CN110458256B

CN110458256B - RFID-based cargo management method, electronic equipment and system

Info

Publication number: CN110458256B
Application number: CN201910619976.XA
Authority: CN
Inventors: 陈建祥; 郑磊
Original assignee: Ruijie Networks Co Ltd
Current assignee: Ruijie Networks Co Ltd
Priority date: 2019-07-10
Filing date: 2019-07-10
Publication date: 2022-11-11
Anticipated expiration: 2039-07-10
Also published as: CN110458256A

Abstract

The embodiment of the invention provides a goods management method, electronic equipment and a system based on RFID. The method comprises the following steps: acquiring the distance between the front goods acquired by the ultrasonic sensor and the goods acquisition equipment; determining the time period for collecting the goods according to the distance; acquiring the signal intensity and the signal phase value of each cargo tag after the RFID identification equipment scans the RFID cargo tag attached to the surface of the cargo in the time period; and inputting the signal intensity and the signal phase value of each cargo tag into a pre-trained classifier, and determining the code information of the collected cargo according to the output result of the classifier. According to the embodiment of the invention, the ultrasonic sensor is utilized to judge the collected goods, the RFID label information is continuously collected for a period of time during collection, the collected goods RFID label is accurately identified according to the pre-trained classifier, threshold setting is not needed, the collection requirements of various collection devices can be met under a complex wireless environment, and the high reliability requirement of a storage system is met.

Description

RFID-based cargo management method, electronic equipment and system

Technical Field

The embodiment of the invention relates to the technical field of communication, in particular to a goods management method, electronic equipment and system based on Radio Frequency Identification (RFID).

Background

The warehouse management system generally adopts a goods coding technology to classify and manage goods, generally adopts media such as bar codes, two-dimensional codes, radio Frequency Identification (RFID) and characters to install on the goods, and manages the goods according to information on the media.

For large goods, the transportation of the goods and the storage management of the large goods are generally carried out by using a forklift, and the two schemes of traditional bar code type registration and RFID (radio frequency identification) assisted bar code type registration are mainly adopted. In the scheme of managing by adopting the RFID technology, the RFID tag is arranged on each large cargo, the RFID identification equipment is arranged on the forklift, and when the forklift forks the cargo, the cargo code is automatically read, so that the cargo forking information can be automatically recorded theoretically.

However, in practical situations, due to complexity of wireless signals and uncontrollable spatial transmission attenuation, the RFID identification device is generally capable of identifying too many RFID tags, and even if goods are not forked, the RFID tags on the goods have a high probability of being read by the RFID identification device, resulting in a low goods identification rate. For such problems, filtering is currently performed by reducing the power of the RFID identification device or adjusting the received signal threshold. However, due to the inaccuracy of the installation position of the RFID tag and the uncertainty of the position of the goods fork-picked by the forklift, it is difficult to call out a proper threshold value, and then, in consideration of the situation that multiple layers of goods are required to be picked up inevitably in an actual scene, the goods in the rear row are shielded by the goods in the front row, so that the signal is weak, and the threshold value adjusting mode is adopted, so that the situation is not applicable to such a scene.

Therefore, how to provide a more efficient cargo management system becomes an important issue to be solved urgently.

Disclosure of Invention

Aiming at the defects in the prior art, the embodiment of the invention provides a goods management method, electronic equipment and a system based on RFID.

In a first aspect, an embodiment of the present invention provides a cargo management method based on RFID, including:

acquiring the distance between the front goods acquired by the ultrasonic sensor and the goods acquisition equipment;

determining the time period for collecting the goods according to the distance;

acquiring the signal intensity and the signal phase value of each goods label after the RFID identification equipment scans the RFID goods labels attached to the surfaces of the goods in the time period;

and inputting the signal intensity and the signal phase value of each cargo tag into a pre-trained classifier, and determining the code information of the collected cargo according to the output result of the classifier.

As above method, optionally, the determining the time period for which the cargo is collected according to the distance includes:

if the distance is larger than the first distance threshold value at the time T0, determining that no goods are collected at the time T0;

if the distance is smaller than a second distance threshold value at the time T1, determining that goods are collected at the time T1, wherein the second distance threshold value is smaller than the first distance threshold value;

and if the distance is larger than the first distance threshold at the moment Tn and T0< T1< Tn, determining that the Tn-T1 time period is the time period of the goods collected by the goods collecting device.

As with the method above, optionally, the cargo collection device is a forklift, and the first distance threshold is equal to a length of a boom of the forklift.

As in the above method, optionally, the classifier is determined in advance according to the following steps:

the method comprises the steps of obtaining signal intensity and signal phase values of various cargo tags in a time period in which cargos are collected in advance;

the signal intensity and the signal phase value of each cargo tag are used as the input of a neural network, the cargo is collected or not collected is used as the output of the neural network, and the parameters of the neural network are determined by utilizing a deep learning algorithm;

and determining the classifier according to the parameters of the neural network.

As in the above method, optionally, the determining parameters of the neural network by using the deep learning algorithm, where the signal strength and the signal phase value of each cargo tag are used as inputs of the neural network, and the cargo is collected or the cargo is not collected as outputs of the neural network, includes:

taking the signal intensity and the signal phase value of each cargo label as horizontal axis pixel points, taking the acquisition time point of each cargo label as vertical axis pixel points, and obtaining m input images of 2*n pixels, wherein n = f T, f is the sampling frequency of RFID identification equipment, T is a time period acquired by cargo acquisition equipment in advance, and m is the number of the cargo labels acquired in the T time period;

respectively taking the input images of the m 2*n pixels as the input of a neural network, taking the goods corresponding to the input images as the output of the neural network, or taking the goods corresponding to the input images as the output of the neural network, and determining the parameters of the neural network according to an image recognition deep neural network algorithm;

correspondingly, the inputting the signal strength and the signal phase value of each cargo tag into a pre-trained classifier includes:

converting the signal intensity and the signal phase value of each cargo tag into p input images of 2*n pixels, and then respectively inputting the input images into a pre-trained classifier, wherein p is the number of the cargo tags acquired in a time period in which the cargo is acquired by the cargo acquisition equipment;

correspondingly, the determining the code information of the collected goods according to the output result of the classifier comprises the following steps:

and determining an input image of the collected goods according to the output result of the classifier, and determining the coding information of the collected goods according to the goods label of the input image of the collected goods.

In a second aspect, an embodiment of the present invention provides an electronic device, including:

the distance acquisition module is used for acquiring the distance between the front goods acquired by the ultrasonic sensor and the goods acquisition equipment;

the sampling period determining module is used for determining the time period for collecting the goods according to the distance;

the tag information acquisition module is used for acquiring the signal intensity and the signal phase value of each cargo tag after the RFID identification equipment scans the RFID cargo tag attached to the surface of the cargo in the time period;

and the classification module is used for inputting the signal intensity and the signal phase value of each cargo tag into a pre-trained classifier and determining the code information of the collected cargo according to the output result of the classifier.

As with the electronic device above, optionally, the sampling period determination module is specifically configured to:

As above, optionally, the electronic device further includes a training module, where the training module is configured to:

taking the signal intensity and the signal phase value of the cargo tag as the input of a neural network, taking the collected cargo or the non-collected cargo as the output of the neural network, and determining the parameters of the neural network by utilizing a deep learning algorithm;

Optionally, the training module is specifically configured to:

correspondingly, the classification module is specifically configured to:

In a third aspect, an embodiment of the present invention provides an RFID-based cargo management system, including: the electronic equipment, the ultrasonic sensor arranged on the goods collecting equipment, the RFID identifying equipment arranged on the goods collecting equipment and the RFID goods label attached to the surface of the goods are arranged;

the ultrasonic sensor is used for acquiring the distance between the front goods and the goods acquisition equipment in real time according to the sampling frequency and transmitting the distance to the electronic equipment;

the RFID identification device is used for scanning the RFID goods labels in the time period when the goods are collected, acquiring the signal intensity and the signal phase value of each goods label and sending the signal intensity and the signal phase value of each goods label to the electronic device.

According to the RFID-based cargo management method provided by the embodiment of the invention, the ultrasonic sensor is utilized to judge the collected cargos by the cargo collection equipment, the continuous RFID label information collection is carried out for a period of time during the collection, the cargo RFID labels collected by the cargo collection equipment are accurately identified according to the pre-trained classifier, the threshold setting is not required, the collection requirements of various cargo collection equipment can be met under the complex wireless environment, and the high reliability requirement of the warehousing system is met.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.

FIG. 1 is a schematic diagram of an RFID-based cargo management system according to an embodiment of the present invention;

FIG. 2 is a schematic flow chart of a RFID-based cargo management method according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a cargo collection process provided by an embodiment of the invention;

fig. 4 is a schematic diagram illustrating distance recognition in the RFID-based cargo management method according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of a deep learning algorithm model according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Fig. 1 is a schematic view of a cargo management system based on RFID according to an embodiment of the present invention, as shown in fig. 1, a storage area is an area for transferring or storing cargos, and usually a large amount of cargos are stacked in the storage area, such as cargos C1 and cargos D3, and the storage area transfers the cargos by using a cargo collection device, such as a forklift. Each cargo is encoded in advance, an Electronic Product Code (EPC) of each cargo is determined, the EPC of each cargo is written into an EPC storage area of an RFID cargo tag, and then different RFID cargo tags are respectively pasted on the surfaces of each cargo in the storage areas, so that the RFID identification device can read the EPC storage area of the RFID cargo tag in a wireless mode.

When the goods in the storage area need to be collected, the goods collecting device is moved to the storage area to collect the goods, as shown in fig. 1, the forklift already forks the goods a and the goods B, and in order to automatically determine the specific goods information forked by the forklift, in the embodiment of the invention, the goods collecting device is respectively provided with the RFID identification device and the ultrasonic sensor, and is connected with the electronic device with an information processing function, such as an edge computing device, and the electronic device can be arranged on the goods collecting device or at other positions as long as the electronic device can receive the information collected by the RFID identification device and the ultrasonic sensor. The electronic equipment belongs to a main control computing unit of the whole system, can control the RFID identification equipment and the ultrasonic sensor, and identifies the RFID label of the collected goods.

The RFID identification device reads the EPC codes on the RFID cargo tags in real time, and obtains the signal strength and Phase information of the received signals of the cargo tags that can be scanned, which are respectively recorded as RSSI and Phase, for example, the RFID identification device collects tag information of cargo a, cargo B, and cargo C1, which are respectively recorded as RSSI _ a, phase _ A, RSSI _ B, phase _ B, and RSSI _ C1 and Phase _ C1. And after the information reading is completed, transmitting the relevant data to the electronic equipment. The ultrasonic sensor is used for identifying whether the goods are collected by the goods collecting device or not, feeding back distance information of a front object and the ultrasonic sensor in real time by the ultrasonic sensor, and transmitting related data to the electronic device. The ultrasonic sensor can only identify the distance between the front object nearest to the ultrasonic sensor and the ultrasonic sensor, but cannot identify the distance between each object, as shown in fig. 1, the distance acquired by the ultrasonic sensor is the distance between the ultrasonic sensor and the nearest cargo a.

Based on the cargo management system, an embodiment of the present invention provides a schematic flow chart of a cargo management method based on RFID, where the method is applied to the electronic device shown in fig. 1, and as shown in fig. 2, the method includes:

s21, acquiring the distance between the front goods acquired by the ultrasonic sensor and the goods acquisition equipment;

specifically, the ultrasonic sensor mounted on the cargo acquisition equipment acquires the distance between the front cargo and the cargo acquisition equipment in real time, and then sends acquired distance information to the electronic equipment. In practical application, electronic equipment can also when needing to acquire distance information, issue the acquisition instruction to ultrasonic sensor, ultrasonic sensor receives the acquisition instruction after, the distance of the place ahead goods and goods collection equipment begins to gather in real time, because the distance of place ahead goods and goods collection equipment when gathering the goods and the distance difference of place ahead goods and goods collection equipment when not gathering the goods are great, consequently can gather the goods through the preliminary judgement of the distance of place ahead goods and goods collection equipment.

S22, determining the time period for collecting the goods according to the distance;

specifically, when the goods are present or absent on the goods acquisition device, the distances between the goods in front acquired by the ultrasonic sensor and the goods acquisition device are different, so that the time period during which the goods are acquired can be determined according to the distance information, for example, the distance in the T time period is the same as the distance characteristic during which the goods are acquired, and the T time period is the time period during which the goods are acquired.

S23, acquiring the signal intensity and the signal phase value of each goods label after the RFID identification equipment scans the RFID goods labels attached to the surfaces of the goods in the time period;

specifically, the RFID identification device mounted on the cargo acquisition device scans the RFID cargo tags attached to the surface of the cargo in real time according to the scanning frequency, and determines the signal strength and the signal phase value of each cargo tag acquired at each scanning time point, where it should be noted that, at each scanning time point, the RFID identification device can only scan the cargo tag signals of a part of the cargo. The RFID identification equipment sends the signal intensity and the signal phase value of each cargo tag acquired at each scanning time point to the electronic equipment, and the electronic equipment intercepts the signal intensity and the signal phase value of the cargo tag which can be identified at each scanning time point in a cargo acquisition time period. In practical application, the electronic device can also issue a collecting instruction to the RFID identification device at the first time after judging that the goods are collected according to the received distance, and after receiving the collecting instruction, the RFID identification device starts to scan the RFID goods tags attached to the surfaces of the goods according to the scanning frequency, and sends the signal intensity and the signal phase value of each goods tag acquired at each scanning time point to the electronic device. When the electronic equipment judges that no goods are collected through the distance, the electronic equipment sends a collection stopping instruction to the RFID identification equipment at the first time so as to reduce resource waste and information interference caused by unnecessary scanning.

And S24, inputting the signal intensity and the signal phase value of each cargo tag into a pre-trained classifier, and determining the code information of the collected cargo according to the output result of the classifier.

Specifically, after the electronic device acquires the signal intensity and the signal phase value of each cargo tag scanned in the time period in which the cargo is collected, it is necessary to determine which cargo in the cargo is actually collected. The electronic device can train a classifier of the second classification in advance, the output of the classifier is that cargoes are collected or not collected, and after the electronic device inputs the signal intensity and the signal phase value of each cargo label scanned in the acquired cargo collection time period into the classifier, which cargoes are collected can be determined, so that EPC (electronic product code) code information of the collected cargoes is determined, and whether the cargoes are collected or not is automatically identified.

Fig. 3 is a schematic diagram of a cargo collection process according to an embodiment of the present invention, and as shown in fig. 3, each cargo is installed with an encoded RFID cargo tag and stored in a specific area, such as a storage area 1, a storage area 2, and the like. Due to the business requirements, the goods A2 need to be transferred from the storage area 1 to the storage area 2. Adopt goods collection equipment to gather goods A2 this moment, RFID identification equipment on the goods collection equipment reads goods RFID information to accurately discern the A2 goods information of being gathered through electronic equipment, confirm that goods collection equipment has carried out correct collection to the goods, goods collection equipment removes goods A2 to position 2 after, whole service system can be according to electronic equipment's recognition result, the new position of accurate renewal goods A2.

On the basis of the foregoing embodiment, further, the determining the time period for which the cargo is collected according to the distance includes:

Optionally, the cargo collection device is a forklift, and the first distance threshold is equal to a fork arm length of the forklift.

Specifically, fig. 4 is a schematic distance identification diagram in the RFID-based cargo management method according to the embodiment of the present invention, as shown in fig. 4, the ultrasonic sensor is used to identify a cargo on a fork arm of the forklift, the ultrasonic sensor feeds back distance information of a front article, and when the forklift does not fork the cargo, the ultrasonic sensor feeds back distance information d0 that is greater than a first distance threshold value because the cargo does not exist on the fork arm, and optionally, the first distance threshold value is equal to the length d of the fork arm, so if at a certain time, for example, at time T0, the distance acquired by the ultrasonic sensor is greater than the first distance threshold value, it indicates that the forklift does not fork the cargo at time T0; after the goods are forked, the goods exist on the fork arms, at the moment, the d0 data are changed, the d0 values are different according to the different quantity of the forked goods, after the d0 value is smaller than a second distance threshold value, the electronic equipment judges that the forklift forks the goods, wherein the second distance threshold value is smaller than the first distance threshold value, and therefore if the distance acquired by the ultrasonic sensor is smaller than the second distance threshold value at a certain moment, such as the T1 moment, the forklift forks the goods at the T1 moment.

In fig. 4, tag _ An represents a forked item Tag closer to the RFID identification device, tag _ Bn represents a forked item Tag farther from the RFID identification device, tag _ Cn represents An undisciplined item Tag closer to the RFID identification device, and Tag _ Dn represents An undisciplined item Tag farther from the RFID identification device. According to the traditional identification scheme, after the goods are judged to be forked, which goods are forked is judged according to the received signal strength RSSI threshold value. As shown in fig. 4, due to the fact that Tag _ Bn is blocked, the signal strength of the RFID cargo Tag is weaker than that of Tag _ Cn, and even though Tag _ An is not blocked, due to the complexity of wireless transmission, the signal strength of Tag _ An cannot be guaranteed to be stronger than that of Tag _ Cn, so that there is a significant misjudgment.

In the embodiment of the invention, after judging to fork goods, the electronic equipment informs the RFID identification equipment to start signal acquisition for a period of time, the reading time is from T1 to Tn, the RFID goods label is continuously read, the reading time length is related to the time of moving out of a storage position after the fork truck forks the goods, the time point of taking the fork goods is T1, and the time point of completely separating the fork arm of the fork truck from the storage position is Tn. As shown in fig. 4, after the forklift forks the goods and moves, in the time period from T1 to Tn, the distance d1 between the forked goods and the RFID identification device is relatively constant, and the distance d2 between the undisputed goods and the RFID identification device is significantly changed. According to the relative motion characteristics of the goods and the forklift, the time period for the goods to be forked can be determined, threshold setting is not needed, pattern recognition is carried out according to the relative motion characteristics under various complex wireless environments, and the RFID tags of the goods forked by the forklift can be accurately recognized.

On the basis of the above embodiments, further, the classifier is determined in advance according to the following steps:

Specifically, due to the relative motion characteristics of the goods and the goods collecting device, the signal strength and the phase of the RFID goods label in the continuous time period can be selected for pattern recognition, so that the collected label code is analyzed. Specifically, the signal strength and the signal phase value of each cargo tag in a collected time period are collected in advance, the RSSI values and the phase values of all identifiable tags are collected at the time T1, the tags are collected continuously according to the collection frequency of RFID identification equipment until the time Tn, the tags are finally integrated into a database, whether the cargo is collected or not is marked, the marked collection information is used as a sample set, one part of samples in the sample set are used as a training sample subset, and the other part of samples are used as a testing sample subset.

Table 1 cargo tag signal information table

Time T1	…	Time Tn
			Tag_A1_RSSI,Tag_A1_Phase	…	Tag_A1_RSSI,Tag_A1_Phase
Tag_B1_RSSI,Tag_B1_Phase	…	Tag_B1_RSSI,Tag_B1_Phase
			Tag_C1_RSSI,Tag_C1_Phase	…	Tag_C1_RSSI,Tag_C1_Phase
Tag_D1_RSSI,Tag_D1_Phase	…	Tag_D1_RSSI,Tag_D1_Phase
			…	…	…
Tag_Aq_RSSI,Tag_Aq_Phase	…	Tag_Aq_RSSI,Tag_Aq_Phase
			Tag_Bq_RSSI,Tag_Bq_Phase	…	Tag_Bq_RSSI,Tag_Bq_Phase
Tag_Cq_RSSI,Tag_Cq_Phase	…	Tag_Cq_RSSI,Tag_Cq_Phase
			Tag_Dq_RSSI,Tag_Dq_Phase	…	Tag_Dq_RSSI,Tag_Dq_Phase

Table 1 is a table of collected tag signal information of the cargo, as shown in table 1, 4*q tag signals can be scanned from time T1 to time Tn, the RSSI and Phase value of the same cargo tag at each time may be the same or different, and if a part of 4*q tag signals is not scanned at a certain time, both the RSSI and Phase value of the tag signal scanned at that time are marked as 0 in table 1. And then, taking the RSSI and the phase value of the cargo label in the training sample subset as the input of the neural network, taking the collected cargo or not as the output of the neural network, carrying out deep learning, determining the parameters of the neural network, and finally optimizing the parameters of the neural network by using the test sample subset to obtain a final classifier, thereby analyzing the collected cargo label.

On the basis of the foregoing embodiments, further, the determining parameters of the neural network by using a deep learning algorithm, where the signal strength and the signal phase value of the cargo tag are used as inputs of the neural network, and the cargo is collected or the cargo is not collected as outputs of the neural network, includes:

respectively taking the input images of the m 2*n pixels as the input of a neural network, taking the collected goods corresponding to the input images or the non-collected goods corresponding to the input images as the output of the neural network, and determining the parameters of the neural network according to an image recognition deep neural network algorithm;

converting the signal intensity and the signal phase value of each cargo tag into p input images of 2*n pixels, and then respectively inputting the input images into a pre-trained classifier, wherein p is the number of the cargo tags acquired in the time period in which the cargo is acquired by the cargo acquisition equipment;

Specifically, if the sampling frequency of the RFID identification device is f, the time period during which the goods are collected by the goods collection device is T during offline training, and m goods label signals are identified in the T time period, then each goods label signal strength RSSI and signal Phase value Phase in table 1 may be further used as a horizontal axis pixel, and the collection time point of each goods label is used as a vertical axis pixel, so as to obtain m input images of 2*n pixels, where n is the number of sampling points and n = f T, and if m different RFID goods labels are identified in the time period from T1 to Tn, then the m input images may be converted into m images.

Fig. 5 is a schematic diagram of a deep learning algorithm model provided in an embodiment of the present invention, and as shown in fig. 5, input images of m 2*n pixels are respectively used as inputs of a neural network, goods corresponding to the input images are collected or goods corresponding to the input images are not collected as outputs of the neural network, and parameters of the neural network are determined by using an image recognition deep neural network algorithm, where fig. 5 only shows the deep learning algorithm model when Tag1 is used as the input image, and if m different RFID goods tags are recognized within time T1 to Tn, the obtained goods are converted into m pictures, and determination and calculation are performed in a time division manner, so as to obtain a collected Tag set.

After the electronic device acquires the RSSI and Phase values of the scanned cargo tags sent by the RFID identification device, each cargo tag is converted into a corresponding input image, for example, the RSSI and Phase values of p cargo tags are acquired, and then the input images are converted into p 2*n pixels and respectively input into a pre-trained classifier, so as to obtain the information of the acquired cargo tags, read the EPC code from the cargo tags, and further locate the acquired cargo.

Experiments show that after the scheme is adopted, compared with the traditional scheme, the recognition rate of collected goods is greatly improved. In the traditional scheme, when gathering the large-scale goods of singleton, the discernment rate of accuracy is generally more than 98%, when gathering many multilayer overlapping cargos, and the discernment rate of accuracy is generally less than 50%, after this scheme of adoption, the rate of accuracy of gathering the large-scale goods of singleton is close 100%, when gathering many multilayer overlapping cargos, the rate of accuracy is greater than 99.9%.

According to the RFID-based cargo management method provided by the embodiment of the invention, the ultrasonic sensor is utilized to judge the collected cargos by the cargo collection equipment, the continuous RFID label information collection is carried out for a period of time during the collection, the cargo RFID labels collected by the cargo collection equipment are accurately identified according to the pre-trained classifier, the reliable automatic storage identification operation is realized, the threshold setting is not needed, the collection requirements of various cargo collection equipment can be met under the complex wireless environment, and the high reliability requirement of the storage system is met.

Based on the same inventive concept, an embodiment of the present invention further provides an electronic device, and fig. 6 is a schematic structural diagram of the electronic device provided in the embodiment of the present invention, as shown in fig. 6, the electronic device includes: a distance acquisition module 61, a sampling period determination module 62, a tag information acquisition module 63, and a classification module 64, wherein:

the distance acquisition module 61 is used for acquiring the distance between the front goods acquired by the ultrasonic sensor and the goods acquisition equipment; the sampling period determining module 62 is used for determining the time period for which the goods are collected according to the distance; the tag information acquisition module 63 is configured to acquire the signal strength and the signal phase value of each cargo tag after the RFID identification device scans the RFID cargo tag attached to the surface of the cargo within the time period; the classification module 64 is configured to input the signal strength and the signal phase value of each cargo tag into a pre-trained classifier, and determine the encoding information of the collected cargo according to the output result of the classifier.

Optionally, the sampling period determination module is specifically configured to:

Optionally, the system further comprises a training module, wherein the training module is configured to:

taking the signal intensity and the signal phase value of each cargo tag as the input of a neural network, taking the collected cargo or not as the output of the neural network, and determining the parameters of the neural network by utilizing a deep learning algorithm;

Optionally, the training module is specifically configured to:

correspondingly, the classification module is specifically configured to:

Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.

The above-described embodiments of the apparatuses and the like are merely illustrative, wherein the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the embodiments of the present invention, and are not limited thereto; although embodiments of the present invention have been described in detail with reference to the foregoing embodiments, those skilled in the art will understand that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An RFID-based cargo management method, comprising:

determining the time period for collecting the goods according to the distance;

2. The method of claim 1, wherein determining a time period for which the cargo is collected based on the distance comprises:

3. The method of claim 2, wherein the cargo acquisition device is a forklift and the first distance threshold is equal to a boom length of the forklift.

4. A method according to any of claims 1-3, characterized in that the classifier is determined beforehand according to the following steps:

5. The method of claim 4, wherein the determining parameters of the neural network using the signal strength and the signal phase value of each cargo tag as inputs to the neural network and cargo collected or cargo not collected as outputs from the neural network comprises:

6. An electronic device, comprising:

and the classification module is used for inputting the signal strength and the signal phase value of each cargo tag into a pre-trained classifier and determining the acquired code information of the cargo according to the output result of the classifier.

7. The electronic device of claim 6, wherein the sampling period determination module is specifically configured to:

8. The electronic device of claim 6 or 7, further comprising a training module to:

9. The electronic device of claim 8, wherein the training module is specifically configured to:

correspondingly, the classification module is specifically configured to:

10. An RFID-based cargo management system, comprising: the electronic device as claimed in any one of claims 6 to 9, an ultrasonic sensor disposed on the cargo collection device, an RFID identification device disposed on the cargo collection device, and an RFID cargo tag attached to a surface of the cargo;

the RFID identification device is used for scanning the RFID goods labels in the time period of goods collection, acquiring the signal intensity and the signal phase value of each goods label, and sending the signal intensity and the signal phase value of each goods label to the electronic device.