Pallet rack physical inventory method and system using multi-sensor complementary error correction

Abstract

Disclosed in the present invention are a pallet rack physical inventory method and system using multi-sensor complementary error correction. The method comprises the following steps: locating an inventory device at a designated position; using a camera to photograph a pallet position to be subjected to inventory, detecting an image, and calculating number data T1 of cigarette cases at the pallet position; using ultrasonic ranging devices at different heights to emit ultrasonic waves to the pallet position, and calculating number data T2 of the cigarette cases on the pallet position; and separately comparing the number data T1 and the number data T2 with inventory number data T3, if said number data are all consistent, indicating that number measurement is up to standards, and otherwise, outputting a corresponding instruction according to a comparison result. The present invention measures the number data of the cigarette cases from two dimensions of image measurement and ultrasonic ranging, obtains the number of the cigarette cases at the pallet position by means of logic operation, and compares the measurement results of the two dimensions of image and sound with the inventory data, thus improving inventory accuracy; when one data identification mode is not available, another identification mode can be used to continue the inventory operation.

Description

A pallet rack physical inventory method and system with multi-sensor complementary error correction Technical Field

[0001] The present invention relates to the technical field of physical inventory counting of pallet racks, and in particular to a physical inventory counting method and system for cigarette pallet racks that utilizes multiple sensors for mutual complementation and error correction. Background Art

[0002] With the rapid development of my country's economy, efficient logistics has become an inevitable requirement for economic development. Logistics management is the "third source of profit" after saving raw materials and improving labor productivity. Through efficient warehouse management, enterprises can reduce operating costs, optimize resource allocation, and enhance their competitiveness.

[0003] Pallet racking is the primary storage method in tobacco logistics distribution centers. Tobacco industry logistics management standards require regular and irregular inventory checks of cigarettes. These physical checks aim to verify the consistency between accounts and physical inventory—that is, to ensure that accounts are consistent with actual inventory. However, traditional manual counting methods are time-consuming, wasteful, and inaccurate, and are gradually being phased out.

[0004] At present, there are some methods on the market that use machines to assist manual inventory counting, but the problem of low inventory accuracy is common. For example, the Chinese invention patent with publication number CN115649732A discloses an intelligent inventory counting device and inventory counting method, which includes two parts: an intelligent inventory counting device and an inventory counting method. The inventory counting device uses a radio frequency method to read item information. This method requires the placement of radio frequency tags on the item packaging in advance. During the identification process, the distance and deflection angle between the radio frequency tag and the radio frequency reading device have a greater impact on the reading, which can easily cause reading errors and affect the inventory accuracy.

[0005] Another example is Chinese invention patent No. 202010528597.2, which discloses an inventory counting method and vending machine. This method uses an electrical detection device installed at the bottom of the aisle to detect the presence of merchandise at a specific location in the aisle. The device then calculates the number of items in the aisle based on the pre-determined length of the ticket. This method is not suitable for physical inventory counting on pallet racks, and therefore has low accuracy.

[0006] Summary of the Invention

[0007] In response to the shortcomings of the existing technology, the present invention aims to provide a pallet rack physical inventory method with multi-sensor complementary error correction to solve the problem of low accuracy of the inventory method in the existing technology. The specific technical solution is as follows:

[0008] A method for physical inventory of pallet racks with multi-sensor complementary error correction, characterized by comprising the following steps:

[0009] S1: Positioning the inventory counting device at a designated position relative to the shelf to be counted;

[0010] S2: Using a camera installed on the inventory counting device to photograph the pallet to be counted, recognizing the captured image, and calculating the number of cigarette boxes on the pallet T1 in combination with pre-stored cigarette box size data;

[0011] Using a plurality of ultrasonic distance measuring devices installed at different heights of the inventory counting device to transmit ultrasonic waves horizontally toward the pallet to be counted, the ultrasonic waves are used to measure the number of stacking layers and stacking method of cigarette cartons on the pallet, and the number data T2 of cigarette cartons on the pallet is calculated based on the pre-stored pallet and cigarette cartons size and specification data;

[0012] S3: Compare the quantity data T1 and the quantity data T2 with the inventory quantity data T3 respectively. If the quantity data T1 and the quantity data T2 are consistent with the inventory quantity data T3, the quantity detection is qualified; otherwise, the quantity detection is unqualified and the corresponding instruction is output according to the comparison result.

[0013] As a preferred embodiment: the method for positioning the inventory device at a specified position relative to the pallet to be counted in S1 is: placing a magnetic positioning mark on the bottom pallet position of the shelf to be tested, installing the inventory device on a mobile carrier, and providing a magnetic locator on the mobile carrier, and the mobile carrier moves to the magnetic positioning mark and stops.

[0014] By setting a magnetic locator on the mobile carrier, the mobile carrier can be accurately positioned at the specified position when it moves to the magnetic positioning mark, which facilitates subsequent data collection operations and ensures the accuracy of the measurement results.

[0015] As a preferred embodiment: in step S2, while recognizing the captured image, the specification data A1 of the cigarette box is identified, and the barcode of the cigarette box is scanned by a barcode scanning device installed on the inventory device to read the specification data A2 of the cigarette box;

[0016] In step S3, the product specification data A1 and product specification data A2 are compared with the inventory product specification data A3 respectively. If the product specification data A1 and product specification data A2 are consistent with the inventory product specification data, the product specification detection is qualified; otherwise, the product specification detection is unqualified and the corresponding instruction is output according to the comparison result.

[0017] The specifications of cigarette boxes are inspected through image recognition and code scanning, and compared with the specifications in stock, which improves the accuracy of inspection. At the same time, image recognition and code scanning are complementary. Image recognition to a certain extent makes up for the situation where code scanning cannot be performed due to reasons such as long distance, thereby improving the accuracy of the detection method.

[0018] As a preferred embodiment: the step of identifying the captured image in step S2 includes: pallet area detection, segmenting the cigarette box area in the image, marking the cigarette box area with a rectangular frame, calculating the aspect ratio of the rectangular frame of the cigarette box area and comparing it with the set aspect ratio, determining the type of cigarette box stacking, and calculating the number of cigarette boxes.

[0019] By calculating the height-to-width ratio of the rectangular box area (including the pallet), we can identify the type of box stacking and thus calculate the number of boxes. While captured images may exhibit some distortion, there may be discrepancies between the actual dimensions and the image. However, the height-to-width ratio is constant, so a relatively accurate ratio can be calculated based on the height-to-width ratio of the box area, thereby determining the number of boxes.

[0020] As a preferred embodiment, the height difference between any two of the ultrasonic sensors is half the height of a single-layer smoke box.

[0021] By setting a certain height difference between each ultrasonic sensor, and setting each to half the height of a single-layer cigarette box, each layer of cigarette boxes corresponds to two ultrasonic sensors, and even half-layer stacked cigarette boxes will be captured by the ultrasonic sensor, thereby improving the accuracy of cigarette box number detection.

[0022] As a preferred embodiment: in S2, after the ultrasonic ranging device is used to transmit ultrasonic waves to the pallet position to be counted, the inventory counting device is driven to move horizontally in a direction parallel to the shelf, and then the above ultrasonic measurement process is repeated multiple times.

[0023] Since the inventory counting device moves horizontally parallel to the shelf, the distance between the ultrasonic ranging device and the cigarette box is always the same. Multiple measurements at different positions can effectively avoid errors. By measuring once at a certain interval, the number of unfilled cigarette boxes on the top layer can be measured, and the number of unfilled cigarette boxes can be calculated based on the width occupied by the cigarette boxes on the top layer.

[0024] As a preferred implementation method: when the quantity detection fails in step S3, if the quantity data T1 and the quantity data T2 are inconsistent with the inventory quantity data T3, the operator performs manual verification; if one of the quantity data T1 and the quantity data T2 is inconsistent with the inventory quantity data T3, re-detection is performed. If the quantity detection still fails after multiple re-detections, it is determined that the detection is wrong and the staff is reminded to debug the equipment.

[0025] If both quantity data T1 and quantity data T2 are inconsistent with inventory quantity data T3, it means that the number of cigarette boxes on the pallet may not match the inventory data, that is, the account does not match the actual situation, so staff need to check. If only one of the items does not match the inventory data, it may be a detection error. If it still fails after multiple re-tests, there may be a problem with the detection equipment and it needs to be re-debugged.

[0026] As a preferred implementation method: when the product specification inspection fails in step S3, if the product specification data A1 and the product specification data A2 are inconsistent with the inventory product specification data A3, the operator performs manual verification; if one of the product specification data A1 and the product specification data A2 is inconsistent with the inventory product specification data A3, re-inspection is performed. If the quantity inspection still fails after multiple re-inspections, it is determined that the inspection is wrong and the staff is reminded to debug the equipment.

[0027] If both specification data A1 and specification data A2 are inconsistent with inventory specification data A3, it means that the specifications of the cigarette boxes on the pallet may not match the inventory data, that is, the accounts do not match the actual situation, so staff need to check. If only one of the items does not match the inventory data, it may be a detection error. If it still fails after multiple re-tests, there may be a problem with the detection equipment and it needs to be re-debugged.

[0028] As a preferred implementation method: when both the quantity inspection and the product specification inspection are qualified, the mobile carrier is driven to move to the next pallet inspection point.

[0029] The inventory counting device is driven by a mobile carrier to move, which can achieve manual operation. After one pallet position passes the inspection, the mobile carrier moves to the next pallet position, and then is positioned at the designated position by the magnetic positioning device, ready for the next round of inspection.

[0030] The present invention also provides a pallet rack physical inventory system with multi-sensor complementary error correction, including a data acquisition module, a sensor data verification module and a measurement result judgment module. The data acquisition module is used to collect cigarette box data on the pallet to be measured, and the sensor data verification module is used to calculate and compare the collected cigarette box data and pass the comparison result to the measurement result judgment module. The measurement result judgment module is used to output the judgment result and give the next operation to be performed based on the judgment result.

[0031] The multi-sensor complementary error correction pallet rack physical inventory system uses the sensor data verification module to compare the collected data, which improves the accuracy of the inventory. The result judgment module outputs the judgment result and gives the next operation instruction, which greatly improves the inventory efficiency.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The present invention collects data on the number of cigarette cartons on a pallet through image detection and ultrasonic ranging, and calculates the number of cigarette cartons on the pallet through logical operations. The detection results of the two dimensions of image and acoustics are compared with the inventory data, thereby improving the accuracy of inventory counting. In addition, the data of the two dimensions can compensate for each other's errors. In some special occasions, when one data collection method is unavailable, the other data collection method can be used to allow the inventory work to continue. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] FIG1 is a flow chart of a system provided by the present invention;

[0035] FIG2 is a two-dimensional plan view of the space between the ultrasonic sensor and the smoke box inside the ultrasonic distance measuring device provided by the present invention;

[0036] FIG3 is a diagram showing the stacking shape of three-layer cigarette boxes provided by the present invention;

[0037] FIG4 is a diagram showing the stacking shape of two and a half layers of cigarette boxes provided by the present invention;

[0038] FIG5 is a diagram showing the stacking shape of two-layer cigarette boxes provided by the present invention;

[0039] FIG6 is a diagram showing the stacking shape of one and a half layers of cigarette boxes provided by the present invention;

[0040] FIG7 is a diagram showing the stacking shape of a single layer of cigarette boxes provided by the present invention;

[0041] FIG8 is a diagram showing the stacking shape of half-layer cigarette boxes provided by the present invention; DETAILED DESCRIPTION

[0042] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0043] In the description of the present invention, it should be noted that the terms "upper," "lower," "inner," "outer," "front end," "rear end," "both ends," "one end," "the other end," and the like, indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limiting the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0044] Example 1: A method for physical inventory of pallet racks with multi-sensor complementary error correction. First, the inventory device is installed on a mobile carrier. The mobile carrier can use equipment such as an AGV cart to achieve automatic movement. The mobile carrier travels in a straight line parallel to each shelf, so that its distance from each cigarette box is always the same. The bottom of the shelf is a pallet, and the cigarette boxes are placed on the pallet. A magnetic positioning mark is set on the pallet at the bottom of each shelf, and a magnetic locator is set on the mobile carrier. When the mobile carrier moves to the magnetic positioning mark, the magnetic locator senses the change in magnetic flux and positions the mobile carrier at the specified position (the magnetic positioning device is existing technology, so it will not be described in detail) to ensure that the angle and distance of the inventory device relative to the shelf to be inventoried are always consistent.

[0045] The inventory device is equipped with a camera to capture images of cigarette cases on pallets. After capture, the image is identified by marking the area of ​​the cigarette case (including the bottom pallet) in the image with a rectangular frame. The cigarette case dimensions used in this embodiment are 465 × 250 × 580 mm. Based on these dimensions, the corresponding heights of the cigarette cases for different stacking heights can be calculated: 3-layer height is 1740 mm, 2.5-layer height is 1660 mm, 2-layer height is 1160 mm, 1.5-layer height is 1080 mm, 1-layer height is 580 mm, and 500 mm for half-layer height. The width is 965 mm. Since there are horizontal gaps between the cigarette cases during stacking, the error in this gap is within 50 mm. The pallet height is 150 mm.

[0046] In addition to directly comparing the stacking shapes in the captured images with the stacking shapes of cigarette boxes at various layers in the database, we can also calculate the aspect ratio of the cigarette box area. The aspect ratio of the image is measured in pixels. As shown in Figures 3-8, the number of cigarette boxes stacked on the pallet is 10 per layer, and 5 for a half-layer.

[0047] If the height-to-width ratio is greater than 1.95, it means the cigarette boxes are stacked in 3 layers and the number of cigarette boxes is 30;

[0048] If the height-to-width ratio is between 1.71 and 1.95 (inclusive), it means the cigarette boxes are stacked in 2.5 layers and the number of cigarette boxes is 25;

[0049] If the height-to-width ratio is between 1.35 and 1.70 (inclusive), it means the cigarette boxes are stacked in two layers and the number of cigarette boxes is 20;

[0050] If the height-to-width ratio is between 1.27 and 1.34 (inclusive), it means the cigarette boxes are stacked in one and a half layers and the number of cigarette boxes is 15;

[0051] If the height-to-width ratio is between 0.75 and 1.26 (inclusive), it means the cigarette boxes are stacked in one layer and the number of cigarette boxes is 10;

[0052] If the height-to-width ratio is between 0.67 and 0.74 (inclusive), it means the cigarette boxes are stacked half-layer and the number of cigarette boxes is 5;

[0053] If the aspect ratio is between 0.16-0.66 (inclusive), it means that the tray is empty and the number of cigarette boxes is 0.

[0054] The number data T1 of the cigarette boxes on the pallet is thus obtained.

[0055] The inventory counting device is also equipped with several ultrasonic distance measuring devices at different heights, which include ultrasonic transmitters and receivers. In this embodiment, there are six ultrasonic distance measuring devices, and the height difference between adjacent ones is equal, and each is half the height of a single-layer cigarette box. Therefore, one layer of cigarette box corresponds to two ultrasonic distance measuring devices, and half a layer of cigarette box corresponds to one ultrasonic distance measuring device. Based on the time of sound wave reflection, the distance between the cigarette box and the ultrasonic distance measuring device can be calculated, thereby determining the number of stacked layers and status of the cigarette boxes on the pallet. The principle of ultrasonic distance measurement can be expressed by the following formula: d = v × Δt / 2

[0056] d is the distance between the ultrasonic transmitter and the object being measured;

[0057] v represents the propagation speed of ultrasonic waves in air;

[0058] Δt represents the time difference.

[0059] The propagation speed of ultrasound in air is 340 m / s. The time difference is the time interval between the emission of ultrasound and the reception of the sound wave, measured in microseconds (μs).

[0060] As shown in Figure 2, the distance between the ultrasonic distance measuring device and the tray is x 0，The X-axis represents the distance between the ultrasonic distance measuring device and the smoke box, and the Y-axis represents the height of the ultrasonic distance measuring device. In the ultrasonic distance measuring device, ultrasonic sensor 1 is located at the intersection of line y=300 and the y-axis. The first intersection between line y=300 and the smoke box is x1. Ultrasonic sensor 2 is located at the intersection of line y=600 and the y-axis. The first intersection between line y=600 and the smoke box is x2. Ultrasonic sensor 3 is located at the intersection of line y=900 and the y-axis. The first intersection between line y=900 and the smoke box is x3. Ultrasonic sensor 4 is located at the intersection of line y=1200 and the y-axis. The first intersection between line y=1200 and the smoke box is x4. Ultrasonic sensor 5 is located at the intersection of line y=1500 and the y-axis. The first intersection between line y=1500 and the smoke box is x5. Ultrasonic sensor 6 is located at the intersection of line y=1800 and the y-axis. The first intersection between line y=1800 and the smoke box is x6.

[0061] If there are more than 5 cigarette cases, they are stacked vertically, with 10 cases per layer. In special cases, the top layer is stacked horizontally, with 5 cases per layer. The maximum number of layers of cigarette cases is 3.

[0062] When the top layer of cigarette boxes is stacked horizontally, the area occupied by the top layer of cigarette boxes on the pallet is reduced compared to the area occupied by the top layer stacked vertically. The reduced area is set to be 10000mm. 2 and 810000mm 2 .

[0063] Changes in the positions of cigarette boxes stacked on pallets or pallets placed on shelves will result in placement errors, and the present invention sets the error to 150 mm.

[0064] If the values ​​of x1, x2, x3, x4, x5, and x6 are all x0±150 mm, the three-layer stack of cigarette boxes is full. The stacking shape of the cigarette boxes is shown in Figure 3, and the total number of cigarette boxes is 30.

[0065] If the values ​​of x1, x2, x3, and x4 are all x0 ± 150 mm, the value of x5 is greater than x0 ± 150 mm and less than (x0 ± 150 mm) + 100 mm, and the value of x6 is greater than (x0 ± 150 mm) + 100 mm and less than (x0 ± 150 mm) + 900 mm, then the first and second layers of cigarette boxes are full, the third layer is full horizontally, and there is one cigarette box on top of the third layer. The stacking configuration is shown in Figure 4, and the total number of cigarette boxes is 25.

[0066] If the values ​​of x1, x2, x3, and x4 are all x0 ± 150 mm, and the values ​​of x5 and x6 are all greater than (x0 ± 150 mm) + 900 mm, then the first and second layers of cigarette boxes are full, and the stacking shape of the cigarette boxes is as shown in Figure 5, with a total of 20 cigarette boxes.

[0067] If the values ​​of x1 and x2 are both x0±150mm, the value of x3 is greater than x0±150mm and less than (x0±150mm)+100mm, the value of x4 is greater than (x0±150mm)+100mm and less than (x0±150mm)+900mm, and the values ​​of x5 and x6 are both greater than (x0±150mm)+900mm, then the first layer of cigarette boxes is full, the second layer is full horizontally, and there is one cigarette box above the second layer. The cigarette box stacking shape is shown in Figure 6, and the total number of cigarette boxes is 15.

[0068] If the values ​​of x1 and x2 are both x0 ± 150 mm, and the values ​​of x3, x4, x5, and x6 are all greater than (x0 ± 150 mm) + 900 mm, then the first layer of cigarette boxes is full. The stacking shape of the cigarette boxes is shown in Figure 7, and the total number of cigarette boxes is 10.

[0069] If the value of x1 is greater than x0 ± 150mm and less than (x0 ± 150mm) + 100mm, the value of x2 is greater than (x0 ± 150mm) + 100mm and less than (x0 ± 150mm) + 900mm, and the values ​​of x3, x4, x5, and x6 are all greater than (x0 ± 150mm) + 900mm, then the first layer of cigarette boxes is fully stacked horizontally, with one cigarette box above the first layer. The cigarette box stacking configuration is shown in Figure 8, and the total number of cigarette boxes is 5.

[0070] If the values ​​of x1, x2, x3, x4, x5, and x6 are all greater than (x0±150mm)+900mm, the number of cigarette boxes on that pallet position is zero.

[0071] The quantity data T2 of the cigarette boxes on the pallet is thus obtained.

[0072] Compare the obtained quantity data T1 and quantity data T2 with the inventory quantity data T3 respectively. If the quantity data T1 and quantity data T2 are both the same as the inventory quantity data T3, the quantity detection is qualified and the mobile carrier starts to move to the next pallet position; if the quantity data T1 and quantity data T2 are both different from the inventory quantity data T3, it is very likely that the number of cigarette boxes on the pallet does not match the inventory quantity, that is, the account does not match the actual situation, and the staff should be reminded to check; if one of the quantity data T1 and quantity data T2 is different from the inventory quantity data T3, the number of cigarette boxes on the pallet is re-detected. If it is still different after re-detection 3 times, there may be a problem with one of the detection devices, and the staff should be reminded to troubleshoot and debug the machine.

[0073] Example 2: The difference from Example 1 is that product specification detection is added. Barcodes and product specification information are set on the cigarette boxes. After using the camera to shoot the tray and cigarette box, the product specification information in the picture is identified to obtain product specification data A1. Then, the barcode of the cigarette box is scanned by the scanning device on the inventory device to obtain product specification data A2. The product specification data A1 and product specification data A2 are compared with the inventory product specification data A3 respectively. If the product specification data A1 and product specification data A2 are both the same as the inventory product specification data A3, the product specification detection is qualified and the mobile The carrier starts to move to the next pallet position; if the specification data A1 and specification data A2 are all different from the inventory specification data A3, it is very likely that the specifications of the cigarette boxes on the pallet do not match the inventory specifications, that is, the accounts do not match the actuals, and the staff should be reminded to check; if one of the specification data A1 and specification data A2 is different from the inventory specification data A3, the specifications of the cigarette boxes on the pallet should be re-tested. If they are still different after re-testing 3 times, there may be a problem with one of the detection devices, and the staff should be reminded to troubleshoot and debug the machine.

[0074] As shown in Figure 1, the multi-sensor complementary error correction pallet rack physical inventory system of the present invention includes a data acquisition module, a sensor data verification and comparison module, and a test result judgment module. In the present invention, the data acquisition module uses multiple dimensions to collect data such as the number and specifications of cigarette boxes on the pallet. The multi-dimensional data verifies each other to improve the accuracy of the inventory. In special circumstances, when data collection in one dimension cannot be achieved, data collection in another dimension can be used to allow the inventory work to continue; the sensor data verification and comparison module calculates the number of cigarette boxes on the pallet through logical operations based on the collected data combined with the known pallet and cigarette box size data, compares the number and specifications of the cigarette boxes with the inventory quantity data and specification data, and transmits the comparison results to the test result judgment module. The test result judgment module outputs whether the test result is qualified based on different results, and gives the next operation to be performed.

[0075] In addition to saving manpower and improving efficiency, the present invention detects the quantity and specifications of cigarette boxes from two dimensions: image and sound. The two dimensions verify each other to improve detection accuracy. The two dimensions can also complement each other to improve the accuracy of the system, making it suitable for popularization and use.

[0076] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for physical inventory of pallet racks with multi-sensor complementary error correction, characterized in that: The following steps are involved: S1: Positioning the inventory counting device at a designated position relative to the shelf to be counted; S2: Using a camera installed on the inventory counting device to photograph the pallet to be counted, recognizing the captured image, and calculating the number data T1 of cigarette boxes on the pallet in combination with the pre-stored pallet and cigarette box size data; Using a plurality of ultrasonic distance measuring devices installed at different heights of the inventory counting device to emit ultrasonic waves in the horizontal direction to the pallet position to be counted, the ultrasonic waves are used to measure the number of stacking layers and stacking method of the cigarette boxes on the pallet position, and the number data T2 of the cigarette boxes on the pallet position is calculated in combination with the pre-stored pallet and cigarette box size and specification data; S3: Compare the quantity data T1 and the quantity data T2 with the inventory quantity data T3 respectively. If the quantity data T1 and the quantity data T2 are consistent with the inventory quantity data T3, the quantity detection is qualified; otherwise, the quantity detection is unqualified and a corresponding instruction is output according to the comparison result.

2. The method for physical inventory counting of pallet racks with multi-sensor complementary error correction as claimed in claim 1, characterized in that: The method for positioning the inventory counting device at a specified position relative to the shelf to be counted in S1 is: placing a magnetic positioning mark on the bottom pallet position of the shelf to be counted, installing the inventory counting device on a mobile carrier, and setting a magnetic locator on the mobile carrier, and the mobile carrier moves to the magnetic positioning mark and stops.

3. The method for physical inventory counting of pallet racks with multi-sensor complementary error correction as claimed in claim 2, characterized in that: In step S2, while recognizing the captured image, the specification data A1 of the cigarette box is recognized, and the barcode of the cigarette box is scanned by a barcode scanning device installed on the inventory counting device to read the specification data A2 of the cigarette box; In step S3, the product specification data A1 and the product specification data A2 are compared with the inventory product specification data A3 respectively. If the product specification data A1 and the product specification data A2 are consistent with the inventory product specification data, the product specification detection is qualified; otherwise, the product specification detection is unqualified and the corresponding instruction is output according to the comparison result.

4. The method for physical inventory counting of pallet racks with multi-sensor complementary error correction as claimed in claim 1, characterized in that: The steps of identifying the captured image in step S2 include: pallet area detection, segmenting the cigarette box area in the image, marking the cigarette box area with a rectangular frame, calculating the aspect ratio of the rectangular frame of the cigarette box area and comparing it with the set aspect ratio, determining the type of cigarette box stacking, and calculating the number of cigarette boxes.

5. The method for physical inventory counting of pallet racks with multi-sensor complementary error correction as claimed in claim 1, characterized in that: The height difference between any two of the plurality of ultrasonic sensors is half the height of a single-layer smoke box.

6. The method for physical inventory counting of pallet racks with multi-sensor complementary error correction as claimed in claim 1, characterized in that: In S2, after the ultrasonic distance measuring device is used to transmit ultrasonic waves to the shelf to be counted, the inventory counting device is driven to move horizontally in a direction parallel to the shelf, and then the above ultrasonic measurement process is repeated multiple times.

7. The method for physical inventory counting of pallet racks with multi-sensor complementary error correction as claimed in claim 1, characterized in that: In step S3, when the quantity inspection fails, if the quantity data T1 and the quantity data T2 are inconsistent with the inventory quantity data T3, the operator performs manual verification; If one of the quantity data T1 and the quantity data T2 is inconsistent with the inventory quantity data T3, re-testing is performed. If the quantity test still fails after multiple re-tests, it is determined to be a test error and the staff is reminded to debug the equipment.

8. The method for physical inventory counting of pallet racks with multi-sensor complementary error correction as claimed in claim 3, characterized in that: In step S3, when the product specification inspection fails, if the product specification data A1 and the product specification data A2 are inconsistent with the inventory product specification data A3, the operator performs manual verification; if one of the product specification data A1 and the product specification data A2 is inconsistent with the inventory product specification data A3, re-inspection is performed. If the quantity inspection still fails after multiple re-inspections, it is determined that the inspection is wrong and the staff is reminded to debug the equipment.

9. The method for physical inventory counting of pallet racks with multi-sensor complementary error correction as claimed in claim 3, characterized in that: When both the quantity inspection and the product specification inspection are qualified, the mobile carrier is driven to move to the next pallet inspection point.

10. A system for a pallet rack physical inventory method with multi-sensor complementary error correction as claimed in any one of claims 1 to 9, characterized in that: It includes a data acquisition module, a sensor data verification module and a measurement result judgment module. The data acquisition module is used to collect the cigarette box data on the pallet to be measured. The sensor data verification module is used to calculate and compare the collected cigarette box data and pass the comparison result to the measurement result judgment module. The measurement result judgment module is used to output the judgment result and give the next operation to be performed according to the judgment result.

Images