of the algorithm. Only the edge line features need to be extracted instead of the full image pixel analysis, reducing the calculation amount and meeting the real-time requirements of the AGV embedded system (processing delay < 200ms). Through the hierarchical control strategy of "single-point over-limit warning + comprehensive value speed regulation", both false alarms of slight bulges are avoided, and rapid response to significant risks can be achieved, reducing the incidence of transportation accidents.

[0044] Refer to Figure 6 , in the step of obtaining the reference edge line, introducing the vertical edge line of the transportation pallet 3 as an auxiliary verification benchmark, further includes the following sub-steps: Extract the vertical edge line of the transportation pallet 3 from the first image data and the second image data; through the Hough line detection algorithm, identify the vertical edge line of the transportation pallet 3 (i.e., the vertical contour line of the pallet edge) from the left and right view images, and this edge line reflects the vertical benchmark of the pallet bearing surface. Taking the left view as an example, assume that the coordinates of the extracted vertical edge line of the pallet are Ltray(x0, y1)-(x0, y2), and the corresponding edge line in the right view is Rtray(xn, y1)-(xn, y2).

[0045] Verify the one-to-one correspondence between the vertical edges and the calculated edges based on the relationship of adjacent positions; establish the spatial mapping relationship between the vertical edges of the tray and the calculated edges of the goods: for a certain calculated edge LcEdge(xi, ya)-(xj, yb) in the left view, if the horizontal distance between its lowest point (xi, ya) and the tray edge Ltray is less than the preset threshold (such as 5 cm), it is determined that they are adjacent in position, and a preliminary correspondence is established. Similarly, process the right view data.

[0046] If the verification passes, extend the vertical edge to the highest point of the corresponding calculated edge; if the correspondence verification passes, extend the vertical edge of the tray upward to the highest point of the calculated edge of the goods. For example, in the left view, the upper endpoint (x0, y2) of the tray edge Ltray is extended to the highest point (xi, yb) of the calculated edge, generating a candidate reference edge Lcand(x0, y2)-(xi, yb).

[0047] Calculate the similarity data between the vertical edge and the reference edge; use the cosine similarity algorithm to calculate the direction consistency between the candidate reference edge and the original reference edge (generated based on the temporary edge). Sim = vector v1 × vector v2 / (the modulus of vector v1 × the modulus of vector v2); where vector v1 is the original reference edge vector (from the lowest point to the highest point), and vector v2 is the extended edge vector of the tray. The similarity value sim ranges from [-1, 1], and the larger the value, the closer the two edges are to being parallel (ideally 1).

[0048] If the similarity data is greater than the preset similarity reference value, use the vertical edge as the new reference edge; otherwise, prompt that there is an error in the temporary edge matching. Preset similarity reference threshold (such as 0.9): If sim > 0.9, it means that the trend of the tray edge and the goods edge is consistent. Replace the original reference edge with the extended edge of the tray to improve the benchmark fitting degree; If sim ≤ 0.9, it is determined that there is an error in the temporary edge matching (such as the goods being tilted, image distortion, etc.). Prompt "Temporary edge matching error" through the system interface, and mark that the goods need to be manually reviewed.

[0049] Through the objective benchmark of the physical edge of the tray, exclude the influence of irregular shapes of the goods (such as top depressions, side arc protrusions) and perspective distortion during image acquisition. For example, when the calculated edge is tilted due to loose stacking of the goods, the vertical edge of the tray can still provide a vertical benchmark, avoiding miscalculation of the bulge data caused by incorrect matching of the temporary edge. Experimental data shows that after introducing the verification of the tray edge, the vertical accuracy of the reference edge is improved, the standard deviation of the lateral bulge data is reduced, and the accuracy of the collision warning is improved. This mechanism is especially applicable to scenarios with different specifications of trays (such as wooden trays, metal trays) and changes in the stacking height of goods. For example, when the height of the tray increases, the extended edge of the tray can automatically adapt to the height of the goods without re-calibrating the algorithm parameters.

[0050] Referring to Figure 7 , when the lateral bulge data does not exceed the preset threshold, spatial distribution feature analysis is introduced. The method further includes the following steps: If the lateral bulge data is less than the preset bulge reference data, the bulge discrete data between multiple lateral bulge data is calculated; the bulge discrete data uses the standard deviation algorithm to calculate the degree of dispersion of multiple lateral bulge data. In the definition of lateral bulge data, it is set that the outward convexity of the edge is positive and the inward concavity is negative. For example, if an edge bulges outward by 3 cm, it is recorded as +3 cm, and if it concaves inward by 2 cm, it is recorded as -2 cm, so that the data directly reflects the "expansion" or "contraction" trend of the contour relative to the ideal state.

[0051] If the bulge discrete data is greater than the preset reference discrete data, it is determined that the bulge on the surface of the goods is unevenly distributed, and there may be a risk of local protrusion (such as the stacking of a certain layer of traditional Chinese medicine packages being skewed). The secondary control strategy needs to be activated, and then the absolute value of the sum of multiple lateral bulge data is calculated. The absolute values of all lateral bulge data are taken and summed to obtain the cumulative bulge amount.

[0052] Since there are multiple edges, the average value of the multiple absolute values corresponding to each edge is calculated to obtain the comprehensive absolute value. The transportation speed of the AGV robot 4 is adjusted in an inverse correlation according to the comprehensive absolute value; an inverse correlation function between the transportation speed v and the comprehensive absolute value is established, such as: v = vbase×(1 - k1×comprehensive absolute value); where vbase is the reference speed (such as 1.5 m / s), and k1 is the adjustment coefficient (such as 0.8). When the comprehensive absolute value = 0, the speed remains unchanged. When the comprehensive absolute value = 1, the speed drops to 30% of the reference value to avoid edge rubbing caused by local bulges.

[0053] Since there are multiple edges, the average value of the multiple bulge discrete data corresponding to each edge is calculated to obtain the comprehensive discrete data. The adjustment step size of the transportation speed of the AGV robot 4 is adjusted in a positive correlation according to the comprehensive discrete data. The adjustment step size Δv of the transportation speed is positively correlated with the comprehensive discrete data, that is: Δv = Δvmin + k2×comprehensive discrete data; where Δvmin is the minimum step size (such as 0.05 m / s), and k2 is the step size coefficient (such as 0.2). When the value of the comprehensive discrete data is relatively high (such as multiple scattered protrusions on the surface of the goods), the step size automatically shrinks, making the speed adjustment more precise (such as from 0.1 m / s / time to 0.02 m / s / time); when the value of the comprehensive discrete data is relatively low, the step size increases to quickly complete the adjustment and improve the efficiency.

[0054] Aiming at the transportation problem of loose packaging and irregular shapes of traditional Chinese medicine dry herbs, this method innovatively introduces uplift discrete data and comprehensive absolute value analysis to achieve intelligent dynamic regulation of transportation speed. By using the standard deviation algorithm to quantify the uplift discreteness, it can sensitively capture the uneven deformation on the surface of the goods and accurately identify the potential risks of local slight uplifts with uneven distribution. When the discrete data exceeds the standard, the system combines the horizontal uplift data and the absolute value of the sum to calculate the comprehensive absolute value, and adjusts the transportation speed in the reverse direction based on this - the larger the comprehensive absolute value, the lower the transportation speed, thus effectively avoiding collision risks. At the same time, the speed adjustment step size is positively adjusted according to the comprehensive discrete data: when the uplift discreteness of the goods is high, it means that the surface shape is complex and changeable, and the system automatically reduces the step size for refined speed regulation; when the discreteness is low, the adjustment process is accelerated to ensure transportation efficiency. This two - dimensional collaborative control strategy not only builds a solid defense line for transportation safety but also takes into account the operation efficiency, providing a stable and reliable technical guarantee for the AGV robot 4 to transport traditional Chinese medicine dry herbs in a complex warehouse environment and significantly improving the intelligent level of logistics transportation.

[0055] To further ensure the accuracy and safety of cargo transportation, a density detection and warning mechanism is added. The specific method also includes the following steps: Based on the position data of the calculated edges on the transportation pallet 3 and combined with the size information of the pallet itself, the three - dimensional modeling algorithm is used to accurately outline the three - dimensional contour of the goods. By integrating and analyzing the spatial coordinates and length data of multiple calculated edges, the volume value of the transported goods can be accurately calculated. At the same time, the AGV robot 4 uses the built - in high - precision weighing sensor to collect the weight value of the transported goods in real time. This sensor has an automatic calibration function to eliminate measurement errors caused by the weight of the pallet and uneven ground.

[0056] After obtaining the volume and weight data, the system quickly calculates the real - time density value of the goods. Subsequently, by docking with the warehouse management database, the standard density value corresponding to this batch of traditional Chinese medicine dry herbs is retrieved. The standard density value is a reference value preset based on parameters such as the type of herbs and the degree of dryness. For example, the standard density of dried astragalus is 0.35 g / cm³, and the standard density of dried wolfberries is 0.42 g / cm³. The real - time density value is compared with the standard density value to calculate the density difference between the two. If the absolute value of the density difference exceeds the preset reference difference (generally set at 10% of the standard density value), the system immediately triggers a warning prompt for the type of goods, and reminds the warehouse management personnel to conduct a verification through means such as sound and light alarms and pop - up notifications.

[0057] This mechanism can accurately identify the actual density state of the transported goods. When there is packaging damage and material leakage, the weight of the goods will decrease, resulting in the deviation of the real-time density value from the standard value; if there are cases of wrong loading or mixed loading of goods, it will also cause abnormal fluctuations in density. Through density difference early warning, it can effectively ensure the accuracy of the logistics link in the traditional Chinese medicine warehouse, reduce transportation accidents caused by damaged and leaked goods, wrong loading and mixed loading, and avoid economic losses and environmental pollution caused by the scattering of medicinal materials. Through actual tests, after introducing this mechanism, the discovery time of abnormal situations of warehouse goods is shortened, the incidence rate of transportation accidents is reduced, significantly improving the standardization and safety of the transportation and management of traditional Chinese medicine dry medicinal materials, providing strong support for the intelligent development of traditional Chinese medicine warehousing logistics.

[0058] In the step of calculating the position data of the edge on the transport pallet 3, the following sub-steps are further included: Obtain the size data of the transport pallet 3; the system pre-enters the length, width, and height parameters of various types of pallets in the warehouse (such as 1200mm×1000mm standard pallets, customized wooden pallets) into the database, and automatically retrieves the corresponding pallet size through the RFID tags on the pallet edge or visual recognition technology when the AGV robot 4 approaches the goods. Taking the pallet carried by the metal support frame 2 as an example, the detailed dimensions such as the bottom hollow structure and the side reinforcement strips are also included in the calculation to ensure the integrity of the pallet reference.

[0059] Identify the height value of the smallest packaging unit in the transported goods; adopt a method combining machine vision and deep learning to perform hierarchical analysis on the goods image. For example, for stacked traditional Chinese medicine woven bags, the system uses image recognition to identify the contour of a single bag, corrects the image distortion by combining the perspective transformation algorithm, and accurately extracts the height value of the smallest packaging unit (single bag). If there are differences in the packaging height of different batches (such as insufficient filling in some bags), the system automatically marks the abnormal units and calculates the average height by weighting.

[0060] Calculate the shooting height value of the smallest packaging unit in contact with the transport pallet 3; based on the installation parameters of the robot camera (such as the height from the ground and the pitch angle), combined with the pixel ratio relationship between the pallet and the goods in the image, calculate the actual distance from the top of the smallest packaging unit in contact with the pallet to the camera through the principle of triangulation. For example, when the robot camera is 1.2 meters from the ground and the depression angle is 30°, by identifying the pixel coordinates of the pallet edge and the bottom of the goods in the image, the shooting height value is calculated in reverse, and the error can be controlled within ±5mm.

[0061] Calculate the position data of the calculated edge on the transport pallet 3 based on the shooting height value and the height value. Calculate the vertical position of the edge on the pallet according to the shooting height value and the height of the minimum packaging unit. For example, if the height of a single bag is 20 cm and the shooting height value is 80 cm, the actual height of the bottom of the edge from the pallet surface is 80 cm - 20 cm = 60 cm. Combine the pallet size and the image distortion correction parameters to map the horizontal and vertical pixel coordinates of the edge in the image to actual physical coordinates. For example, given that the pallet width is 1000 mm and the horizontal pixel proportion of the edge in the image is 30%, the actual abscissa of the edge is 1000 mm × 30% = 300 mm.

[0062] Integrate multi-source data such as the size of the transport pallet 3 and the height of the minimum packaging unit of the goods, and deeply analyze the complex form and spatial relationship of the goods stacked on the pallet. By accurately calculating the shooting height value, convert the relative position relationship between the goods and the pallet into quantitative data, comprehensively covering influencing factors such as pallet specification differences and irregular packaging stacking. This multi-dimensional data collaborative analysis mode effectively avoids position data deviation caused by missing details of goods stacking and misjudgment of pallet size, greatly improves the calculation accuracy of the volume of goods based on the calculated edge, provides reliable data support for subsequent goods density detection and transportation status regulation, and enhances the accuracy and reliability of the transportation management in the traditional Chinese medicine warehouse.

[0063] The method further includes the following steps: The AGV robot 4 is equipped with a distributed pressure sensor array, and 8 - 12 detection points (such as the four corners and the midpoints of the four sides) are evenly distributed at the bottom of the pallet. When receiving the elimination warning instruction, the sensors synchronously collect the pressure data of each point at a sampling frequency of 200 Hz, and eliminate the interference signals caused by robot vibration and uneven ground through a filtering algorithm to ensure the real-time and accuracy of the weight data.

[0064] Calculate the center of gravity position based on multiple weight data; based on the principle of statics, the system uses a weighted average algorithm to calculate the center of gravity coordinates of the goods.

[0065] The system automatically obtains the geometric center coordinates of the transport pallet 3.

[0066] Calculate the position distance between the center of gravity position and the center position; calculate the position distance D between the center of gravity and the center through the Euclidean distance formula.

[0067] Adjust the loading and unloading speed of the AGV robot 4 for loading and transporting the pallet 3 in an inverse correlation with the position distance, and adjust the steering speed when the AGV robot 4 is turning in an inverse correlation with the position distance. Establish an inverse correlation function between the loading and unloading speed vload and the position distance D: vload = v0 × 1 / (1 + k1 × D); where, v0 is the reference speed (such as 0.3 m / s), and k1 is the adjustment coefficient (such as 4 / m). For example, when the center of gravity offset distance reaches 10 cm, the speed automatically drops to 0.214 m / s, so that the robot maintains a slow and uniform motion during the lifting and lowering processes, avoiding the dumping or slipping of the goods due to the imbalance of the center of gravity.

[0068] For the steering scenario, adopt a segmented adjustment strategy: When D ≤ safety threshold, the steering speed is reduced in a normal proportion (such as reduced to 80% of the reference speed); When D > safety threshold, enable the "progressive deceleration" mode: the robot first pre-decelerates with an acceleration of 0.5 m / s², and then completes the turning at a lower speed (such as 0.2 m / s). By extending the turning time (from the conventional 2 seconds to 4 - 6 seconds), the impact of inertia on the overweight goods is offset.

[0069] Convert the offset of the center of gravity of the goods into a quantifiable control parameter, realizing the upgrade from passive response to active prevention. Compared with the traditional fixed-speed transportation mode, it innovatively couples the position distance and the depth of speed adjustment, not only significantly improving the safety and stability of the transportation of traditional Chinese medicine dry medicinal materials, but also extending the service life of the AGV equipment by reducing mechanical wear and tear, providing technical guarantee for the efficient operation of intelligent warehousing.

[0070] To cope with the potential risks brought by the offset of the center of gravity of the goods and the movement inertia, an intelligent acceleration adjustment mechanism based on vector analysis is constructed. By quantifying the spatial relationship between the movement direction of the robot and the center of gravity of the goods, the dynamic stability control of the transportation process is realized. The specific method also includes the following steps: Obtain the movement direction vector of the AGV robot 4; The AGV robot 4 is equipped with a high-precision inertial navigation system (INS) and a lidar, which can collect the traveling direction, speed and position information of the robot in real time. Based on the robot coordinate system, abstract its movement direction as a unit vector vmove = (xm, ym, zm), where (xm, ym, zm) represents the components of the movement direction in three-dimensional space. For example, when driving straight, vmove = (1, 0, 0).

[0071] Calculate the center of gravity vector of the goods based on the center position and the center of gravity position; after obtaining the center position (Xt, Yt, Zt) of the transport pallet 3 and the center of gravity position (Xc, Yc, Zc) of the goods, construct the center of gravity offset vector vcg = (Xc - Xt, Yc - Yt, Zc - Zt) through vector operations. This vector not only reflects the distance of the center of gravity offset but also contains information on the offset direction. For example, when the center of gravity of the goods is biased to the right, the x-component of vcg is positive.

[0072] Calculate the vector angle between the movement direction vector and the center of gravity vector of the goods; use the vector dot product formula to calculate the included angle θ between the movement direction vector and the center of gravity vector; where the value range of θ is [0, π], and the larger the angle, the higher the degree of deviation between the center of gravity offset direction and the movement direction. For example, when θ = π / 2, the center of gravity of the goods is severely side-offset, and it is easy to cause tipping due to centrifugal force during movement.

[0073] Adjust the acceleration for accelerating the AGV robot 4 in an inverse correlation according to the vector angle; establish an inverse correlation function between the acceleration for accelerating a1 and the vector angle θ: a1 = a0 × 1 / (1 + k1 × θ); where a0 is the reference acceleration (such as 0.5 m / s²), and k1 is the adjustment coefficient (such as 0.6 / rad). When θ approaches π (the center of gravity is almost perpendicular to the movement direction), the acceleration drops to 0.1 m / s², causing the robot to accelerate slowly to avoid the goods from slipping or tipping due to inertia during rapid startup.

[0074] Adjust the deceleration for decelerating the AGV robot 4 in a positive correlation according to the vector angle. Use a positive correlation function to adjust the deceleration a2: a2 = amin + k2 × θ; where amin is the minimum deceleration (such as 0.3 m / s²), and k2 is the coefficient (such as 0.4 m / s² / rad). When θ is large, the deceleration increases to more than 1.2 m / s², causing the robot to quickly reduce the speed during braking, reducing the impact of inertia on the goods, and preventing the goods with a severely offset center of gravity from being thrown off due to sudden braking.

[0075] In the right-angle turning transportation test of a certain traditional Chinese medicine warehouse, for a batch of wolfberry medicinal materials with a center of gravity offset of 12 cm, when the robot turns, the vector included angle θ reaches π / 3. The system automatically reduces the acceleration for accelerating from 0.5 m / s² to 0.2 m / s² and increases the deceleration from 0.4 m / s² to 0.8 m / s², effectively avoiding the goods from scattering due to centrifugal force during turning.

[0076] By quantifying the spatial relationship between the cargo's center of gravity offset and its direction of motion through vector angles, this technology upgrades traditional empirical speed control to data-driven precision control. Compared to a fixed acceleration mode, this system implements a dynamic response strategy where larger offset angles result in slower acceleration and faster deceleration. This strategy is particularly suitable for transporting loosely packaged dried Chinese medicinal herbs with a variable center of gravity. This technology not only ensures safe transportation but also extends the mechanical life of the AGV robot 4 by reducing sudden acceleration and deceleration, ensuring efficient and stable operation of smart warehousing and logistics.

[0077] The method also includes the following data: Position sensors located on the top edge of transport pallet 3 calculate the height of the cargo's edge ridges. Not all pallets have these sensors; only those used for calibration have them. High-precision laser displacement sensors (with an accuracy of ±0.1mm) are evenly distributed along the four top edges of transport pallet 3. These sensors scan the cargo's edge contours in real time at a frequency of 100Hz. When AGV robot 4 approaches the cargo, the sensors automatically activate, emitting laser beams perpendicular to the pallet's edge. By measuring the time difference between reflected light beams, they obtain the height of each point on the cargo's edge relative to the pallet's top surface. For example, for stacked woven bags of traditional Chinese medicine, the sensors can accurately capture subtle changes in ridges, such as wrinkles at the bag opening and spillage of medicinal materials. AGV robot 4 simultaneously collects image data at different positions, extracts the calculated edges of the cargo using an edge detection algorithm, and calculates the lateral ridges. This process can be affected by variations in lighting and the cargo's surface texture, leading to errors in the detection results.

[0078] Real-time heave data is calculated based on the height data. After filtering and noise reduction on the height data collected by the position sensor, a sliding window algorithm is used to calculate real-time heave data for the cargo edge. Specifically, using a 10cm window length, the difference between the maximum height data within each window and the pallet's top surface height is calculated. This difference represents the real-time heave measurement for that area. For example, if the maximum height within a window is 23.5cm (for a 20cm pallet height), the real-time heave data is 3.5cm.

[0079] Calculate the edge difference based on the lateral uplift data and the real-time uplift data; compare the lateral uplift data obtained by image analysis with the real-time uplift data calculated by the sensor point by point to calculate the edge difference between the two. If the edge difference of a certain area exceeds the preset reference edge value (such as 3mm), the system immediately triggers the three-level warning mechanism: Level 1 prompt: "Data verification abnormality" will be displayed on the AGV display screen; Secondary alarm: push error area coordinates and deviation values to the warehouse management system; Level 3 Suspension: If the error exceeds the standard in three consecutive areas, the transportation task will be suspended and await manual verification.

[0080] Calibrate the lateral bulge data according to the edge difference. The adaptive gain calibration algorithm is used to correct the lateral bulge data. Dynamically adjust the gain coefficient k according to the magnitude of the edge difference: The calibrated data = the original data + k × the edge difference; where, when the edge difference is less than 1 mm, k = 0.5; when the difference is between 1 - 3 mm, k = 1; when the difference is greater than 3 mm, k = 1.5. For example, if the original lateral bulge data of a certain edge is 4 cm and the edge difference is 2 mm, then the calibrated data is 4 cm + 1×0.2 cm = 4.2 cm.

[0081] Through the complementary verification of the position sensor and image analysis, a dynamic adaptive data calibration system is constructed, which not only greatly improves the reliability of the detection of the goods bulge state, but also provides a solid data foundation for the intelligent decision-making of the AGV robot 4, and strongly promotes the development of traditional Chinese medicine warehousing logistics towards high precision and high safety.

[0082] The embodiment of the present application also discloses a state control system of an AGV robot in a traditional Chinese medicine warehouse, including a processor, and the processor executes the steps of the state control method of the AGV robot in a traditional Chinese medicine warehouse as described in any one of the above.

[0083] The embodiment of the present application also discloses a storage medium, in which a program is stored, and when the program is executed by a processor, the steps of the state control method of the AGV robot in a traditional Chinese medicine warehouse as described in any one of the above are implemented.

[0084] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A method for controlling the state of an AGV robot in a traditional Chinese medicine warehouse, characterized in that, It includes the following steps: Based on the transportation goods instruction, call the AGV robot (4); Obtain first image data at a first position and second image data at a second position beside the transported goods by the AGV robot (4), wherein, with the center of the transported goods as the center of the circle, the radian between the first position and the second position is within a preset first radian range; Extract multiple calculated edges of the transported goods from the first image data and the second image data; Match corresponding temporary edges based on the lowest points of the calculated edges, associate the temporary edges with the calculated edges one by one based on the adjacent position relationship, and extend the temporary edges to the highest points of the corresponding calculated edges to obtain reference edges; Calculate the lateral bulge data of the calculated edges relative to the corresponding reference edges, and the lateral bulge data corresponds to the degree of bulge of the calculated edges on the side away from the center of the circle of the reference edges; If the lateral bulge data is greater than a preset bulge reference data, issue a transportation collision warning; According to multiple pieces of the lateral bulge data corresponding one by one to the calculated edges, calculate a comprehensive bulge data, and inversely adjust the transportation speed of the AGV robot (4) according to the comprehensive bulge data.

2. The state control method of the AGV robot in the traditional Chinese medicine warehouse according to claim 1, characterized in that In the step of obtaining the reference edges, the following sub-steps are further included: Extract the vertical edges of the transportation tray (3) from the first image data and the second image data; Verify the one-to-one correspondence between the vertical edges and the calculated edges based on the adjacent position relationship; If the verification is passed, extend the vertical edges to the highest points of the corresponding calculated edges; Calculate the similarity data between the vertical edges and the reference edges; If the similarity data is greater than a preset similarity reference value, use the vertical edges as the new reference edges; Otherwise, prompt that the temporary edge matching is incorrect.

3. The state control method of the Chinese medicine warehouse AGV robot according to claim 2, characterized in that The method further includes the following steps: If the lateral bulge data is less than the preset bulge reference data, calculate the bulge dispersion data between multiple pieces of the lateral bulge data; If the bulge dispersion data is greater than a preset reference dispersion data, calculate the absolute value of the sum of multiple pieces of the lateral bulge data; Calculate a comprehensive absolute value according to multiple pieces of the absolute values, and inversely adjust the transportation speed of the AGV robot (4) according to the comprehensive absolute value; Calculate a comprehensive dispersion data according to multiple pieces of the bulge dispersion data, and positively adjust the adjustment step size of the transportation speed of the AGV robot (4) according to the comprehensive dispersion data.

4. The state control method of the AGV robot for traditional Chinese medicine warehouse according to claim 2, characterized in that The method further includes the following steps: Calculate the position data of the calculated edges on the transportation tray (3); Calculate the volume value of the transported goods according to the position data and size data of multiple pieces of the calculated edges; Collect the weight value of the transported goods; Calculate the real-time density value according to the volume value and the weight value; Obtain the standard density value corresponding to the transported goods; Calculate the difference between the real-time density value and the standard density value as the density difference; If the absolute value of the density difference is greater than a preset reference difference, issue a warning prompt for the goods type.

5. The state control method of the AGV robot for traditional Chinese medicine warehouse according to claim 4, characterized in that In the step of calculating the position data of the calculation edge on the transport pallet (3), the following sub-steps are further included: Obtain the dimension data of the transport pallet (3); Identify the height value of the smallest packaging unit in the transported goods; Calculate the shooting height value of the smallest packaging unit in contact with the transport pallet (3); Calculate the position data of the calculation edge on the transport pallet (3) according to the shooting height value and the height value.

6. The state control method of the AGV robot for traditional Chinese medicine warehouse according to claim 4, characterized in that The method further includes the following steps: Based on the elimination instruction for eliminating the warning of the goods type, obtain the weight data at multiple positions on the transport pallet (3); Calculate the center of gravity position according to the multiple weight data; Obtain the center position of the transport pallet (3); Calculate the position distance between the center of gravity position and the center position; Inversely adjust the loading and unloading speed of the AGV robot (4) for loading and unloading the transport pallet (3) according to the position distance, and inversely adjust the steering speed of the AGV robot (4) when turning according to the position distance.

7. The state control method of the AGV robot for traditional Chinese medicine warehouse according to claim 6, characterized in that, The method further includes the following steps: Obtain the movement direction vector of the AGV robot (4); Calculate the center of gravity vector of the goods according to the center position and the center of gravity position; Calculate the vector angle between the movement direction vector and the center of gravity vector of the goods; Inversely adjust the acceleration for accelerating the AGV robot (4) according to the vector angle; Directly adjust the deceleration for decelerating the AGV robot (4) according to the vector angle.

8. The state control method of the AGV robot for the traditional Chinese medicine warehouse according to claim 2, characterized in that The method further includes the following data: Calculate the height data of the edge bulge of the transported goods based on the position sensor arranged at the upper edge of the transport pallet (3); Calculate the real-time bulge data according to the height data; Calculate the edge difference according to the lateral bulge data and the real-time bulge data; If the edge difference is greater than the preset reference edge value, issue a data error warning; Correct the lateral bulge data according to the edge difference.

9. A state control system for an AGV robot in a traditional Chinese medicine warehouse, characterized in that, Includes a processor, and the processor executes the steps of the state control method of the traditional Chinese medicine warehouse AGV robot according to any one of claims 1-8.

10. A storage medium, characterized in that, A program is stored in the medium, and when the program is executed by the processor, the steps of the state control method of the traditional Chinese medicine warehouse AGV robot according to any one of claims 1-8 are realized.

Images