Inbound and outbound information collection method applied to smart warehouses

Abstract

The present invention relates to the field of data processing technology, and in particular to a method for collecting in-and-out warehouse information applied to a smart warehouse, comprising: using multiple RFID receiving antennas arranged in the smart warehouse to identify the RFID tags of goods, and determining at least one garbled goods whose identity information cannot be identified in the smart warehouse; determining the cargo collection characteristics of at least one garbled goods, the cargo collection characteristics at least including the total number of garbled goods, the total weight of garbled goods, and the edge curve data of each garbled goods; determining the target cargo area where the garbled goods are suspected to be located according to the track line data when the staff takes and places the goods in the smart warehouse; determining the cargo identity information of the garbled goods according to the cargo collection characteristics, and the preset cargo weight and preset edge curve data of a single cargo under the corresponding storage cargo type in the target cargo area. The present invention can realize the accurate collection of the cargo identity information corresponding to the garbled goods.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method for collecting in-and-out warehouse information applied to a smart warehouse. Background Art

[0002] With the rapid development of the Internet of Things, automation technology, and artificial intelligence, the methods of collecting inbound and outbound information in smart warehouses have made significant progress. At present, a variety of technologies such as barcode / QR code scanning, radio frequency identification (RFID), smart wearable devices, machine vision technology, voice recognition technology, and Internet of Things sensors are widely used in warehouse management, realizing the rapid and accurate collection of cargo information. The integrated application of these technologies not only improves warehouse operation efficiency, but also provides enterprises with real-time and transparent inventory data, provides strong support for supply chain management, and promotes the intelligent transformation of the logistics industry.

[0003] In the overall process of collecting inbound and outbound information of conventional smart warehouses, RFID can realize non-contact multi-tag parallel identification, and the RFID tags on individual goods can remain relatively stable for a long time during storage. However, its RFID signal is easily interfered by liquid or metal in the monitoring process, which leads to deviation of the collected data or inability to accurately identify, and cannot guarantee the accurate collection of inbound and outbound information of smart warehouses. Summary of the invention

[0004] In order to solve the technical problem that the RFID signal cannot guarantee the accurate collection of the in-and-out information of the smart warehouse when it is interfered by liquid or metal in the monitoring link, the purpose of the present invention is to provide a method for collecting in-and-out information applied to the smart warehouse, and the technical scheme adopted is as follows:

[0005] In a first aspect, the present invention provides a method for collecting inbound and outbound information applied to a smart warehouse, the method comprising:

[0006] Using multiple RFID receiving antennas deployed in the smart warehouse to identify the RFID tags of the goods, determining at least one garbled goods whose identity information cannot be identified in the smart warehouse, wherein each of the goods in the smart warehouse is respectively configured with an independent RFID tag of the goods;

[0007] Determine at least one cargo collection feature of the garbled cargo, the cargo collection feature at least including the total quantity of the garbled cargo, the total weight of the garbled cargo, and edge curve data of each of the garbled cargo;

[0008] Determine the target cargo area where the garbled cargo is suspected to be located based on the trajectory data of the staff when picking up and placing the cargo in the smart warehouse;

[0009] The cargo identity information of the garbled cargo is determined according to the cargo collection characteristics, and the preset cargo weight and preset edge curve data of a single cargo under the cargo type corresponding to the target cargo area, wherein different cargo types correspond to different preset cargo weights and preset edge curve data.

[0010] Optionally, before determining the target cargo area where the garbled cargo is suspected to be located based on the trajectory data of the staff when picking up and placing the cargo in the smart warehouse, the method further includes:

[0011] Determine the real-time location information of the staff when picking up and placing goods in the smart warehouse during a preset continuous time period, wherein the real-time location information is obtained by identifying the identity card of the staff using the multiple RFID receiving antennas;

[0012] According to the pickup time points corresponding to the different real-time location information, the trajectory route data of the staff when picking up and placing goods in the smart warehouse is generated.

[0013] Optionally, the real-time location information of the staff when picking up and placing goods in the smart warehouse during the preset continuous time period is determined, including:

[0014] At each pickup time point in the preset continuous time period, respectively collect the real-time signal monitoring strength of the multiple RFID receiving antennas for the identity card;

[0015] Among the multiple RFID receiving antennas, extract three target RFID receiving antennas corresponding to the highest real-time signal monitoring strength;

[0016] Obtaining antenna position coordinates corresponding to the three target RFID receiving antennas and their estimated distances relative to the ID card;

[0017] According to the triangulation positioning method, a circle is constructed with the antenna position coordinates of the three target RFID receiving antennas as the center and the estimated distance relative to the identity card as the radius, and the intersection of the three circles is determined as the real-time position information of the staff when picking up and placing the goods at the current pickup time.

[0018] Optionally, determining the target cargo area where the garbled cargo is suspected to be located according to the trajectory data of the staff when taking and placing the cargo in the smart warehouse includes:

[0019] According to the trajectory route data, the length of time the staff stays in different cargo areas is counted;

[0020] By analyzing the pickup habit information of the staff, determining the pickup time characteristics of the staff in different cargo areas, the pickup time characteristics at least include the pickup time and the time deviation value;

[0021] Based on the length of time the staff stays in different cargo areas and the characteristics of the cargo pickup time, a first probability value of the presence of the garbled cargo in each cargo area is calculated;

[0022] The cargo area corresponding to the first probability value greater than the first preset threshold among the multiple cargo areas is determined as the target cargo area where the garbled cargo is suspected to be located.

[0023] Optionally, the determining of at least one cargo collection feature of the garbled cargo, the cargo collection feature including at least the total quantity of the garbled cargo, the total weight of the garbled cargo and the edge curve data of each of the garbled cargo, comprises:

[0024] Taking statistics of abnormal coding signals for garbled goods whose identity information cannot be identified in the smart warehouse;

[0025] The number of abnormal coding signals obtained by counting is determined as the total number of garbled goods;

[0026] Using a pressure sensor configured in the smart warehouse to detect the total weight of at least one of the garbled goods;

[0027] The edge curve data of each of the garbled goods is scanned using an infrared laser scanning device configured in the smart warehouse, wherein the edge curve data at least includes the edge curve length, the average edge curvature and the maximum edge curvature.

[0028] Optionally, determining the cargo identity information of the garbled cargo according to the cargo collection characteristics, and the preset cargo weight and preset edge curve data of a single cargo under the cargo type corresponding to the storage in the target cargo area, includes:

[0029] Based on the preset cargo weight, the total number of garbled cargo and the total weight of the garbled cargo, a plurality of cargo combinations are determined in the target cargo area, wherein the number of first target cargo included in each cargo combination is the total number of garbled cargo, and the sum of the weights of the first target cargo corresponding to the preset cargo weight is equal to the total weight of the garbled cargo;

[0030] According to the preset edge curve data and the edge curve data of each garbled product, feature matching is performed on each garbled product and each combination of products to obtain edge feature similarity;

[0031] Based on the edge feature similarity, calculating a second probability value of each of the goods combinations being a garbled goods combination;

[0032] By analyzing the second probability values ​​corresponding to different combinations of the goods, the goods identity information of the garbled goods is obtained.

[0033] Optionally, the step of performing feature matching on each of the garbled goods and each of the goods combinations according to the preset edge curve data and the edge curve data of each of the garbled goods to obtain edge feature similarity includes:

[0034] Determine each of the goods combinations as the current goods combination in turn, and perform a cosine similarity calculation process with each edge curve in each of the garbled goods and the current goods combination, until a target cosine similarity between each edge curve and the current goods combination is obtained;

[0035] Calculate the average of the target cosine similarities of all the edge curves in each of the garbled goods to obtain the first edge feature similarity of each of the garbled goods and the current goods combination;

[0036] The mean of the first edge feature similarity corresponding to at least one of the garbled goods is calculated to obtain a second edge feature similarity of at least one of the garbled goods and the current goods combination.

[0037] Optionally, the target cosine similarity calculation process includes:

[0038] Determine each edge curve in each of the garbled goods as the current edge curve in turn;

[0039] The first cosine similarity between the current edge curve and any target edge curve in the current goods combination is calculated until the calculation of the cosine similarity of all the target edge curves in the current goods combination is completed, and the largest first cosine similarity is determined as the target cosine similarity between the current edge curve and the current goods combination.

[0040] Optionally, the calculating, based on the edge feature similarity, a second probability value of each of the goods combinations being a garbled goods combination includes:

[0041] Determine at least one target cargo area where a plurality of first target cargoes included in each cargo combination are located in the smart warehouse;

[0042] Calculating a mean value of the at least one target cargo area with respect to the first probability value;

[0043] A weighted sum is taken for the second edge feature similarity and the mean of the first probability value to obtain a second probability value that each of the goods combinations is a garbled goods combination.

[0044] Optionally, the obtaining of the cargo identity information of the garbled cargo by analyzing the second probability values ​​corresponding to different cargo combinations includes:

[0045] Extracting a garbled cargo combination corresponding to the second probability value being greater than a second preset threshold value from the plurality of cargo combinations;

[0046] Scattering the first target goods included in the random code goods combination, and using the second probability value of the random code goods combination as the second probability value of each of the first target goods;

[0047] The second probability values ​​of the same first target goods included in the plurality of the garbled code goods combinations are summed to obtain a third probability value of each different first target goods;

[0048] According to the order of the third probability values ​​from large to small, a preset number of first target goods are selected as second target goods, and the preset number is equal to the total number of the garbled goods;

[0049] The preset cargo type of the second target cargo is determined as the cargo identity information of the garbled cargo.

[0050] The present invention has the following beneficial effects: through the technical solution provided by the present invention, the RFID tags of goods can be firstly identified by using multiple RFID receiving antennas arranged in the smart warehouse, and at least one garbled goods whose goods identity information cannot be identified in the smart warehouse can be determined; further, the goods collection characteristics of at least one of the garbled goods can be determined, and the goods collection characteristics at least include the total number of garbled goods, the total weight of garbled goods and the edge curve data of each garbled goods; and according to the track line data when the staff takes and places goods in the smart warehouse, the target goods area where the garbled goods are suspected to be located is determined; finally, according to the goods collection characteristics, and the preset goods weight and preset edge curve data of a single goods under the corresponding storage type of the target goods area, the goods identity information of the garbled goods is determined. The technical solution in this application can comprehensively determine the possible types of garbled goods from multiple angles according to the entire process of the staff taking and placing goods, as well as the weight monitoring data and the edge curve data of the infrared scan, when the RFID signal of the goods in the monitoring link is interfered by liquid or metal, so as to realize the accurate collection of the goods identity information corresponding to the garbled goods. This will reduce repeated monitoring caused by monitoring failure during the goods outbound process, and greatly improve the reliability and convenience of smart warehouse inbound and outbound operations.

[0051] It should be understood that the above general description and the following detailed description are only exemplary and explanatory and cannot limit the present invention. Other features and advantages of the present invention will be described in detail in the following specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 A flowchart of a method for collecting inbound and outbound information applied to a smart warehouse provided by an embodiment of the present invention;

[0054] Figure 2 A flowchart of a method for collecting in-and-out warehouse information applied to a smart warehouse is provided as another embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a method for collecting in-and-out information applied to a smart warehouse proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0056] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0057] The following is a detailed description of a specific solution of a method for collecting in-and-out warehouse information applied to a smart warehouse provided by the present invention in conjunction with the accompanying drawings.

[0058] See also Figure 1 , which shows a method flow chart of a method for collecting inbound and outbound information applied to a smart warehouse provided by an embodiment of the present invention, the method comprising the following steps:

[0059] Step 110: Use multiple RFID receiving antennas deployed in the smart warehouse to identify the RFID tags of the goods, and determine at least one garbled goods whose identity information cannot be identified in the smart warehouse.

[0060] In specific application scenarios, in order to deal with the problems of inefficiency and identification interference in the smart warehouse collection system, this solution adopts Radio Frequency Identification (RFID) signal collection technology to configure an independent cargo RFID tag for each cargo in the smart warehouse, that is, an RFID coded electronic tag. In addition, multiple RFID antennas can be installed in the entrance and exit areas of the smart warehouse. However, during the identification process, due to the stacking of goods during entry and exit or other factors, the tag may be attached to the metal surface or close to the water source, which will significantly reduce the reading effect. In order to solve the problem of signal collection failure in the smart warehouse entry and exit information collection method, multiple RFID antennas can be deployed inside the factory area. Specifically, an RFID antenna matrix can be installed above the smart warehouse to achieve full coverage and reception of the external coded signal of the electronic tag.

[0061] For the disclosed embodiment, if the RFID tag of the goods is missing or damaged, the RFID signal may be abnormal and the reading may fail, and the identity information of the goods may not be recognized. In order to realize the contactless full-intelligent goods outbound identification, at least one garbled goods whose identity information cannot be identified in the smart warehouse may be first determined.

[0062] Step 120: determine at least one cargo collection feature of the garbled cargo, where the cargo collection feature at least includes the total quantity of the garbled cargo, the total weight of the garbled cargo, and edge curve data of each garbled cargo.

[0063] In a specific application scenario, when the RFID tag of the goods cannot be recognized normally, the RFID receiving antenna will monitor the abnormal coding signal for the garbled goods, and the total number of garbled goods can be obtained by counting the number of abnormal coding signals received. And a pressure sensor device and an infrared monitoring device can be set at the entrance and exit area of ​​the factory warehouse. The pressure sensor device can be used to monitor the weight of goods entering and leaving the warehouse, and the infrared monitoring device is used to perform an overall scan of the goods entering and leaving the warehouse to obtain the three-dimensional contour features of different goods. For the embodiments of the present disclosure, the pressure sensor can be used to collect the weight information of abnormally coded goods, and the total number of garbled goods of the garbled goods can be further obtained, and the edge curve data of the garbled goods can be monitored by the infrared monitoring device.

[0064] Step 130: Determine the target cargo area where the garbled cargo is suspected to be located based on the trajectory data of the staff when picking up and placing the cargo in the smart warehouse.

[0065] For the disclosed embodiment, the identity cards of the staff can be monitored throughout the process to obtain their trajectory routes and the time they stay in different areas, and then combined with the staff's picking habits, the target cargo area where the garbled cargo is suspected to be located can be analyzed.

[0066] Step 140: Determine the cargo identity information of the garbled cargo according to the cargo collection characteristics, and the preset cargo weight and preset edge curve data of a single cargo under the cargo type corresponding to the target cargo area.

[0067] Among them, different cargo types correspond to different preset cargo weights and preset edge curve data.

[0068] For the disclosed embodiment, the trajectory data of the staff when picking up and placing the goods can be first analyzed to determine the corresponding probability of the goods in the area where they stay in the warehouse, and the target cargo area where the garbled goods are suspected to be located can be obtained by preliminary analysis. Next, the weight data collected by the pressure sensor is used to construct different combinations of suspected garbled goods. Subsequently, the edge features of the suspected garbled goods are obtained by laser radar scanning technology. Finally, the preset edge curve data of a single cargo under the corresponding storage cargo type in the target cargo area is combined to judge the matching degree with the cargo edge features, and the cargo category of the current suspected garbled goods is screened out from both subjective and objective dimensions, that is, the cargo identity information of the garbled goods is obtained.

[0069] In summary, according to a method for collecting in-and-out information applied to a smart warehouse provided by the present invention, the RFID tags of goods can be firstly identified by using multiple RFID receiving antennas arranged in the smart warehouse to determine at least one garbled goods whose identity information cannot be identified in the smart warehouse; further, the cargo collection characteristics of at least one garbled goods can be determined, and the cargo collection characteristics at least include the total number of garbled goods, the total weight of garbled goods and the edge curve data of each garbled goods; and according to the trajectory line data when the staff takes and places goods in the smart warehouse, the target cargo area where the garbled goods are suspected to be located is determined; finally, according to the cargo collection characteristics, and the preset cargo weight and preset edge curve data of a single cargo under the corresponding storage cargo type in the target cargo area, the cargo identity information of the garbled goods is determined. The technical solution in this application can comprehensively determine the possible types of garbled goods from multiple angles according to the entire process of the staff taking and placing goods, as well as the weight monitoring data and the edge curve data of the infrared scan, when the RFID signal of the goods is interfered by liquid or metal in the monitoring link, so as to realize the accurate collection of the cargo identity information corresponding to the garbled goods. This will reduce repeated monitoring caused by monitoring failure during the goods outbound process, and greatly improve the reliability and convenience of smart warehouse inbound and outbound operations.

[0070] based on Figure 1 The embodiment shown is a refinement and extension of the above embodiment. In order to fully illustrate the specific implementation process of the method of this embodiment, this embodiment provides the following Figure 2 The specific method shown. Figure 2 based on Figure 1 As shown in 2, the method comprises the following steps:

[0071] Step 210: Use multiple RFID receiving antennas deployed in the smart warehouse to identify the RFID tags of the goods, and determine at least one garbled goods whose identity information cannot be identified in the smart warehouse.

[0072] For the embodiments of the present disclosure, the embodiment steps may refer to the relevant description in embodiment step 110, and no specific limitation is made here.

[0073] Step 220: determine at least one cargo collection feature of the garbled cargo, where the cargo collection feature at least includes the total quantity of the garbled cargo, the total weight of the garbled cargo, and edge curve data of each garbled cargo.

[0074] For the embodiment of the present disclosure, determining the cargo collection features of at least one garbled cargo in step 220 may include the following steps:

[0075] Step 220-1: count abnormal coding signals of garbled goods whose identity information cannot be identified in the smart warehouse, and determine the number of abnormal coding signals obtained by counting as the total number of garbled goods.

[0076] In a specific application scenario, when the RFID receiving antenna identifies the RFID tag of the goods, if the RFID tag of the goods is blocked or damaged, the RFID receiving antenna will not be able to identify the coded signal corresponding to the RFID tag of the goods. At this time, the coded signal corresponding to the RFID tag of the goods can be marked as an abnormal coded signal. After completing the identification of the RFID tags of goods in all areas of the smart warehouse, the total number of garbled goods can be obtained by counting the number of abnormal coded signals.

[0077] Step 220-2: Use the pressure sensor configured in the smart warehouse to detect the total weight of at least one garbled cargo.

[0078] In specific application scenarios, pressure sensors can be configured in the cargo entrance and exit areas of smart warehouses. When all cargoes enter and leave the smart warehouse, their weights will be collected, and the collected weights will be bound to the cargo identity information identified by the cargoes. For cargoes whose cargo identity information cannot be identified, they can be marked as garbled cargoes, and their weights can be recorded and accumulated. After the weight collection of all cargoes is completed, the total weight of at least one garbled cargo can be obtained.

[0079] Step 220-3: Use the infrared laser scanning device configured in the smart warehouse to scan the edge curve data of each garbled product, where the edge curve data at least includes the edge curve length, the average edge curvature, and the maximum edge curvature.

[0080] In specific application scenarios, infrared laser scanning equipment can be configured in the cargo entrance and exit area of ​​the smart warehouse. When all goods enter and exit the smart warehouse, edge curve data will be collected and bound to the cargo identity information identified by the goods. For goods whose identity information cannot be identified, they can be marked as garbled goods and the edge curve data can be recorded. After completing the edge data collection of all goods, the edge curve data of each garbled goods can be obtained.

[0081] Step 230: determine the real-time location information of the staff when picking up and placing goods in the smart warehouse during a preset continuous time period. The real-time location information is obtained by identifying the staff's identity card using multiple RFID receiving antennas.

[0082] For the embodiment of the present disclosure, determining the real-time location information of the staff when picking up and placing goods in the smart warehouse during the preset continuous time period in step 230 may include the following steps:

[0083] Step 230-1: At each pickup time point in a preset continuous time period, real-time signal monitoring strength of multiple RFID receiving antennas for the identity card is collected respectively.

[0084] In a specific application scenario, the RFID receiving antenna will continuously receive signals from the staff's identity electronic tag (i.e., identity card). Different signal monitoring intensities indicate that the staff is at different distances from the RFID receiving antenna.

[0085] Step 230 - 2 : extract three target RFID receiving antennas corresponding to the highest real-time signal monitoring strength from among the multiple RFID receiving antennas.

[0086] At each pickup time point, each RFID receiving antenna can monitor the identity card signal under different signal monitoring strengths. The greater the signal monitoring strength, the closer the staff is to the RFID receiving antenna. In order to more accurately and comprehensively locate the real-time position of the staff, the three target RFID receiving antennas with the highest real-time signal monitoring strength can be extracted at each pickup time point, so as to further determine the real-time position information of the staff when picking up and placing the goods at the current pickup time point based on the antenna position information of the three target RFID receiving antennas.

[0087] Step 230-3: Obtain antenna position coordinates corresponding to the three target RFID receiving antennas and estimated distances relative to the ID card.

[0088] Step 230-4: According to the triangulation positioning method, a circle is constructed with the antenna position coordinates of the three target RFID receiving antennas as the center and the estimated distance relative to the ID card as the radius, and the intersection of the three circles is determined as the real-time position information of the staff when picking up and placing the goods at the current pickup time.

[0089] Step 240: Generate trajectory data of the staff when picking up and placing goods in the smart warehouse according to the pickup time points corresponding to different real-time location information.

[0090] After the real-time location information of the staff when picking up and placing the goods at each pickup time point is determined through step 230 of the embodiment, the multiple real-time location information can be route-fitted according to the time sequence of the pickup time points from front to back to obtain the trajectory route data of the staff when picking up and placing the goods in the smart warehouse.

[0091] Step 250: Determine the target cargo area where the garbled cargo is suspected to be located based on the trajectory data of the staff when taking and placing the cargo in the smart warehouse.

[0092] By analyzing the staff's picking habits, we can obtain the staff's regional residence time for a single pick-up in each cargo area when there are no garbled goods, as well as the pick-up time characteristics of different goods. Based on the trajectory data of the staff when they actually pick up and put goods in the smart warehouse, we can statistically obtain the staff's actual residence time in each cargo area. When the actual residence time deviates greatly from the staff's regional residence time when there are no garbled goods, it can be determined that the cargo area is the target cargo area where the garbled goods are suspected to be located.

[0093] For the disclosed embodiment, determining the target cargo area where the garbled cargo is suspected to be located according to the trajectory data of the staff when picking up and placing the cargo in the smart warehouse in step 250 may include the following steps:

[0094] Step 250-1: Count the length of time that staff members stay in different cargo areas based on the track route data.

[0095] For the embodiment of the present disclosure, the regional location range of each cargo area can be determined. When it is determined that the real-time location of the staff falls within the regional location range, the staff's stay time statistics in the cargo area are started. Until it is determined that the real-time location of the staff moves outside the regional location range, the stay time statistics in the cargo area are stopped to obtain the final stay time statistical value.

[0096] Step 250 - 2 : By analyzing the pickup habit information of the staff, the pickup time characteristics of the staff in different cargo areas are determined. The pickup time characteristics at least include the pickup time and the time deviation value.

[0097] Among them, the pickup time can be the total pickup time of the staff in picking up n items in a certain pickup area during historical pickup; the time deviation value is the time offset sequence spent by the staff in picking up n items in a certain area. This time offset sequence is different. When the degree of concentration is greater, it means that the staff's pickup time in this place is more concentrated. The time offset value can be used to measure the difference between the actual pickup time of the staff and the mean pickup time. This value can be positive, negative or zero, indicating that the actual pickup time is later, earlier or exactly the same as the mean pickup time. If the mean pickup time is 5s, and the actual pickup time of the staff in the cargo area is 4.5s, then the time deviation value = actual pickup time - mean pickup time = -0.5s.

[0098] Step 250-3: Calculate a first probability value of the presence of garbled goods in each cargo area based on the staff's stay time in different cargo areas and the characteristics of the pickup time.

[0099] In a specific application scenario, when analyzing the pickup time of the mth area, if the difference between the stay time in the area and the pickup time required for the corresponding coded goods is larger, the first probability value of the garbled goods belonging to the area is higher:

[0100]

[0101] In the formula, Indicates the first probability value of the existence of garbled goods in the mth area; represents the normalization function; Indicates the length of time the staff stays in the mth area; represents the time required to pick up v pieces of goods in the mth area, and v represents the number of coded goods shipped in the mth area; It represents the deviation of the time taken to pick up v pieces of goods in the mth area.

[0102] Step 250-4: Determine the cargo area among the multiple cargo areas whose corresponding first probability value is greater than the first preset threshold as the target cargo area where the garbled cargo is suspected to be located.

[0103] The closer the first probability value is to 1, the greater the probability that the cargo area corresponds to the cargo area with garbled codes. In a specific application scenario, the first preset threshold is a value between 0 and 1, and the specific value can be set according to actual application needs and is not specifically limited here.

[0104] Step 260: Determine the cargo identity information of the garbled cargo according to the cargo collection characteristics, and the preset cargo weight and preset edge curve data of a single cargo under the cargo type corresponding to the target cargo area.

[0105] For the disclosed embodiment, determining the cargo identity information of the garbled cargo in step 260 according to the cargo collection characteristics, and the preset cargo weight and preset edge curve data of a single cargo under the cargo type corresponding to the target cargo area may include the following steps:

[0106] Step 260-1: Determine multiple cargo combinations in the target cargo area based on the preset cargo weight, the total number of garbled cargo, and the total weight of garbled cargo.

[0107] Among them, the quantity of the first target goods in each cargo combination is the total quantity of the garbled goods, and the sum of the weights of the first target goods corresponding to the preset cargo weights is equal to the total weight of the garbled goods.

[0108] Step 260-2: Based on the preset edge curve data and the edge curve data of each garbled product, feature matching is performed between each garbled product and each product combination to obtain edge feature similarity.

[0109] Since different categories of goods have different geometric features, the edge curve data obtained by infrared laser scanning is used to extract its edge features. Feature matching is performed through the relevant information (i.e., the preset edge curve data) displayed in the database of the above-screened goods combination to obtain the matching degree of the edge geometric features between the single goods combination and the actual garbled goods, that is, the second edge feature similarity described below is obtained.

[0110] For the embodiments of the present disclosure, the embodiment steps may include: determining each cargo combination as the current cargo combination in turn, and performing a cosine similarity calculation process with each edge curve in each garbled cargo and the current cargo combination, respectively, until a target cosine similarity between each edge curve and the current cargo combination is obtained; calculating the mean of the target cosine similarities corresponding to all edge curves in each garbled cargo, and obtaining a first edge feature similarity between each garbled cargo and the current cargo combination; calculating the mean of the first edge feature similarity corresponding to at least one garbled cargo, and obtaining a second edge feature similarity between at least one garbled cargo and the current cargo combination.

[0111] The target cosine similarity calculation process includes: determining each edge curve in each garbled cargo as the current edge curve in turn; calculating the first cosine similarity between the current edge curve and any target edge curve in the current cargo combination, until all target edge curves in the current cargo combination complete the calculation of cosine similarity, and determining the largest first cosine similarity as the target cosine similarity between the current edge curve and the current cargo combination.

[0112] The similarity of the first edge feature of each garbled product and the current product combination can be recorded as:

[0113]

[0114] In the formula, is the similarity of the first edge feature between the nth garbled product and the vth current product combination; Indicates the number of edge curves for the nth garbled goods; The multi-dimensional features of the lth edge curve representing the nth garbled goods; Represents the multi-dimensional features of any target edge curve in the vth current cargo combination; Indicates the first cosine similarity between the lth edge curve of the nth garbled goods and any target edge curve in the vth current goods combination about the multi-dimensional features, Indicates the target cosine similarity between the lth edge curve and the vth current cargo combination. The multi-dimensional features may include edge curve length features, edge average curvature features, and edge maximum curvature features.

[0115] The similarity of the second edge feature between at least one garbled product and the current product combination can be recorded as:

[0116]

[0117] In the formula, is the second edge feature similarity of the vth current goods combination as the garbled goods combination; The quantity of goods for all garbled goods; is the similarity of the first edge feature between the nth garbled product and the vth current product combination.

[0118] Step 260-3: Based on the edge feature similarity, calculate the second probability value of each cargo combination being garbled cargo.

[0119] For the embodiments of the present disclosure, the embodiment steps may include: determining at least one target cargo area in the smart warehouse where multiple first target cargoes contained in each cargo combination are located; calculating the mean of the at least one target cargo area with respect to the first probability value; and performing weighted summation on the second edge feature similarity and the mean of the first probability value to obtain a second probability value that each cargo combination is garbled cargo.

[0120] The second probability value of each cargo combination being a garbled cargo can be recorded as:

[0121]

[0122] In the formula, is the second probability value of the vth cargo combination being garbled cargo; is the second edge feature similarity of the vth current goods combination as the garbled goods combination; The quantity of goods for all garbled goods; represents the first target product in the vth product combination The first probability value of being in the target cargo area. and are the corresponding weights respectively. Since edge similarity is more objective in comparing goods similarity, and the analysis of trajectory route data may have certain subjective influence in some cases, you can set The value is 0.9. The value is 0.1.

[0123] Step 260-4: Analyze the second probability values ​​corresponding to different cargo combinations to obtain cargo identity information of the garbled cargo.

[0124] Currently, the similarity of the garbled information of each combination of goods after screening is obtained, that is, the second probability value of each combination of goods being a garbled goods. Further, a part of the combination that may be the best combination can be screened out from the current combination of goods, and then the common goods information in the part of the combination can be extracted to infer the final garbled combination.

[0125] To this end, we first conduct a preliminary screening of all current similar cargo combinations, and filter out the high-probability related cargo combinations in the combination by setting a threshold. The combination of is used as the combination of random code goods for preliminary screening. Among them, M is the second preset threshold, which is a value between 0 and 1. The specific value can be set according to the actual application needs, such as 0.85, and is not specifically limited here.

[0126] Among these random code goods combinations, the actual random code goods combination is one of them. However, due to some subjective factors in the screening process, probability deviation may occur. Therefore, directly taking the combination with the highest probability as the current random code combination goods information may be wrong. Therefore, the goods information of the random code goods combination after preliminary screening is scattered, and the second probability value of the corresponding random code goods combination is used as the dimension of its goods information. Then the second probability value of the first target goods v under the i-th combination can be recorded as :

[0127] The second probability values ​​of the same first target goods contained in multiple garbled goods combinations are summed to obtain the third probability value of each different first target goods:

[0128]

[0129] In the formula, is the third probability value of the first target cargo f, is the second probability value of the first target product f under the i-th combination, It is the number of times the same first target product appears in multiple garbled product combinations.

[0130] The corresponding third probability value can be further ranked The first target goods are determined as the second target goods, and the preset goods type of the second target goods is determined as the goods identity information of the garbled goods.

[0131] Correspondingly, for the embodiment of the present disclosure, the embodiment steps may include: extracting a garbled cargo combination whose corresponding second probability value is greater than a second preset threshold value from multiple cargo combinations; breaking up the first target cargo contained in the garbled cargo combination, and using the second probability value of the garbled cargo combination as the second probability value of each first target cargo therein; summing the second probability values ​​of the same first target cargo contained in multiple garbled cargo combinations to obtain a third probability value of each different first target cargo; screening a preset number of first target cargoes as second target cargoes in descending order of the third probability values, the preset number being equal to the total number of garbled cargoes; and determining the preset cargo type of the second target cargo as the cargo identity information of the garbled cargoes.

[0132] In summary, the technical solution in this application can first use multiple RFID receiving antennas arranged in the smart warehouse to identify the RFID tags of the goods, and determine at least one garbled goods in the smart warehouse whose identity information cannot be identified; further, the goods collection characteristics of at least one garbled goods can be determined, and the goods collection characteristics at least include the total number of garbled goods, the total weight of garbled goods, and the edge curve data of each garbled goods; and according to the trajectory line data when the staff takes and places the goods in the smart warehouse, determine the target goods area where the garbled goods are suspected to be located; finally, according to the goods collection characteristics, and the preset goods weight and preset edge curve data of a single goods under the corresponding storage type of the target goods area, determine the goods identity information of the garbled goods. The technical solution in this application can comprehensively determine the possible types of garbled goods from multiple angles when the RFID signal of the goods is interfered by liquid or metal in the monitoring link, according to the entire process of the staff taking and placing the goods, as well as the weight monitoring data and the edge curve data of the infrared scan, and realize the accurate collection of the corresponding goods identity information of the garbled goods. In addition, in the process of goods leaving the warehouse, repeated monitoring caused by monitoring failure is reduced, and the reliability and convenience of the smart warehouse in and out of the warehouse are greatly improved.

[0133] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0134] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0135] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for collecting inbound and outbound information applied to a smart warehouse, characterized in that: The method comprises: Using multiple RFID receiving antennas deployed in the smart warehouse to identify the RFID tags of the goods, determining at least one garbled goods whose identity information cannot be identified in the smart warehouse, wherein each of the goods in the smart warehouse is respectively configured with an independent RFID tag of the goods; Determine at least one cargo collection feature of the garbled cargo, the cargo collection feature at least including the total quantity of the garbled cargo, the total weight of the garbled cargo, and edge curve data of each of the garbled cargo; Determine the target cargo area where the garbled cargo is suspected to be located based on the trajectory data of the staff when picking up and placing the cargo in the smart warehouse; Determine the cargo identity information of the garbled cargo according to the cargo collection characteristics, and the preset cargo weight and preset edge curve data of a single cargo under the cargo type corresponding to the target cargo area, wherein different cargo types correspond to different preset cargo weights and preset edge curve data; Among them, the method for determining the target cargo area where the garbled cargo is suspected to be located is: According to the trajectory route data, the length of time the staff stays in different cargo areas is counted; by analyzing the staff's pickup habit information, the pickup time characteristics of the staff in different cargo areas are determined, and the pickup time characteristics at least include the pickup time and the time deviation value; based on the length of time the staff stays in different cargo areas and the pickup time characteristics, a first probability value of the presence of the garbled cargo in each cargo area is calculated; and the cargo area among the multiple cargo areas whose corresponding first probability value is greater than a first preset threshold is determined as the target cargo area where the garbled cargo is suspected to be located.

2. The method for collecting inbound and outbound information applied to a smart warehouse according to claim 1 is characterized in that: Before determining the target cargo area where the garbled cargo is suspected to be located based on the trajectory data of the staff when picking up and placing the cargo in the smart warehouse, the method further includes: Determine the real-time location information of the staff when picking up and placing goods in the smart warehouse during a preset continuous time period, wherein the real-time location information is obtained by identifying the identity card of the staff using the multiple RFID receiving antennas; According to the pickup time points corresponding to the different real-time location information, the trajectory route data of the staff when picking up and placing goods in the smart warehouse is generated.

3. The method for collecting inbound and outbound information applied to a smart warehouse according to claim 2 is characterized in that: The real-time location information of the staff when picking up and placing goods in the smart warehouse during the preset continuous time period includes: At each pickup time point in the preset continuous time period, respectively collect the real-time signal monitoring strength of the multiple RFID receiving antennas for the identity card; Among the multiple RFID receiving antennas, extract three target RFID receiving antennas corresponding to the highest real-time signal monitoring strength; Obtaining antenna position coordinates corresponding to the three target RFID receiving antennas and their estimated distances relative to the ID card; According to the triangulation positioning method, a circle is constructed with the antenna position coordinates of the three target RFID receiving antennas as the center and the estimated distance relative to the identity card as the radius, and the intersection of the three circles is determined as the real-time position information of the staff when picking up and placing the goods at the current pickup time.

4. The method for collecting inbound and outbound information applied to a smart warehouse according to claim 1 is characterized in that: The determining of at least one cargo collection feature of the garbled cargo, wherein the cargo collection feature at least includes the total quantity of the garbled cargo, the total weight of the garbled cargo, and the edge curve data of each of the garbled cargo, comprises: Taking statistics of abnormal coding signals for garbled goods whose identity information cannot be identified in the smart warehouse; The number of abnormal coding signals obtained by counting is determined as the total number of garbled goods; Using a pressure sensor configured in the smart warehouse to detect the total weight of at least one of the garbled goods; The edge curve data of each of the garbled goods is scanned using an infrared laser scanning device configured in the smart warehouse, wherein the edge curve data at least includes the edge curve length, the average edge curvature and the maximum edge curvature.

5. The method for collecting inbound and outbound information applied to a smart warehouse according to claim 1 is characterized in that: Determining the cargo identity information of the garbled cargo according to the cargo collection characteristics, and the preset cargo weight and preset edge curve data of a single cargo under the cargo type corresponding to the target cargo area includes: Based on the preset cargo weight, the total number of garbled cargo and the total weight of the garbled cargo, a plurality of cargo combinations are determined in the target cargo area, wherein the number of first target cargo included in each cargo combination is the total number of garbled cargo, and the sum of the weights of the first target cargo corresponding to the preset cargo weight is equal to the total weight of the garbled cargo; According to the preset edge curve data and the edge curve data of each garbled product, feature matching is performed on each garbled product and each combination of products to obtain edge feature similarity; Based on the edge feature similarity, calculating a second probability value of each of the goods combinations being a garbled goods combination; By analyzing the second probability values ​​corresponding to different combinations of the goods, the goods identity information of the garbled goods is obtained.

6. The method for collecting inbound and outbound information applied to a smart warehouse according to claim 5 is characterized in that: According to the preset edge curve data and the edge curve data of each garbled product, feature matching is performed on each garbled product with each combination of products to obtain edge feature similarity, including: Determine each of the goods combinations as the current goods combination in turn, and perform a cosine similarity calculation process with each edge curve in each of the garbled goods and the current goods combination, until a target cosine similarity between each edge curve and the current goods combination is obtained; Calculate the average of the target cosine similarities of all the edge curves in each of the garbled goods to obtain the first edge feature similarity of each of the garbled goods and the current goods combination; The mean of the first edge feature similarity corresponding to at least one of the garbled goods is calculated to obtain a second edge feature similarity of at least one of the garbled goods and the current goods combination.

7. The method for collecting inbound and outbound information applied to a smart warehouse according to claim 6 is characterized in that: The target cosine similarity calculation process includes: Determine each edge curve in each of the garbled goods as the current edge curve in turn; The first cosine similarity between the current edge curve and any target edge curve in the current goods combination is calculated until the calculation of the cosine similarity of all the target edge curves in the current goods combination is completed, and the largest first cosine similarity is determined as the target cosine similarity between the current edge curve and the current goods combination.

8. The method for collecting inbound and outbound information applied to a smart warehouse according to claim 6 is characterized in that: The calculating, based on the edge feature similarity, a second probability value of each of the goods combinations being a garbled goods combination comprises: Determine at least one target cargo area where a plurality of first target cargoes included in each cargo combination are located in the smart warehouse; Calculating a mean value of the at least one target cargo area with respect to the first probability value; A weighted sum is taken for the second edge feature similarity and the mean of the first probability value to obtain a second probability value that each of the goods combinations is a garbled goods combination.

9. The method for collecting inbound and outbound information applied to a smart warehouse according to claim 5 is characterized in that: The obtaining of the cargo identity information of the garbled cargo by analyzing the second probability values ​​corresponding to different cargo combinations includes: Extracting a garbled cargo combination corresponding to the second probability value being greater than a second preset threshold value from the plurality of cargo combinations; Scattering the first target goods included in the random code goods combination, and using the second probability value of the random code goods combination as the second probability value of each of the first target goods; The second probability values ​​of the same first target goods included in the plurality of the garbled code goods combinations are summed to obtain a third probability value of each different first target goods; According to the order of the third probability values ​​from large to small, a preset number of first target goods are selected as second target goods, and the preset number is equal to the total number of the garbled goods; The preset cargo type of the second target cargo is determined as the cargo identity information of the garbled cargo.

Images